AN INTERPRETIVE GUIDE

~TO THE PERSONALITY

ASSESSMENT INVENTORY
(PAI*)

Leslie C. Morey, PhD

AR Psychological Assessment Resources, Inc.

Since it was first introduced in 1991, the
Personality Assessment Inventory™
(PAI®) has been heralded as one of the
most important innovations in the field
of clinical assessment. In this book, Dr.
Morey has consolidated the results of
clinical and empirical observations into
a volume that describes the state of the
art in PAI interpretation. An Interpretive
Guide to the Personality Assessment Inven-
tory™ (PAI®) provides an extensive exam-
ination of the following topics:
e The meaning of the individual PAI
scales, with particular emphasis on
subscale configurations
e The meaning and correlates of two-
point codetypes of the PAI clinical
scales
e The strategies for the detection of
sources of negative distortion on the
PAI, including malingering, exagger-
ation, or random responding
e The identification of defensive
responding or positive impression
management, including the denial of
substance abuse problems
e The use of PAI configuration infor-
mation to arrive at DSM-IV diagnoses
of mental disorders
e The evaluation of PAI profiles for the
potential for suicidal or assaultive
behavior
e The assessment of self-concept and
interpersonal adaptation with the
PAI scales
e The use of the PAI for planning and
monitoring treatment

This book provides the most compre-

hensive resource to date on the use and
interpretation of the PAI in clinical
assessment. It is recommended for all
practitioners and researchers who wish
to maximize their use of this important
instrument.
 Digitized by the Internet Archive
in 2022 with funding from
Kahle/Austin Foundation

https://archive.org/details/interpretiveguid(0000more |
AN INTERPRETIVE GUIDE
TO THE PERSONALITY
ASSESSMENT INVENTORY
M

(PAI®)

Leslie C. Morey, PhD

my XPtNe ") hs j
AN INTERPRETIVE GUIDE
TO THE PERSONALITY
ASSESSMENT INVENTORY —
(PAI®)

Leslie C. Morey, PhD

PAR
Ft
(ag
Psychological Assessment Resources, Inc.
 This one is for my Dad.

Library of Congress Cataloging-in-Publication Data

Morey, Leslie Charles, 1956-
An interpretive guide to the Personality Assessment Inventory/Leslie C. Morey
Dae acid,
Includes bibliographical references.
ISBN 0-911907-22-X
1. Personality Assessment Inventory. I. Title.
RC473.P56M67 1996
616.89'075--de20 96-8181
GIP

“Personality Assessment Inventory” is a trademark and “PAI” is a registered trademark owned by

Psychological Assessment Resources, Inc.
Copyright © 1996 by Psychological Assessment Resources, Inc. All rights reserved. May not be repro-
duced in whole or in part in any form or by any means without written permission of Psychological
Assessment Resources, Inc.

SW eiTs Gh alee saPaal Reorder #RO-3341 Printed in the U.S.A.

 PREFACE

The topics covered in this Interpretive Guide grew out of questions that fre-
quently arose as I presented talks and workshops about using the PAI in many dif-
ferent settings. Participants were always interested in the latest research, conducted
by myself and by others, that would help them address important issues in their
assessment practice. As I began making presentations to groups who were increas-
ingly sophisticated and experienced in using the PAI, I found it more and more dif-
ficult to provide concise answers to their questions; such complex issues required
comprehensive explanations. It eventually became clear that an interpretive guide
was needed to consolidate what can be determined about some of the most com-
mon concerns of the assessment practitioner and researcher who is using the PAI.
This volume is an attempt to address that need.
A wealth of information about the use of the PAI has emerged since the test
manual was published. Some of this information grew out of my own use of the
PAI with individual clinical cases, and some was based on my work with data sets
gathered by colleagues in a wide variety of settings. Over time, a number of pat-
terns associated with various referral questions began to emerge. For example, a
particular PAI configuration appeared to be common among individuals who had
a history of violence, and a series of unusual profile elements tended to be seen
only in situations where the motivation to deny problems was quite high. From
such observations, I gradually began to articulate informal checklists, indicators to
look for when considering appropriate steps to take with a particular patient. Once
the lists of indicators were assembled, the next step was testing these informal
observations against the assembled data—at times, I found that some element |
had thought was unusual was in reality a fairly common finding among people in
general. Through this process of observation and verification, my informal hunches
became more explicit clinical guidelines for which norms and validity data could
be provided.
At the same time, I found I was becoming more inclined to view assessment
problems within a PAI framework. When participating in a discussion of a clinical
case for whom no PAI results were available, | would find myself wishing for three
or four specific bits of PAI profile data to address an unanswered question. When
reading a research article about the description, etiology, or treatment of a particu-
lar clinical condition, I would internally translate the relevant variables into the lan-
guage of the PAI scales. For example, when reading an article about the prediction

ili
of suicidal behavior, I would visualize how the author’ findings would appear if
mapped onto the PAI profile.
Among the most important clinical developments presented in this book are
several new interpretive indices based on this process of observation and verifica-
tion, using elements of the PAI profile to address specific assessment questions.
These indices are derived not only from my own work, but also from ground-
breaking research conducted by others since the PAI was introduced. These indices
include a number of actuarial checklists that may help to identify (a) individuals
who are attempting to distort their self-presentation, (b) individuals who deny
substance misuse, (c) individuals at risk for suicide or for violence, and (d) indi-
viduals who may be particularly difficult to engage in therapy. These indices pro-
vide a valuable supplement to the standard PAI scales, in many cases providing
alternative means of answering some of the most challenging issues in assessment.
I owe a debt of gratitude to the numerous students and PAI workshop partici-
pants whose challenging questions provided the stimulus for many of the ideas
presented here. I have tried to answer their questions as completely as I can, but
in many instances the answers should be considered only partial ones. The vali-
dation of a psychological test is never complete; for this reason, this book should
be regarded as an “interim report” on the state of the art of PAI interpretation.
Although the empirical database that serves as the foundation of this material is
already substantial, I look forward to watching it grow. As the literature on the PAI
continues to increase, subsequent editions of this book will encompass these new
developments. Hopefully, the book will serve as a catalyst for such developments,
as there are as many hypotheses remaining to be tested as there are clients in need
of assessment. I encourage all readers to contribute to the examination of these
hypotheses and to let me know about interesting results so that these new findings
can be incorporated into later editions of this Interpretive Guide.

L. Morey
Cambridge, Massachusetts

lv
 TABLE OF CONTENTS
[LTS V2 SUIS) os Ve vii
|SSVS1P Gove [LIT
UTENGYck ag pal ee cl ea ix
CHAPTER |
GENERAL INTRODUCTION AND OVERVIEW ..00-..cccccccccccececceccecceccecceceee ]
i esrALaRatonalerand Developimerituen)<. enemies
ALA. i
PS)Crepeye Wal ieee Rete heptane <ceenc gee Mee ie ok Sh LINO RR I, 7
FoetiAo iii moet HIE Le ews caicerre ores ree nies ae es) Pees Bate 10
Va lreity 0 thion, emit eh, Na at teh Ne be eo helt ade lip!
see DLC YC il ALCOVaoen 2°. Sa meet oe dee oe cas 18
CHAPTER 2
INTERPRETING PAI CLINICAL SCALE ELEVATIONS .........0.cccccccccscee- 23
Somat omiplamis (SONA: PA eT,SHS Les 23
CL CANS ye unentmnete
PALES arctan S ubeA CR Pd EOD Me be ae 29
PREE@ety = isorcers (ARD Wh
helateaib eeu oa iherect a ee 32
Depresstor (1)EP.) si 0m, Meenas’ earn ii be he eh i Me peri eca: 38
DViareamd AINE) Syme wiert Bi due PAS. Saami 4. Sh nn, Misa pts ear eh 42.
ERB VatG) GoAWG ecard gor LEER Oe ec Re Se ee 47
SO abVA) BURSAHh Wo!oy2) deyAa SERRE Fs «Rel eh ae PRO OACIE DET»Ge MORRORS REDE ce es a al
BoE eMIME Re aLUitC Sel LoWIN teat et ee eI ee err ee on eee ea 56
PMc tee CALUITCS CaeLa) er rece rere teeeca ieee nee 65
Sabsriie waists SCALCS teeter te ee et non ae eee a ee nen tenet Te
Perrinieeocotalice .Oilse Weial . wemema. Mer ces ct ca eee tare Me:
PICOMOMETOLIEMISN ALG tr earn en eee eee eee 78
Mrmrome LOD LeItion(Ci) erect renee etre eet ene cee eee Ae 79
CHAPTER 3
Two-POINT CODETYPES IN PROFILE INTERPRETATION .................. 81
CHAPTER 4
NEGATIVE DISTORTION: RANDOM RESPONDING AND
IVERTEI ING On
ERNG ae er en eet NT teakoctane tee Rnae 105
Detecting Careless or Idiosyncratic’ Responding niin oe reaaeg: 106
Détectine: Malineerino Wai, Jer epi i PAG Se ata ge htt rapls 111
Me gate ealumpness10nyUN IM) )-geatiemieis Set Bye borin Hee aoe jae
Die Am Malin geting Index (MAL) ssitin werercst sss. srscucndanersaave rene.
nsledan 120
PAL Malingering Index:(MAL) Interpretation a.2.1.. 0.0... 125
CHAPTER 5
IDENTIFYING DEFENSIVENESS ON THE PAL ....0.......0.0 cece 129
 Positive Impression (PIM) ........c:ccccccssseceeseeseesetaesteeneenseesenterecearesenaeenes 130
Defensive Responding Profile Configuration.............:::::c ccs 132
The PAL, Defensiveness. Index (DER). 2 euceee eset nee teeeeeeec Sie

CHAPTER 6
eceesc
teeoneee-
USE OF THE.PAI IN DIAGNOSIS «.0:c4::cstt scic:
acaneteteeeae see 143
Depression and Related Disorders ..........-:1::1:ceceees eseeee
tenet
tees ttre 146
Anxiety DisOrders <.c:.c2 eseicaawevegns (nc gta, ace ee 15)
Somatoform Disorders...... 0824 0) Ae ee eee 161
Psychotic Disorders is......+:--.v-:1s--2 saipmen
beeeal 1 Waehenehe mntein
eeecetera meee 164
Personality Disorders .crase:.21,1:-:.7> unrsdensa ites core eoipiessi bonne ee eee PS)
<a ean 2-<:005+5
Substance Abuse DisSOrdefs.............cpes-2sne+2¢¢ ee +
oe 189
aesss
casieee tora
Other Digenoses’ .....2. fer. sdeceesses ogee eetaeda
a s 194
CHAPTER 7
2:.ch
acoso sn
EVALUATING SUICIDE POTENTIAL .....22.2. aa ae 197
Suicidal: ldéation (SUI) neem 5 ee eee ee ee 197
PAI Profile Configurations and Suicidal Behavior ................:0::cecceeee 199
ThesPALSuicidesPotéential: nclex (SPL) 2 ee ree ee eee 201
CHAPTER 8
EVALUATING POTENTIAl FOR AGGRESSION eet eee 209
OT TAGGe eee
Ap oressi een ee er ne ee ee ee 210
PAI Protile*@onfiguration-and Ageressive Potential Sv ee ee 214
ThePALViolencerRotentialéindea Pl) me eaee eeee 215
CHAPTER 9
EVALUATING SPECIFIC “PSYCHOLOGICAT ISSUES 7 eee 223
SSeSSrienit OF SCL -CGOTICEDL sco. toaneee teen: ARM cere | ote amen 225
Ascesciment of Interpersonal Style V2 See. seer eer eee eo 232
ssesctient Ol Perceptlom Ol ENViITOnIetith nae. eeeeec es anne ieee ons Pal
CHAPTER 10
TREATMENT PLANNING AND MONITORING .........:ccccceceecceseeeeeeeeees 249
ren timents REJCCLORM CRA Late cect darectiieee Sie satcatre tee eee: en sce 249
Predicting Treatment Process: Impediments and Assets.............0..00.00084 ol
The PAT Ireatment’ Process Index (EPID 0"). ee ne ee eee peafe
Ditterential ‘Ireatment» Planning :::seess eene
isnt retoe: ea neree 262
specifying Therapeutic Targets), 20 tet AG BeSha eae 2th
The-PAl in the Evaluation-of Change !y::.2a0 Slee ee..9ee 204
REFERENCES stosuntecc victimes tea as eigen Mee ee 281
APPENDIX
INDEX CORRELATIONS WITH CLINICAL, PERSONALITY, AND
VALIDITY INDICATORS ¥ eet a cas eee ayy 291
ENDEX jonni sis tok cdedgiis Cee retest Asay NaN talc ney ee 2 305
 LIST OF TABLES

1-1 PAI Full Scales and Their Descriptions

1-2 PAI Subscales and Their Descriptions
1-3 Summary of Significant Gender, Race, and Age Influences on
PEE OOM Ores MAE Rens besceceichl sive eeeey Pimemteet et cou cian RA, ICONS
2-1 Cleckley’s (1941) 16 Diagnostic Indicators of Psychopathy .........0.0.00ccccccecseeee
2-2 Predicted ALC and DRG Scores from the Sum of BOR-S, ANT-A, ANT-E,
AN lesmancen( Grew) abel demerit sncenliltl. LOO mele tes gona leyers” Lette’...
Observed and Estimated ALC and DRG Scores for Groups from the
PEEPS SITE BUST) SUNThapa er ey aes een
Discriminant Function Weights Used in Computation of the Discriminant
PUIMCHOMAON Irie \ssessient Of NIA GETING vcs: cagcseseceneqes-<arcqearenepteangitatexasts
Descriptive Statistics for the Rogers Discriminant Function (RDF) for
Relevant Samples from the PAI Professional Manual ...............cccccceccesseseseeseeeees
T-Score Conversions for Rogers Discriminant Function (RDF) Scores,
Standardized Against the PA] -Nonnative Samplent: sndissndits savaitarts-eraeouebeads.
Selected Correlates of the Rogers Discriminant Function (RDF) Score..............
4-5 Prevalence of Features of the PAI Malingering Index (MAL) in Community,
Giaiea Rand imulatin champ)essen sere. garsrtuete Pee aecds NES ae snes one
T-Score Equivalents for the PAI Malingering Index (MAL) Standardized
Against Community and Clinical Normative Samples..........0......::c:eeeeees
Selected Correlates of the PAI Malingering Index (MAL) Total Score ................
Discriminant Function Means for Honest and Dissimulating Groups ...............
T-Score Conversions for the Cashel Discriminant Function (CDF) Results
Siancandized Against the PAL Normative Samples Se. tea carinasiate
Selected Correlates of the Cashel Discriminant Function (CDF) Score .............
Prevalence of Features of the PAI Defensiveness Index (DEF) in Community,
GsMnatea erie iieii
lat iesSATIS Le. aa acce ten.apace e gues ig ban cae set toen omycecanriataasn te nsued:
T-Score Equivalents for the PAI Defensiveness Index (DEF) Standardized
AGarisied GommMinity NOlMalive Sat Plea reemt te elena nea gcean cane
5-6 Selected Correlates of the PAI Defensiveness Index (DEF) Total Score..............
Summary of Key PAI Diagnostic Indicators for Depressive Disorders ...............
6-2 Summary of Key PAI Diagnostic Indicators for Anxiety Disorders ..................-.
Summary of Key PAI Diagnostic Indicators for Somatoform Disorders.............
6-4 Summary of Key PAI Diagnostic Indicators for Psychotic Spectrum
Pster NN lh es hs ee bea gin cadet tat castdessia ventions inagans pasate
Summary of Key PAI Diagnostic Indicators for Personality Disorders...............
7-1 PiietPAll eaicide Potential Index (SEL iam tes ecs havastsccbeyncdancnndenoecontenae eseeaeen
7-2 PAI Suicide Potential Index (SPI) Means and Standard Deviations in
Ree atte SANE [CS estateee apon ne rence eae lara rere rerde ans ars neaning sd cancaatenngtce néedingsneoehs
7-3 T-Score Equivalents for the PAI Suicide Potential Index (SPI)
Standardized Against Community and Clinical Normative Samples ..............-.. 204
7-4 Factors of the PAI Suicide Potentialidindex (SPI) Sixres ena
res ere eet 205
Characteristics of the PAI Suicide Potential Index (SPI) Factor Scores .............. 206

7-6 Selected Correlates of the PAI Suicide Potential Index (SPI) Total Score ........... 207,
The PAI Violence Potential Index (VP) sz ae ee eee ee 218
PAI Violence Potential Index (VPI) Means and Standard Deviations in
Relevant: Sari ples gcg2sctecacsycsess aguas iv tauaee duit pases ese ee eee ere neta 219
T-Score Equivalents for the PAI Violence Potential Index (VPI)
Standardized Against Community and Clinical Normative Samples ................. 220
Selected Correlates of the PAI Violence Potential Index (VPI) Total Score......... 221
indicators of Suitability for Exploratory Therapy gees eee ee ee 252
Operationalization of the Items of the PAI Treatment Process Index (TPI)........ 256
T-Score Equivalents for the PAI Treatment Process Index (TPI)
Standardized Against Community and Clinical Normative Samples ................. 257
Selected Correlates of the PAI Treatment Process Index (TPI) Total Score......... 258
Selective Patient Variables for Psychodynamic Therapy ............:::ccte:eeeeeeteeeees 268
Selective Patient Variables for Gognitive: Therapy......e
eee eee ee 269
Selective PatientrVariables tor interpersonal@inerapy 05 ee eee ee 270
InGicauions tor ¢-barmacothetapy Of Depression ana arte eee 250
Pre=Post Chaniges On. PA Scales: ment tile, Cube te ete) le aah eeta eee 279
index Cormelations With Al Full. Scale-ccores a5 esate
ee ee 292
index, CorrelationsiWithjPAL.Subscale,Scotes® (A ar 4 eee ee ee 293
Index Correlations With MMPI Clinical and Validity Scales..........0..000cccccecee 294
Index Correlations With Wiggins MMPI Content Scales .........0..0.cccccecceceeeeeee: 295
Index Correlations With Indicators of Clinical Symptomatology ..............0.0.00.. 296
Index Correlations With MMPI Personality Disorder Scales...............0cccccecesees 297
Index Correlations With Personality Disorder Questionnaire—Revised
bie Il-E Personality, DisorderyscaleyScores tte ose ee ee 298
Index Correlations With Diagnostic Interview for Personality Disorder
DSM-lI-RPersonality Misorder Diagnoses, seme. 5 eres too ae 299
Index Correlations With Brief Psychiatric Rating Scale (BPRS)
CUINIGiaPRAUINPos cece, Pamala ecanccet oer ate cet a aera 300
Index Correlations With Indicators of Personality and Environment................ 301
Index Correlations With Indicators of Clinical Symptomatology .............:c0.00.-- 302
Index Intercorrelations in Clinical and Community Samples............cccccccc0c00.:. 303

Vili
 LIST OF FIGURES

1-1 Mean PAI T-Scores for a clinical sample of adults (N = 1,246) and the
skyline at 2 SD above the mean in that clinical sample
2-1 Mean profiles for groups denying substance abuse problems, adapted from
Fals-Stewart (1996)
4-] PAI profile for 1,000 protocols (Morey, 1991) using random responding
simulations
4-2 PAI profiles for malingered mental disorder (Morey, 1991) and malingered
clepressiom, mvormec amd naive (Gales, 1903) yee caccicseneeeriestvansesceroapsve
sees
5-1 Mean “Fake-good” PAI profiles for prison inmates and students...............0...00..
6-1 Mean profiles for Adjustment Disorder, Dysthymic Disorder, and Major
Deptessive Disorder samples on the-PAL (Morey, 1901) io cncceccctezenssortuncrssoheress
6-2 Mean PAI profiles for depression and primary affective disorder samples.........
6-3 Mean PAI profiles for psychiatric patients reporting childhood abuse and
PUIeHbste IACTOSEC VILE PALSL) sys. cpncancn Meee tts Oe esaartek- eee Uaranancs astigs cess vit
6-4 Mean PAI profile for patients with Generalized Anxiety Disorder
UNO cg IE 8 6 sd me act ri en aa tena ce ORR a oe ae
6-5 Mean PAI profiles for male and female general medical patients who
completed the computer-administered version of the PAI (Osborn, 1994).......
6-6 Mean PAI profiles for patients diagnosed with Schizoaffective Disorder
ATICROE HIZO
IT CTIA ba eames Ne.) Claes che aE Aig, Tse Sea Meh L RRR
6-7 Mean PAI profiles for current antipsychotic medication, auditory
hallucinations, and persecutory delusions samples (Morey, 1991)...
6-8 Mean PAI profiles for two groups diagnosed with Borderline Personality
(BLEPECTSES es aps tendae eS he RR take eee ere. oe ete mn Eee eee EOIN
6-9 Mean PAI profiles for patients diagnosed with Antisocial and Dependent
Personality Disoraers (Morey 1991) Re cuerg re. tare ryeueerare erg ee cays ne
6-10 Mean PAI profiles for alcohol and drug abuse samples (Morey, 1991)..............
6-11 Mean PAI profiles for alcoholic and methadone maintenance samples..............
7-1 Mean PAI profiles for patients with a suicide history, patients on current
suicide precautions, and patients with a history of self-mutilation
CMR yal OO ee seeccearen suet ite ene ter<Micaretcnag rencecaat <1 ade teen one ctenasnsere vrs cnetea ca ates
Mean PAI profiles for patients on current assault precautions and patients
with a history of assaultive behavior (Morey, 1991)..........: cece
Mean PAI profiles for incarcerated rapists and spouse abusers in treatment
GIVE yalONL. teents cervtaag). eeenc). cs adores nar tree eee ee ah autem ve cite a ae. Seegereee ads

1x
 ~

a '

| af “y ¢.
' Aa ! a"

j "1a

a 4 ‘ou pe f :

jd geensrer yi !y How prea LJ

i
pL) ae io Sere iO) 7 : ,
rast |
: ‘i hea
entsbwy, ste
We ,
Pages |
Reais , j
(OP eees
IM { iene 8
.

y Dtectatutio dcocad belts eutpicm

baseless
reg tae
ute falilag
rie |

fe .
: .

Abowut
un

| uinod: kee
’

mast eiAMTCTS um ceil staan

<< oY oo s byes at) td ae)

aletoni) snk were jeep 1! otiderng neohd.

7 rer ee Saft Sages Lh ant Suni rane. Ye e
ORM STS ae bevitaT Jotarony daca €bi alent
wi eadihyt: int &
A Fahy ite CiOPeTav thy saath iio fake fs
— r
PZehsitiediase ac mins
_ pa pide i et a aaa OR
ee v=
mean
m7 f= el P Ps
 CHAPTER 1
GENERAL INTRODUCTION
AND OVERVIEW

The Personality Assessment Inventory (PAI; Morey, 1991) is a self-administered,

objective test of personality and psychopathology designed to provide information
on critical client variables in professional settings. From its inception, the PAI was
developed to provide measures of constructs that are central in treatment plan-
ning, implementation, and evaluation. Although it was introduced fairly recently,
the PAI already has generated considerable attention from clinicians and
researchers, and the test has been described as a “substantial improvement from a
psychometric perspective over the existing standard in the area” (Helmes, 1993,
p. 417) and as “one of the most exciting new personality tests” (Schlosser, 1992,
p. 12). The various applications of the test have generated findings that are impor-
tant considerations in the interpretation of the test. The purpose of this interpre-
tive guide is to integrate this recent work, and in doing so, to provide specific
interpretive information about the use of the PAI in addressing questions central
to the clinician and the researcher. This chapter will provide a summary of basic
psychometric information about the test, including reliability and validity studies.
Subsequent chapters will be devoted to the use of the PAI in addressing specific
clinical issues.

The PAI: Rationale and Development

The development of the PAI was based upon a construct validation framework
that emphasized a rational as well as quantitative method of scale development.
This framework placed a strong emphasis on a theoretically informed approach to
the development and selection of items and on the assessment of their stability and
correlates. The theoretical articulation of the constructs to be measured was
assumed to be critical, because this articulation had to serve as a guide to the con-
tent of information sampled and to the subsequent assessment of content validity,
In this process, both the conceptual nature and empirical adequacy of the items
played an important role in their inclusion in the final version of the inventory. Tic
development of the test went through four iterations in a sequential construct
PAI Interpretive Guide

validation strategy similar to that described by Loevinger (1957) and Jackson

(1971), although a number of item parameters were considered in addition to
those described by these authors. Of paramount importance in the development
of the test was the assumption that no single quantitative item parameter should
be used as the sole criterion for item selection. An overreliance on a single para-
meter in item selection typically leads to a scale with one desirable psychometric
property and numerous undesirable ones.
As an example, each PAI scale was constructed to include items addressing the
full range of severity of the construct, including both its milder as well as most
severe forms. Such coverage would not be possible if a single item selection crite-
rion was applied; “milder” items would be most effective in distinguishing clinical
subjects from normal respondents, while items reflecting more severe pathology
would be more useful in discriminating among different clinical groups. Also,
item-total correlations for such different items would be expected to vary as a com-
position of the sample, due to restriction of range considerations; milder items
would display higher biserial correlations in a community sample, whereas more
severe items would do so in an inpatient psychiatric sample. Thus, items selected
according to a single criterion (e.g., discrimination between groups or item-total
correlation) are doomed to provide limited coverage of the full range of sympto-
matology and/or severity of a clinical construct. The PAI sought to include items
that struck a balance between different desirable item parameters, including con-
tent coverage as well as empirical characteristics, so that the scales could be useful
across a number of different applications.
The clinical syndromes assessed by the PAI were selected on the basis of two
criteria: the stability of their importance within the nosology of mental disorder,
and their significance in contemporary diagnostic practice. These criteria were
assessed through a review of the historical and contemporary literature as well as
through a survey of practicing diagnosticians. In generating items for these syn-
dromes, the literature on each clinical syndrome was examined to identify those
components most central to the definition of the disorder, and items were written
directed at providing an assessment of each component of the syndrome in question.
The test itself contains 344 items that are answered on a four-alternative scale,
with the anchors Totally False, Slightly True, Mainly True, and Very True. Each
response is weighted according to the intensity of the feature that the different
alternatives represent. Thus, a client who answers Very True to the item “Sometimes
I think I'm worthless” adds 3 points to his or her raw score on the Depression scale,
whereas a client who responds Slightly True to the same item adds only 1 point.
The use of this four-alternative scaling is justified psychometrically in that it allows
 General Introduction and Overview

a scale to capture more true variance per item, meaning that even scales of mod-
est length can achieve satisfactory reliability. It is also justified clinically, because
sometimes even a Slightly True response to some constructs (e.g., as suicidal
ideation) may merit clinical attention. Furthermore, clients themselves often
express dissatisfaction with forced choice alternatives, expressing the belief that
the true state of affairs lies somewhere “in the middle” of the two extremes
presented.
The 344 items of the PAI comprise 22 nonoverlapping full scales: 4 validity, 11
clinical, 5 treatment consideration, and 2 interpersonal scales. Ten of the full scales
contain conceptually derived subscales that were designed into the test to facilitate
interpretation and coverage of the full breadth of complex clinical constructs. A
brief description of the PAI full scales is provided in Table 1-1; Table 1-2 presents
a description of the PAI subscales.

Table 1-1
PAI Full Scales and Their Descriptions
Scale (designation) Description
Validity Scales
Inconsistency (/CN) Determines if client is answering consistently throughout
inventory. Each pair consists of highly correlated (positively
or negatively) items.
Infrequency (/NF) Determines if client is responding carelessly or randomly.
Items are neutral with respect to psychopathology and have
extremely high or low endorsement rates.
Negative Impression (N/M) Suggests an exaggerated unfavorable impression or malin-
gering. Items have relatively low endorsement rates among
respondents in clinical settings.
Positive Impression (P/M) Suggests the presentation of a very favorable impression or
reluctance to admit minor flaws.

Clinical Scales
Somatic Complaints (SOM) Focuses on preoccupation with health matters and somatic
complaints associated with somatization and conversion
disorders.
Anxiety (ANX) Focuses on phenomenology and observable signs of anxiety
with an emphasis on assessment across different response
modalities.
Anxiety-Related Disorders (ARD) Focuses on symptoms and behaviors related to specific
anxiety disorders, particularly phobias, traumatic stress, and
obsessive-compulsive symptoms.
Depression (DEP) Focuses on symptoms and phenomenology of depressive
disorders.
(continued)
PAI Interpretive Guide

Table 1-1 (continued)

PAI Full Scales and Their Descriptions
Scale (designation) Description

Clinical Scales (continued)

Mania (MAN ) Focuses on affective, cognitive, and behavioral symptoms of
mania and hypomania.
Paranoia (PAR) Focuses on symptoms of paranoid disorders and more
enduring characteristics of paranoid personality.
Schizophrenia (SCZ) Focuses on symptoms relevant to the broad spectrum of
schizophrenic disorders.
Borderline Features (BOR) Focuses on attributes indicative of a borderline level of per-
sonality functioning, including unstable and fluctuating inter-
personal relations, impulsivity, affective lability and instability,
and uncontrolled anger.
Antisocial Features (ANT) Focuses on history of illegal acts and authority problems,
egocentrism, lack of empathy and loyalty, instability, and
excitement-seeking.
Alcohol Problems (ALC) Focuses on problematic consequences of alcohol use and
features of alcohol dependence.
Drug Problems (DRG) Focuses on problematic consequences of drug use (both
prescription and illicit) and features of drug dependence.
Treatment Scales
Aggression (AGG) Focuses on characteristics and attitudes related to anger,
assertiveness, hostility, and aggression.
Suicidal Ideation (SU/) Focuses on suicidal ideation, ranging from hopelessness to
thoughts and plans for the suicidal act.
Stress (STR) Measures the impact of recent stressors in major life areas.
Nonsupport (VON ) Measures a lack of perceived social support, considering both
the level and quality of available support.
Treatment Rejection (RXR) Focuses on attributes and attitudes theoretically predictive
of interest and motivation in making personal changes of a
psychological or emotional nature.
Interpersonal Scales
Dominance (DOM) Assesses the extent to which a person is controlling and inde-
pendent in personal relationships. A bipolar dimension with a
dominant style at the high end and a submissive style at the
low end.
Warmth (WARM)
Assesses the extent to which a person is interested in sup-
portive and empathic personal relationships. A bipolar dimen-
sion with a warm, outgoing style at the high end and a cold,
rejecting style at the low end.
eee
 General Introduction and Overview

Table 1-2
Ee PAI Subscales and Their Descriptions
RES 0 sie niece eae
Subscale (designation) Description

Somatic Complaints
Conversion (SOM-C) Focuses on symptoms associated with conversion disorder,
particularly sensory or motor dysfunctions.
Somatization (SOM-S) Focuses on the frequent occurrence of various common
physical symptoms and vague complaints of ill health and
fatigue.
Health Concerns (SOM-H) Focuses on a preoccupation with health status and physical
problems.
Anxiety
Cognitive (ANX-C) Focuses on ruminative worry and concern about current
issues that result in impaired concentration and attention.
Affective (ANX-A) Focuses on the experience of tension, difficulty in relaxing,
and the presence of fatigue as a result of high perceived
stress.
Physiological (ANX-P) Focuses on overt physical signs of tension and stress, such
as sweaty palms, trembling hands, complaints of irregular
heartbeats, and shortness of breath.

Anxiety-Related Disorders
Obsessive-Compulsive (ARD-O) Focuses on intrusive thoughts or behaviors, rigidity, indeci-
sion, perfectionism, and affective constriction.
Phobias (ARD-P) Focuses on common phobic fears, such as social situations,
public transportation, heights, enclosed spaces, or other
specific objects.
Traumatic Stress (ARD-T) Focuses on the experience of traumatic events that cause
continuing distress and that are experienced as having left
the client changed or damaged in some fundamental way.

Depression
Cognitive (DEP-C) Focuses on thoughts of worthlessness, hopelessness, and
personal failure, as well as indecisiveness and difficulties in
concentration.
Affective (DEP-A) Focuses on feeling of sadness, loss of interest in normal
activities, and anhedonia.
Physiological (DEP-P) Focuses on level of physical functioning, activity, and energy,
including disturbance in sleep pattern and changes in appetite
and/or weight loss.

Mania
Activity Level (MAN-A) Focuses on overinvolvement in a wide variety of activities in a
somewhat disorganized manner and the experience of accel-
erated thought processes and behavior.
Grandiosity (MAN-G) Focuses on inflated self-esteem, expansiveness, and the
belief that one has special and unique skills or talents.
(continued)
PAI Interpretive Guide

Table 1-2 (continued)

ions
PAI Subscales and Their DescriptGA rere
Tr Rip tp a na, 1 SS
Subscale (designation) Description

Mania (continued)
Irritability (MAN-/) Focuses on the presence of strained relationships due to the
respondent's frustration with the inability or unwillingness of
others to keep up with their plans, demands, and possibly
unrealistic ideas.

Paranoia
Hypervigilance (PAR-H) Focuses on suspiciousness and the tendency to monitor the
environment for real or imagined slights by others.
Persecution (PAR-P) Focuses on the belief that one has been treated inequitably
and that there is a concerted effort among others to under-
mine one’s interests.
Resentment (PAR-R) Focuses on a bitterness and cynicism in interpersonal rela-
tionships, and a tendency to hold grudges and externalize
blame for any misfortunes.

Schizophrenia
Psychotic Experiences (SCZ-P) Focuses on the experience of unusual perceptions and
sensations, magical thinking, and/or other unusual ideas that
may involve delusional beliefs.
Social Detachment (SCZ-S) Focuses on social isolation, discomfort and awkwardness in
social interactions.
Thought Disorder (SCZ-T) Focuses on confusion, concentration problems, and disorga-
t
nization of thought processes.

Borderline Features
Affective Instability (BOR-A) Focuses on emotional responsiveness, rapid mood changes,
and poor emotional control.
Identity Problems (BOR-/) Focuses on uncertainty about major life issues and feelings
of emptiness, unfulfillment, and an absence of purpose.
Negative Relationships (BOR-N) Focuses on a history of ambivalent, intense relationships in
which one has felt exploited and betrayed.
Self-Harm (BOR-S) Focuses on impulsivity in areas that have high potential for
negative consequences.
Antisocial Features
Antisocial Behaviors (ANT-A) Focuses on a history of antisocial acts and involvement in
illegal activities.
Egocentricity (ANT-E) Focuses on a lack of empathy or remorse and a generally
exploitive approach to interpersonal relationships.
Stimulus-Seeking (ANT-S) Focuses on a craving for excitement and sensation, a low
tolerance for boredom, and a tendency to be reckless and
risk-taking.
: (continued)
 General Introduction and Overview

Table 1-2 (continued)

PAI Subscales and Their Descriptions
Subscale (designation) Description
Aggression
Aggressive Attitude (AGG-A) Focuses on hostility, poor control over anger expression, and
a belief in the instrumental utility of aggression.
Verbal Aggression (AGG-V) Focuses on verbal expressions of anger ranging from
assertiveness to abusiveness, and a readiness to express
anger to others.
Physical Aggression (AGG-P) Focuses on a tendency to physical displays of anger,
including damage to property, physical fights, and threats of
violence.

Normative Data
The PAI was developed and standardized for use in the clinical assessment of
individuals in the age range of 18 through adulthood. The initial reading level
analyses of the PAI test items indicated that reading ability at the fourth-grade level
was necessary to complete the inventory. Subsequent studies of this issue (e.g.,
Schinka & Borum, 1993) have supported the conclusion that the PAI items are
written at a grade equivalent lower than estimates for comparable instruments.
PAI scale and subscale raw scores are transformed to T scores in order to provide
interpretation relative to a standardization sample of 1,000 community-dwelling
adults. This sample was carefully selected to match 1995 U.S. census projections
on the basis of gender, race, and age; the educational level of the standardization
sample was selected to be representative given the required fourth-grade reading
level. The only stipulation for inclusion in the standardization sample (other than
stratification fit) was that the respondent had to endorse more than 90% of PAI
items (i.e., no more than 33 items could be left blank). No other restrictions based
upon PAI data were applied in creating the census-matched standardization sample.
The PAI T scores are calibrated to have a mean of 50 and a standard deviation
of 10, using a standard linear transformation from the community sample norms.
Thus, a T-score value greater than 50 lies above the mean in comparison to the
scores of respondents in the standardization sample. Roughly 84% of nonclinical
respondents will have a T score below 60 (i.e., 1 SD above the mean) on most
scales, whereas 98% of nonclinical respondents will have scores below 70 (i.e., 2
SD above the mean). Thus, a T score at or above 70 represents a pronounced devi-
ation from the typical responses of adults living in the community.
PAI Interpretive Guide

For each scale and subscale, the T scores were linearly transformed from the
means and standard deviations derived from the census-matched standardization
sample. Unlike many other similar instruments, the PAI does not calculate T scores
differently for men and women; instead, the same (combined) norms are used for
both genders. This is because separate norms distort natural epidemiological dif-
ferences between genders. For example, women are less likely than men to receive
a diagnosis of antisocial personality, and this is reflected in lower mean scores for
women on the Antisocial Features (ANT) scale. A separate normative procedure for
men and women would result in similar numbers of each gender scoring in the
clinically significant range, a result that does not reflect the established gender
ratio for this disorder. The PAI development included several procedures designed
to eliminate items that might be biased due to demographic features (e.g., race,
gender, or age), and items that displayed any signs of being interpreted differently
as a function of these features were eliminated in the course of selecting the final
test items. As it turns out, with relatively few exceptions, differences as a function
of demography were negligible in the community sample. Table 1-3 lists all PAI
variables for which any of three demographic variables (i.e., race, gender, or age)
accounted for more than 5% of the variance in the PAI score and the resulting
effect (in terms of T-score units) of that variable.

Table 1-3
Summary of Significant Gender, Race, and
Age Influences on PAI Scale Scores
Demographic Primary subscales
PAI Scale influences affected
PAR Non-White: + 6T PAR-H
18-29 years: + 5T PAR-P
60+ years: -—4T PAR-R
BOR 18-29 years: +67 BOR-I
60+ years: —4T BOR-I
ANT Male: + 37 ANT-A
18-29 years: +/7T ANT-S
60+ years: —-4T ANT-A
AGG 18-29 years: +5T AGG-V
60+ years: —-4T AGG-P
STR 18-29 years: +47 (no subscales)
60+ years: -—4T
eee
 General Introduction and Overview

T scores are derived from a representative community sample; therefore, they

provide a useful means for determining whether certain problems are clinically sig-
nificant, because relatively few normal adults will obtain markedly elevated scores.
However, other comparisons are often of equal importance in clinical decision-
making. For example, nearly all patients report depression at their initial evalua-
tion, the question confronting the clinician considering a diagnosis of major
depression is one of relative severity of symptomatology. Knowing that an individ-
ual’s score on the PAI Depression scale is elevated in comparison to the standard-
ization sample is of value, but a comparison of the elevation relative to a clinical
sample may be more critical in forming diagnostic hypotheses.
To facilitate these comparisons, the PAI profile form (shown in Figure 1-1) also
indicates the T scores that correspond to marked elevations when referenced
against a representative clinical sample. The profile “skyline” indicates the score for
each scale and subscale that represents the raw score that is 2 standard deviations
above the mean for a clinical sample of 1,246 patients selected from a wide vari-
ety of different professional settings. Thus, roughly 98% of clinical patients will
obtain scores below the skyline on the profile form. Therefore, scores above this

PROFILE FORM FOR ADULTS-SIDEA 1 2 3 4 5 6 % 8 9 10 ret A B c D E Y z

— Clinical
sample mean
(Morey, 1991)
Profile skyline

Oosaiog,

io)3°

a 2 3 4 5 6 iv! 8 10 a A 8 c b E y z
scz BOR ANT ALC DRG AGG ‘SUI STR NON RXR DOM WRM
ICN INF NIM PIM SOM ANX ARO DEP MAN PAR

at 2 SD
Figure 1-1. Mean PAI T scores for a clinical sample of adults (N = 1,246) and the skyline
above the mean in that clinical sample.
PAI Interpretive Guide

skyline represent a marked elevation of scores relative to those of patients in clini-

cal settings. Thus, interpretation of PAI profiles can be accomplished in compari-
son to both normal and clinical samples.
The PAI Professional Manual (Morey, 1991) provides normative transformations
for a number of different comparisons. The appendices provide T-score transfor-
mations referenced against the clinical sample and a large sample of college stu-
dents (N = 1,051), as well as for various demographic subgroups of the commu-
nity standardization sample. Although the differences between different
demographic groups were generally quite small, there are occasions where it may
be useful to make comparisons with reference to particular groups. The raw score
means and standard deviations needed to convert raw scores to T scores with ref-
erence to normative data provided by particular groups (men, women, Blacks, and
respondents over age 60) are provided in the manual for this purpose. However,
for most clinical and research applications, the use of the T scores derived from the
full normative data is strongly recommended, because this sample was both large
and representative of the general population.

Reliability of the PAI

The reliability of the PAI has been examined in a number of different studies
that have examined the internal consistency, test-retest reliability, and configural
stability of the instrument.
The internal consistency of the PAI has been examined in a number of differ-
ent populations (Alterman et al., 1995; Boyle @ Lennon, 1994; Morey, 1991;
Rogers, Flores, Ustad, & Sewell, 1995; Schinka, 1995). This has involved the use
of coefficient alpha (Cronbach, 1951), which can be interpreted as an estimate of
the mean of all possible split-half combinations of items. The internal consistency
alphas for the PAI full scales are satisfactory; in the PAI Professional Manual, Morey
reports median alphas for the full scales of .81, .82, and .86 for normative, college,
and clinical samples, respectively. As expected, the scales tend to appear more
internally consistent in more heterogeneous samples. Alterman et al. found a
median alpha of .78 in a sample of methadone maintenance patients; Schinka
found a median alpha of .86 for full scales and .77 for the subscales in an alcoholic
sample. Boyle and Lennon (1994) reported a median alpha of .84 in a mixed
clinical—-normal sample. Internal consistency estimates for the ICN and INF scales
are consistently lower than those for other scales, because these scales do not mea-
sure theoretical constructs; instead, they measure the care with which the respon-
dent completed the test. Lower alphas for such scales would be anticipated, as

10
 General Introduction and Overview

carelessness might vary within a given sitting (e.g., a respondent might complet
e
the first half of the test accurately, but complete the last half haphazardly).
The lowest internal consistency estimates for the PAI reported in the literature
were obtained using the Spanish version of the instrument (Rogers et al., 1995),
where an average alpha of .63 was obtained. Rogers and colleagues concluded that
the internal consistency of the treatment consideration scales seemed to be most
affected by the translation of the test. Examination of internal consistency
estimates for the PAI full scales for groups defined by various demographic char-
acteristics (Morey, 1991) does suggest that there is little variability in internal
consistency (i.e., median scale alphas) as a function of race (i.e., Whites = .77,
non-Whites = .78), gender (i.e., men = .79, women = .75), or age (i.e., under 40
years = ./9, 40 years and over = .75).
The temporal stability of PAI scales has been examined by administering the
test to respondents on two different occasions (Boyle & Lennon, 1994; Morey,
1991; Rogers et al., 1995). For the standardization studies, median test-retest reli-
ability over a 4-week interval for the 11 full clinical scales was .86 (Morey, 1991),
leading to standard error of measurement estimates for these scales on the order of
3 to 4 T-score points, with 95% confidence intervals of +6 to 8 T-score points.
Examination of the mean absolute T-score change values for scales also revealed
that the absolute changes over time were quite small, on the order of 2 to 3 T-score
points for most of the full scales (Morey, 1991). Boyle and Lennon (1994) reported
a median test-retest reliability of .73 in their normal sample over 28 days. Rogers
et al. (1995) found an average stability of .71 for the Spanish version of the PAI,
administered over a 2-week interval.
Because multiple-scale inventories are often interpreted configurally, additional
questions concerning the stability of configurations on the 11 PAI clinical scales
are necessary. One such analysis (Morey, 1991) examined the inverse (or Q-type)
correlation between each respondent’ test and retest profiles. Correlations were
obtained for each of the 155 respondents in the full retest sample, and a distribu-
tion of these within-subject profile correlations was obtained. Conducted in this
manner, the median correlation over time of the clinical scale configuration was
.83, indicating a substantial degree of stability in profile configurations over time.

Validity of the PAI

The validation of measures of clinical constructs is a process that requires accu-
mulation of data concerning convergent and discriminant validity correlates. In
Morey’s (1991) examination of PAI validity, a number of the best available clinical

i
PAI Interpretive Guide

indicators were administered concurrently to various samples to determine their

convergence with corresponding PAI scales. Furthermore, diagnostic and other
clinical judgments concerning clinical behaviors (as rated by the treating clinician)
were also examined to determine whether their PAI correlates were consistent with
hypothesized relationships. Finally, a number of simulation studies were per-
formed to determine the efficacy of the PAI validity scales in identifying response
sets. To date, a number of studies have been conducted examining correlates of
various PAI scales; in the PAI Professional Manual, Morey provides information
about correlations of individual scales with over 50 concurrent indices of psycho-
pathology. Noteworthy findings from these studies are described in the following
paragraphs.
The PAI validity scales were developed to provide an assessment of the poten-
tial influence of certain response tendencies on PAI test performance. Two of these
scales, Inconsistency (ICN) and Infrequency (INF), were developed to assess devi-
ations from conscientious responding, whereas the other two validity scales, Neg-
ative Impression (NIM) and Positive Impression (PIM), were developed to provide
an assessment of efforts at impression management by the respondent.
To model the performance of respondents completing the PAI in a random
fashion, computer-generated profiles were created by generating random
responses to individual PAI items and then scoring all scales according to their
normal scoring algorithms. A total of 1,000 simulated protocols were generated for
this analysis. Comparison of profiles derived from normal respondents, clinical
respondents, and the random response simulations demonstrated a clear separation
between scores of actual respondents and scores from the random simulations:
99.4% of these random profiles were identified as such by either ICN or INF
(Morey, 1991).

To model the performance of respondents attempting to manage their impres-

sions in either a positive or negative direction, studies have been performed
(Morey, 1991) in which respondents were instructed to simulate such response
styles. Comparisons of profiles for normal respondents, clinical respondents, and
the corresponding response style simulation groups demonstrated a clear separa-
tion between scores for the actual respondents and scores for the simulated
response groups. Respondents scoring above the critical level of NIM were 14.7
times more likely to be a member of the malingering group than of the clinical
sample, whereas respondents scoring above threshold on PIM were 13.9 times
more likely to be from the positive dissimulation sample than from a community
sample. Subsequent studies generally support the ability of these scales to distin-
guish simulators from actual protocols under a variety of response set conditions

12
 General Introduction and Overview

(e.g., Cashel, Rogers, Sewell, & Martin-Cannici, 1995; Rogers, Ornduff, & Sewell,
1993). Results of these studies are reviewed in greater detail in chapters 4 and 5.
In addition to such simulation studies, a number of correlational studies have
been performed to determine the convergent and discriminant validity of the PAI
validity scales as measured against other commonly used measures of similar con-
structs (Ban, Fjetland, Kutcher, & Morey, 1993; Costa & McCrae, 1992: Morey,
1991). For example, NIM correlated significantly (r = .54) with the Minnesota
Multiphasic Personality Inventory (MMPI; Hathaway & McKinley, 1967) F scale:
PIM was associated with the Marlowe-Crowne (Crowne & Marlowe, 1957) Social
Desirability scale (r = .56) as well as with the MMPI K (r = .47) and L (7 = 41)
scales (Morey, 1991). PAI scales INF and ICN displayed negligible correlations with
any measures, an expected result as these scales were designed as relatively pure
indicators of measurement error.

The clinical scales of the PAI were assembled to provide information about
critical diagnostic features of 11 important clinical constructs. A number of dif-
ferent validity indicators have been used to provide information on the conver-
gent and discriminant validity of the PAI clinical scales; these indicators can be
divided into measures of “neurotic features, ” “psychotic features,” and “behavior
disorder features.” Within the neurotic spectrum, correlations with the NEO Person-
ality Inventory (NEO-PI; Costa & McCrae, 1985), the MMPI clinical and research
scales (Hathaway & McKinley, 1967; Morey, Waugh, & Blashfield, 1985; Wiggins,
1966), and several specialized assessment instruments have been examined. These
specialized instruments include the following: the Wahler Physical Symptoms Inven-
tory (Wahler Inventory; Wahler, 1983), a broad measure of somatic complaints; the
Beck Depression Inventory (BDI, Beck & Steer, 1987), the Beck Anxiety Inventory
(BAI; Beck & Steer, 1990) and the Beck Hopelessness Scale (BHS; Beck & Steer,
1988), three widely used and well-validated measures of negative affect; the
Hamilton Rating Scale for Depression (HAM-D: Hamilton, 1960), perhaps the
most widely used measure of outcome in treatment studies of depression; the
State-Trait Anxiety Inventory (STAI; Spielberger, 1983), a widely used measure
that distinguishes between the situational and more enduring elements of anxiety;
the Fear Survey Schedule (FSS; Wolpe & Lang, 1964), a comprehensive assess-
ment of common fears; the Maudsley Obsessive-Compulsive Inventory (Maudsley
Inventory; Rachman & Hodgson, 1980), a measure of severe obsessional ideation
and contamination fears; and the Mississippi Scale for Combat-Related Posttrau-
matic Stress Disorder (Mississippi PTSD; Keane, Caddell, & Taylor, 1988).
Correlations between each of the full scale scores for the four PAI neurotic
cluster scales and the validation measures described above follow hypothesized

13
PAI Interpretive Guide

patterns, demonstrating strong associations with other measures of neuroticism

(Costa & McCrae, 1992; Montag & Levin, 1994; Morey, 1991). The strongest
correlates for Somatic Complaints (SOM) were found with the Wiggins Health
Concerns (r = .80) and Organic Problems (r = .82) content scales, the Wahler
Inventory (r = .72), and the MMPI Hypochondriasis (r = .60) scale. Each of these
measures is a fairly straightforward assessment of complaints regarding physical
functioning, so this pattern of correlations is consistent with expectations. The
SOM scale also displays small-to-moderate relationships with measures of distress,
such as anxiety or depression. The SOM scale is generally the highest point of the
PAI profile in a general medical population, although, even in such populations,
the average score is typically below 70T (Osborne, 1994).
The Anxiety (ANX) scale demonstrated substantial correlations with a number
of measures of negative affect, including the NEO-PI Neuroticism (r = .76) and
Anxiety (r = .76), the STAI Trait Anxiety Inventory (r = .73), and the Wiggins
Depression content (r = .76) scales. This finding is consistent with research results
highlighting the prominent role of anxiety in many mental disorders; such a pat-
tern should be anticipated, as ANX was intended to be a general measure of anxi-
ety rather than a specific diagnostic indicator. In contrast, the Anxiety-Related
Disorders (ARD) scale was designed to provide content relevant to more specific
diagnostic differentiations; hence, the pattern of correlations tends to be more spe-
cific than that observed with ANX. The largest correlation for ARD was with the
Mississippi PTSD scale (r = .81), and the second largest involved the FSS (r = .66);
each of these scales directly parallels a disorder for which ARD was designed to
provide coverage. The ARD scale has also been found to correlate with the proba-
bility of getting nightmares (r = .46), with ARD-T (r = .51), in particular, being
associated with night terrors (Greenstein, 1993). The ARD scale (particularly
ARD-T) also has been found to differentiate between women psychiatric patients
who were victims of childhood abuse and women patients who did not experience
such abuse (Cherepon & Prinzhorn, 1994).

The Depression (DEP) scale demonstrates its highest correlations with various
well validated indicators of depression, such as the BDI (r = .81), the HAM-D (r =
.78), and the Wiggins Depression content scale (r = .81). This is consistent with
expectations, because these measures are widely used in the assessment of depres-
sion and related symptomatology. Other noteworthy correlates of the Depression
scale include the MMPI D scale (r = .66), the Wiggins Poor Morale scale (r = .74),
the NEO-PI Neuroticism (r = .69) and Depression (r = .70) scales, and the Beck
Hopelessness scale (r = .67).

14
 General Introduction and Overview

In addition, correlations with a number of other measures of related constructs

can provide information relevant to the convergent and discriminant validity of the
PAI “psychotic cluster” scales. For example, the MMPI, the NEO-PI, the Interper-
sonal Adjective Scale (IAS-R; Trapnell & Wiggins, 1990) and the clinician-rated
Brief Psychiatric Rating Scale (BPRS; Overall & Gorham, 1962) include scales that
capture the cognitive and interpersonal abnormalities that characterize these dis-
orders. Correlations between each the three PAI psychotic spectrum scales and
these validation measures generally follow the expected pattern (Ban et al., 1993:
Costa & McCrae, 1992; Morey, 1991). The Mania (MAN) scale has demonstrated
its strongest correlations with Wiggins Hypomania (r = .63), Psychoticism (r =
58), and Hostility (r = .55) content scales; with the BPRS clinical ratings of
Grandiosity (r = .48) and Conceptual Disorganization (r = .40); and with the
MMPI Ma scale (r = .53). The Paranoia (PAR) scale demonstrated its largest corre-
lations with the MMPI Paranoid personality disorder scale (r = .70), the Wiggins
Psychoticism scale (r = .60), and various measures of hostility such as the Wiggins
Hostility content scale (r = .54) and the NEO-PI Hostility facet scale (r = .55). A
moderate correlation with the MMPI Pa scale was also observed (r = .45). The
Schizophrenia (SCZ) scale has been found to correlate with the Wiggins Psychoti-
cism content scale (r = .76) and the MMPI Schizotypal (r = .67) and Paranoid (r =
.66) personality disorder scales. The SCZ scale was also positively correlated with
the MMPI Sc scale (r = .55) and negatively associated with indices of sociability
and social effectiveness such as the NEO-PI Agreeableness (r = —.49) and Gregar-
iousness (r = —.57) scales. This pattern indicates that scores on the SCZ scale reflect
disruptions in both the cognitive (e.g., delusions, hallucinations) and the interper-
sonal (e.g., limited social competence) realms of functioning. Finally, the SCZ scale
has been found to distinguish schizophrenic patients from controls (Boyle &
Lennon, 1994). In that study the schizophrenic sample did not differ significantly
from a sample of alcoholics on SCZ scores, although the article suggested that
many of the alcoholic patients completed the PAI during detoxification, which
might complicate differential diagnosis based solely upon SCZ scores.
Information on the convergent and discriminant validity of the PAI scales in
the behavior disorders cluster is also available. In addition to the NEO-PI, the
IAS-R, and the MMPI, the PAI scales have been correlated with a number of spe-
cialized assessment instruments, including the Bell Object Relations Inventory (Bell
Inventory; Bell, Billington, & Becker, 1985), a multifactorial questionnaire con-
structed to measure a variety of interpersonal attitudes and beliefs indicative of early
pathological object relations thought to be at the core of the borderline syndrome

15
PAI Interpretive Guide

(Bell, Billington, Cicchetti, & Gibbons, 1988); the Michigan Alcoholism Screening
Test (MAST; Selzer, 1971), a widely used and well validated measure of problem
behaviors associated with drinking; the Drug Abuse Screening Test (DAST, Skin-
ner, 1982), a measure, patterned after the MAST, that assesses the consequences of
drug abuse; and the Self-Report Psychopathy test designed by Hare (1985) to
assess his model of psychopathy.
Correlations between scores for the four PAI behavior disorder cluster scales
and these validation measures follow expected patterns (Costa @ McCrae, 1992;
Kurtz, Morey, & Tomarken, 1993; Morey, 1991). The strongest correlates of the
Borderline Features (BOR) scale are the MMPI Borderline personality disorder scale
(r = .77), the NEO-PI Neuroticism scale (r = .67), and several different measures
of hostility, such as the NEO-PI Hostility facet (r = .70). The BOR scale also dis-
played substantial correlations with the Bell Inventory Insecure Attachment scale
(r = .63), the NEO-PI Impulsiveness facet (r = .52), and the Wiggins Family Prob-
lems (r = .63) and Psychoticism (r = .63) content scales. This pattern of anger,
impulsiveness, and interpersonal clashes is consistent with the core features of the
borderline syndrome. Other studies have supported the validity and utility of this
scale in a variety of clinical contexts. The BOR scale in isolation has been found to
distinguish borderline patients from unscreened controls with an 80% hit rate; it
successfully identified 91% of these respondents as part of a discriminant function
(Bell-Pringle, 1994). Classifications based on the BOR scale have been validated in
a variety of domains related to borderline functioning, including depression, per-
sonality traits, coping, Axis I disorders, and interpersonal problems (Trull, 1995).
These BOR scale classifications were also found to be predictive of 2-year outcome
on academic indices in college students, even controlling for academic potential
and diagnoses of substance abuse (Trull, Useda, Conforti, @ Doan, 1995).

The PAI Antisocial Features (ANT) scale demonstrated its largest correlations
with the Hare Psychopathy Scale (r = .82) and the MMPI Antisocial personality
disorder scale (r = .77). Other correlates included the Wiggins Hostility (r = .57)
and Family Problems (r = .52) content scales, the NEO-PI Excitement Seeking
facet (r = .56), and the IAS-R cold interpersonal octant (r = .45). This pattern sug-
gests that the ANT scale addresses the personality, interpersonal, and behavioral
elements of psychopathy. The correlation with the MMPI Pd scale is positive, but
not impressive (r = .34), suggesting that the two scales represent the core features
of the disorder somewhat differently. The PAI Alcohol Problems (ALC) and Drug
Problems (DRG) scales each demonstrate a similar pattern of correlates: strong cor-
relations with corresponding measures of substance abuse and moderate associa-
tions with indicators of antisocial personality. ALC yields a correlation of .89 with
 General Introduction and Overview

the MAST, whereas DRG correlates .69 with the DAST. The ALC scale has been
found to differentiate patients in an alcohol rehabilitation clinic from both patients
with schizophrenia and normal controls (Boyle & Lennon, 1994). The DRG scale
has also been found to successfully discriminate drug abusers and methadone
maintenance patients from general clinical and community samples (Alterman et
al., 1995).

The treatment consideration scales of the PAI were assembled to provide indi-
cators of potential complications in treatment that would not necessarily be appar-
ent from diagnostic information. There are five of these scales: two indicators of
potential for harm to self or others, two measures of the respondent's environ-
mental circumstances, and one indicator of the respondent’s motivation for treat-
ment. These scales have been compared to a number of measures of related
constructs. In addition to the NEO-PI, the IAS-R, and the MMPI, the scales have
been correlated with a number of specialized assessment instruments. The BDI,
BAI, and BHS provide convergent correlates for suicidal ideation. Also, the Suicide
Probability Scale (SPS; Cull & Gill, 1982) serves as a concurrent indicator of sui-
cide potential. The SPS has four subscales that assess hopelessness, suicidal
ideation, negative self-evaluation, and hostility, in addition to yielding a total score
for suicide probability. The State-Trait Anger Expression Inventory (STAXI;
Spielberger, 1988) provides a marker for aggression that is broken down into six
scales and two subscales. The Perceived Social Support scales (Procidano & Heller,
1983) provide an assessment of the subjective impact of supportive transactions
between the respondent and his or her social support system; two separate scales
assess support provided by the respondent's family and the respondent’ friends.
Finally, the Schedule of Recent Events (SRE; Holmes @ Rahe, 1967) is a unit-
scoring adaptation of the widely used Holmes and Rahe (1967) checklist of
recent stressors, where respondents are asked to indicate major life changes that
have taken place during the 12 months prior to evaluation.
Correlations between the PAI treatment consideration scales and such valida-
tion measures provide support for the construct validity of these PAI scales (Costa
& McCrae, 1992; Morey, 1991). Substantial correlations have been identified
between the Aggression (AGG) scale and the NEO-PI Hostility (r = .83) and STAXI
Trait Anger (r = .75) scales. The AGG scale also was negatively correlated with the
STAXI Anger Control scale (r = -.57). The Suicidal Ideation (SUI) scale was most
positively correlated with the BHS (r = .64), the BDI (r = .61), the Suicidal Ideation
(r = .56) and Total Score (r = .40) of the SPS; it also was found to be negatively
correlated with the measures of perceived social support. As expected, the Non-
support (NON) scale was found to be highly (and inversely) correlated with the
social support measures: —.67 with PSS-Family and —.63 with PSS-Friends. NON

ad
PAI Interpretive Guide

also was moderately associated with numerous measures of distress and tension.
The Stress (STR) scale displayed its largest correlations with the SRE (r = .50) and
also was associated with various indices of depression and poor morale. Finally,
the Treatment Rejection (RXR) scale was found to be negatively associated with
Wiggins Poor Morale (r = —.78) and the NEO-PI Vulnerability (r = -.54) scales, con-
sistent with the idea that distress can serve as a motivator for treatment. The Treat-
ment Rejection scale has been shown to be positively associated with indices of
social support (r = .26 to .49), suggesting that people are less likely to be motivated
for treatment if they have an intact and available support system as an alternative.
The interpersonal scales of the PAI were designed to provide an assessment of
the interpersonal style of respondents along two dimensions: (a) a warmly affilia-
tive versus a cold rejecting axis, and (b) a dominating and controlling versus a
meekly submissive style. These axes provide a useful way of conceptualizing vari-
ation in normal personality as well as in many different mental disorders, and per-
sons at the extremes of these dimensions may present with a variety of disorders.
The PAI Professional Manual (Morey, 1991) describes a number of studies indicat-
ing that diagnostic groups differ on these dimensions; for example, spouse-abusers
are relatively high on the Dominance (DOM) scale, whereas schizophrenics are low
on the Warmth (WRM) scale. Correlations with related measures also provide sup-
port for the construct validity of these scales. For example, the correlations with
the IAS-R vector scores are consistent with expectations, with PAI DOM associated
with the IAS-R Dominance vector (r = .61) and PAI WRM associated with the IAS-R
Love vector (r = .65). The NEO-PI Extroversion scale roughly bisects the high
DOM/high WRM quadrant, as it is moderately positively correlated with both
scales; this finding is consistent with previous research (Trapnell & Wiggins,
1990). The WRM scale was also correlated with the NEO-PI Gregariousness scale
(r = .46), whereas DOM was associated with the NEO Assertiveness facet (r = .71).

In summary, the PAI scales have been found to associate in theoretically con-
cordant ways with most major instruments for the assessment of diagnosis and
treatment efficacy. Strategies for the interpretation of the PAI profile and its use in
treatment planning and evaluation are presented in following sections.

Basic Interpretive Strategy

Because the development of the PAI emphasized the importance of both conver-
gent and discriminant validity of the instrument, the interpretation of PAI protocols
is relatively straightforward. For example, scales were designed to be generally
 General Introduction and Overview

pure measures of the specific constructs; thus, an elevation on the DEP scale may
be interpreted as indicating that the respondent reports a number of experiences
consistent with the symptomatology of clinical depression. Interpretive hypothe-
ses may be generated at four different levels: the item level, the subscale level, the
full scale level, and the configuration level.
Interpretation of PAI responses at the item level are meaningful because the
content of each item was assumed to be critical in determining its relevance for the
assessment of the construct. For example, each item was reviewed by a panel of
experts to ensure that its content was directly relevant to the specific clinical con-
struct. As a result, a review of item content can provide specific information about
the nature of the difficulties experienced by the respondent. In addition, 27 PAI
items were identified as “critical items” based on two criteria: (a) importance of
their content as an indicator of potential crisis situations, and (b) very low
endorsement rates in normal individuals. Endorsement of any of these items
should be followed by more detailed questioning that can clarify the nature and
severity of these concerns.
The PAI subscales were constructed as an aid in isolating the core elements of
the different clinical constructs measured by the instrument. These subscales can
serve to clarify the meaning of full scale elevations, and may be used configurally
in diagnostic decision-making. For example, many patients typically come to clin-
ical settings with marked distress and dysphoria; this often leads to elevations on
most unidimensional depression scales. However, unless other manifestations of
the syndrome are present, this does not necessarily indicate that Major Depres-
sive Disorder is the likely diagnosis. In the absence of features such as vegetative
signs, lowered self-esteem, and negative expectancies, the diagnosis may not be
warranted even with a prominent elevation on a unidimensional depression scale.
On the PAI, such a pattern would lead to an elevation on DEP-A, representing the
dysphoria and distress, but no elevations on DEP-P (the vegetative signs) and
DEP-C (the cognitive signs). As a result, an overall elevation on DEP in this
instance would not be interpreted as diagnostic of major depression because of the
lack of supporting data from the subscale configuration.
Interpretation of PAI full scale scores is aided by comparison to two referents:
expected scores in the community and expected scores in clinical patients. As
described earlier, the PAI profile form (Figure 1-1) provides a skyline marking an
elevation of 2 standard deviations with respect to the clinical sample. The similar-
ity of expected scores for these two populations varies a great deal across scales.
For example, the interpersonal scales DOM and WRM have distributions that are

19
PAI Interpretive Guide

quite similar in both community and clinical samples; thus, marked elevations (or
very low scores) are noteworthy regardless of the nature of the client. On the other
hand, the RXR scale (which was designed to identify risk for early treatment ter-
mination) has a markedly different distribution in clinical and community sam-
ples. A majority of clinical respondents who are currently in treatment obtain
scores that are considerably lower than those of community respondents, who are
typically not in psychological treatment and have little interest in it. Thus, a T
score of 50 on RXR ina client presenting for psychotherapy, although “average” for
a community sample, is actually considerably above the expected score for respon-
dents in clinical settings. In this instance, the RXR score should be interpreted as
indicating potentially significant resistance to change for this client. In contrast, an
RXR score of 50T in an individual who was administered the PAI for personnel
selection purposes would be unremarkable. In these two examples, the differences
in the assessment question leads to differences in the interpretation of the infor-
mation yielded by a normative transformation.
The broadest level of PAI interpretation involves the analysis of scale configu-
ration. Traditionally, the premise behind multidimensional inventories such as the
PAI has been that the combination of information provided by the multiple scales is
greater than any of its parts; hence, most previous research focused on the profile
yielded by such an inventory, rather than any single scale elevations. There are a
variety of ways to examine profile configuration; to date, there have been five
research approaches to studying the configural use of PAI profile data. These
approaches include the use of mean profiles, profile codetypes, cluster profiles,
actuarial functions, and conceptually driven configural decision rules. These dif-
fering approaches can be applied to different issues in decision-making, including
diagnostic (e.g., Is this a schizophrenic patient or a depressed patient?), interven-
tion (e.g., Does this patient require inpatient treatment?), or protocol-related
(e.g., Is this a valid PAI protocol?) issues. Each of these approaches will be dis-
cussed throughout this guide in the context of these different types of decisions.
The following chapters will focus on the four different interpretive levels in an
effort to resolve certain dilemmas the test user may face in interpreting the PAI.
The initial focus is on understanding the composition and interpretation of the
individual scales; this is followed by a discussion of the meaning of different two-
point combinations of scales (codetypes). The remainder of this interpretive guide
explores specific issues commonly encountered in PAI interpretation: Is this
patient malingering or defensive? What diagnoses should be considered? What is
the person’ characteristic view of self and of others? What initial steps should be
considered in planning treatment? In all cases, the available data are used to
address these questions, but, as is the case with any assessment instrument, many

20
 General Introduction and Overview

questions require further study. It is hoped that current and future PAI users can
help to fill the gaps in this literature, so that subsequent editions of this guide can
incorporate the advances made possible by such work. _

apie see hali

lala Oaret « as
a> f i
(7a) (enh Qaine.: eli Peart
? fae. = od

5 He amew De i
Secintoviping aus ois instil,
nik shone riba. @
-

ole a
~~ yl eo

>
 x.

¢ 4

C7

~ +t

re VERE the
‘ \ .
nat ; = Ck a Te re ,

~ ene
is aad : TI phe =~

ara y vy - Vise aiie

_

pot y i.
eee ee rat je é
re
ma sig! civ aN gin mi edges~ ae "a 1 :
Si
f ' age ae? * 7

eae ie, ew: aa Coes, | Yi oe.

7 yy 7 aks —
 CHAPTER 2
INTERPRETING PAI CLINICAL
SCALE ELEVATIONS

The Starting point in interpreting the PAI lies at the level of the individual
scales that were developed to measure the specific construct implied by the scale
name. Each scale on the test was designed to measure the major facets of a differ-
ent clinical construct, as determined by current theoretical and empirical work on
those constructs. Most of the clinical scales offer subscales. Therefore, configural
interpretation of the test is possible even at the level of the individual scales,
because two identical elevations on a particular scale may be interpreted quite dif-
ferently depending on the configuration of the subscales. The following sections
describe the logic underlying the PAI clinical scales and the interpretations of dif-
ferent ranges and configurations of scores on each scale.

Somatic Complaints (SOM)

The Somatic Complaints scale is precisely what the scale name suggests: the
items reflect complaints and concerns about physical functioning and health mat-
ters in general. Interpretively, there are many things that the SOM scale should not
be expected to do. In isolation, SOM cannot distinguish between functional and
organic somatic features. It is not a neuropsychological assessment instrument, and
it certainly is not sufficient evidence for establishing a diagnosis of a physical con-
dition. However, the scale is useful for assessing the extent to which physical con-
ditions are a central concern in an individual’ life. It is important to recognize that
people with very similar physical conditions can differ drastically in their reactions
to the condition. For example, one person faced with a crippling chronic condi-
tion might react stoically, successfully adapting to any impairment and, perhaps,
refusing to acknowledge the limitations imposed by his or her health. Another per-
son, faced with the same condition, might ruminate bitterly about these limita-
tions, complaining endlessly about physical problems and, perhaps, even using the
problems as a means of controlling other people. There are valid physical problems
in both situations, but the psychological reaction to the problems is quite differ-
ent. The SOM scale provides information about the latter, but it should not be used
in isolation to determine the former.

23
PAI Interpretive Guide

The Diagnostic and Statistical Manual of Mental Disorders (DSM-IV; American

Psychiatric Association, 1994) classification system groups a variety of syndromes
under the concept of Somatoform Disorder; all involve physical symptoms sug-
gestive of some organic disorder, but one for which there are no known physio-
logic mechanisms. The constructs included in this group of disorders (e.g., con-
version hysteria and hypochondriasis) have had a variety of clinical meanings over
the years; for this reason, it is difficult to evaluate the results of diagnostic research
that has accumulated on this topic. One of the central distinctions drawn in recent
years has been between individuals who present with multiple, relatively minor
physiologic symptoms and individuals who complain of major disability of some
sensory or motor function. In the DSM manual, the former individuals are referred
to as having Somatization Disorder, whereas the latter are typically diagnosed with
Conversion Disorder. This distinction can be traced back to the early studies of
Briquet (1859), who found that most patients with “hysteria” displayed few of the
symptoms thought to be pathognomonic of the disorder. More recently, the symp-
tomatic approach of Briquet has been applied to contemporary diagnoses, with
research suggesting that “Briquets syndrome” (or Somatization Disorder, as it is
now called) reflects a distinct diagnostic entity from the traditional construct of
conversion hysteria (Guze, Woodruff, @& Clayton, 1971). Both disorders are
viewed as distinct from Hypochondriasis, which, in contemporary diagnostic prac-
tice, refers to a preoccupation with the fear or belief of having a disease.
The PAI SOM scale was designed to provide a differential assessment of some
of these components of somatoform disorders. The three subscales of SOM reflect
different facets of somatic complaints frequently associated with psychological
conditions. Although two of the subscale names reflect this association, one should
not assume that an elevation on one of these scales indicates that the diagnosis is
present, as for each of those diagnoses the presumption is made that organic
factors have been ruled out. Rather, elevation indicates that the respondent is
reporting symptoms consistent with these disorders. To support such caution in
interpretation, the SOM scale is generally the highest point of the PAI profile in a
general clinical population, although, even in such populations, the average score
is typically below 70T (Osborne, 1994). Perhaps more than on any other scale, the
primary question about discriminant validity (ie., whether these might be valid
physical problems) lies outside of the domains measured by the PAI.

SOM-C: Conversion
The Conversion subscale includes items corresponding to the dramatic physi-
ological symptoms that have been found to be prevalent in conversion disorders
(Watson & Buranen, 1979). As it turns out, most of these symptoms involve

24
 Interpreting PAI Clinical Scale Elevations

unusual sensory-motor problems: impairments in perception (e.g., vision or hear-

ing problems, numbness) or motor problems (e.g., paralysis). The mean raw score
in the normative sample on the Conversion subscale was very low, indicating that,
for the most part, these symptoms are quite unusual in the general population.
Although such symptoms may be rare, there are some populations in which these
symptoms are more common, because there are a variety of physical conditions
that result in sensory-motor problems. For example, people with multiple sclero-
sis, stroke victims, and those with other neurological disorders all may have sensory-
motor problems. It has been observed that the SOM-C subscale is probably the most
sensitive scale on the PAI to various forms of Central Nervous System (CNS)
impairment. One diagnostic group that frequently obtains elevations on SOM-C
are chronic alcoholics who are beginning to experience some neuropsychological
compromise associated with their drinking. Often, clinicians will use indicators on
self-report personality inventories to distinguish a conversion reaction from a “gen-
uine” organic problem or to distinguish functional from organic pain, but, in actu-
ality, this diagnostic distinction should never be based solely on the results of such
tests. In such instances, a thorough medical evaluation is recommended.
Thus, an elevated score on SOM-C indicates a report of problems in physical
functioning due to symptoms often associated with conversion disorders, such as
sensory or motor dysfunctions. Such problems are likely to be unusual ones,
rather than a more severe form of more common problems such as headaches or
dizziness. Perhaps consistent with the notion of la belle indifference, the SOM-C
scale is relatively uncorrelated with other indicators of distress; thus, an isolated
elevation does not necessarily signify that the reported symptoms are of great con-
cern to the respondent (cf. the score on DEP and the SOM-H subscale for such dis-
tress or preoccupation). Marked elevations could be a sign of (a) a debilitating
physical illness leading to marked sensorimotor impairment, (b) a rather dramatic
conversion reaction, or (c) severe hypochondriasis or, perhaps, even somatic delu-
sions. More moderate elevations would be expected in a person with a more cir-
cumscribed sensory or motor impairment, such as those associated with mild cere-
brovascular infarcts. Because of the rarity of these somatic signs in the general
population, SOM-C has a rather “hard floor,” and it is not possible to obtain
extremely low scores.
Individuals with SOM-C elevations are likely to report that their daily func-
tioning has been compromised by one or more serious and rather unusual physi-
cal problems. Although they may feel that their health is good in general, if the
other SOM subscales are not elevated, they will feel that the health problems that
they do have are complex and difficult to treat successfully. Physical complaints are

25
PAI Interpretive Guide

likely to focus on symptoms of distress in neurological and musculoskeletal sys-

tems, and may involve features often associated with conversion disorders, such as
unusual sensory or motor dysfunctions. As scores become extreme (i.e., 2 95T),
the possibility of somatic delusions should also be considered.

SOM-S: Somatization
The Somatization subscale inquires about routine physical complaints, such as
headaches, back problems, pain, or gastrointestinal ailments; these complaints are
diagnostic by virtue of their frequency rather than their presence. In comparison
to SOM-C, the Somatization subscale consists of complaints that are more vague
and diffuse, not localized in any one organ system. There are two components to
elevations on the subscale, one element involving the physical symptoms (which
can include a general lethargy and malaise), and a second element relating to a
more general complaintiveness and dissatisfaction. The SOM-S subscale yields
substantial correlations with measures of both psychological and physical distress;
individuals with SOM-S elevations are likely to have a litany of physical complaints
that they will share with anyone who will listen.
Individuals with SOM-S elevations will report that their daily functioning has
been compromised by numerous and varied physical problems. They will report
particular problems with the frequent occurrence of various minor physical symp-
toms and vague complaints of ill health and fatigue, often accompanied by unhap-
piness and bitterness about their health. This pattern of symptoms is often consis-
tent with a somatization disorder.

SOM-H: Health Concerns

The Health Concerns subscale indicates a preoccupation with health and phys-
ical functioning. Items on this subscale are related to the self-perceived complex-
ity of the individual’s health problems and the intensity of the individual’ efforts
to ameliorate these problems. The SOM-H subscale is a measure of focus rather
than of severity; a general medical population has a very wide distribution, and
individuals with serious health problems can still obtain low scores on this sub-
scale. Such people will tend to strike others as quite stoic about their problems,
whereas individuals with SOM-H elevations will tend to focus a great deal on their
health issues.
Individuals with elevations on SOM-H are likely to report that their daily func-
tioning has been compromised by numerous and varied physical problems. If the
other subscales are not elevated, such individuals may appear to be relatively
healthy to other observers, but they will see themselves as having a history of com-
plex medical problems. They will tend to feel that their health is not as good as

26
 Interpreting PAI Clinical Scale Elevations

that of their age peers, who may view such individuals as rather hypochondriacal.
There are likely to be continuous concerns with health status and physical prob-
lems, and the poor health may be a major component of the self-image, with such
individuals accustomed to being in the patient role.

SOM Full Scale Interpretation

As the sum of these three elements, the full scale of SOM reflects the degree of
concern about physical functioning and health matters and the extent of perceived
impairment arising from somatic symptoms. Average scores on SOM (i.e., < 60T)
reflect a person with few bodily complaints. Such individuals are typically seen as
optimistic, alert, and effective. Scores between 60T and 70T indicate some concern
about health functioning and will not be uncommon in older respondents or in
medical patients with relatively specific organic symptoms. Scores above 70T sug-
gest significant concerns about somatic functioning and probable impairment aris-
ing from somatic symptoms. Such a person will feel that his or her health is not as
good as that of age peers and is likely to believe that the health problems are com-
plex and difficult to treat successfully. For such people, social interactions and con-
versations are likely to focus often on their health problems, and self-image may
be largely influenced by the belief that they are handicapped by poor health. Indi-
viduals scoring in this range may be seen as unhappy, complaining, and pes-
simistic. They may be using somatic complaints to control others in a passive-
aggressive manner.
SOM scores that are markedly elevated (i.e., > 87T) are unusual even in clini-
cal samples; such scores suggest a ruminative preoccupation with physical func-
tioning and health matters and severe impairment arising from somatic symptoms.
In that range, the somatic complaints are likely to be chronic and accompanied by
fatigue and weakness that render the individual incapable of performing even min-
imal role expectations. Such scores require elevations on all three subscales,
reflecting a large number of somatic complaints affecting most organ systems,
including the neurological, gastrointestinal, and musculoskeletal systems. Scores
in this range will reflect a diagnosable somatoform disorder in most instances.
These patients may be resistant to psychological explanations for problems and
may be poor candidates for psychotherapy, particularly if there are few accompa-
nying indications of psychological distress.

SOM Subscale Configurations

The following sections describe some of the implications of particular combi-
nations of elevations on SOM subscales.

oi,
PAI Interpretive Guide

SOM-C high, SOM-S high, SOM-H high

Individuals with this subscale pattern will report that their daily functioning
has been compromised by numerous and varied physical problems. They feel that
their health is not as good as that of their age peers and are likely to believe that
their health problems are complex and difficult to treat successfully. Physical com-
plaints are likely to include symptoms of distress in several biological systems,
including the neurological, gastrointestinal, and musculoskeletal systems. The
pattern indicates the report of unusual sensorimotor symptoms as well as severe
manifestations of more ordinary complaints, such as headaches or pains. Such
individuals are likely to be continuously concerned with their health status and
physical problems, and social interactions and conversations will tend to focus on
their health problems. The self-image may be largely influenced by a belief that
they are handicapped by poor health, and such individuals may be quite accus-
tomed to being in the patient role.

SOM-C high, SOM-S high, SOM-H average

This subscale pattern is rather unusual, as it represents a report of numerous
and varied physical problems but relatively little focus on these problems. The
physical complaints are likely to include symptoms of distress in several biologi-
cal systems, including the neurological, gastrointestinal, and musculoskeletal sys-
tems. The item endorsement pattern indicates the report of symptoms consistent
with both conversion and somatization disorders. The lower scores on SOM-H
suggest less complaintiveness than is typical of individuals with SOM-S elevations.

SOM-C high, SOM-S average, SOM-H high

Individuals with this subscale pattern will report that their daily functioning is
impeded by unusual physical problems. They feel that their health is not as good
as that of their age peers and are likely to believe that their health problems are
particularly challenging and treatment resistant. Physical complaints are likely to
focus on symptoms of distress across varied physical systems, particularly neuro-
logical and musculoskeletal systems; these involve features often associated with
conversion disorders, such as unusual sensory or motor dysfunctions. Such peo-
ple tend to be continuously concerned with their health status and physical prob-
lems, and there may be underlying concerns about the ability of the medical sys-
tem to treat these problems effectively.

SOM-C average, SOM-S high, SOM-H high

People displaying this pattern are likely to report that they cannot function nor-
mally due to various physical problems. They feel that their health is not as good as
that of others, reporting that their health problems are complicated and difficult to

28
 Interpreting PAI Clinical Scale Elevations

treat successfully. They report particular problems with the frequent occurrence of
various minor physical symptoms (e.g., headaches, pain, or gastrointestinal prob-
lems) and vague complaints of ill health and fatigue. Health status and physical
problems are likely to be continuous concerns, and social interactions and conver-
sations will tend to focus on health problems. Marked dissatisfaction with the
quality and effectiveness of the care they have received is also likely.

Anxiety (ANX)
Anxiety is a prominent part of many of the major syndromes of mental disor-
der. Unfortunately, with respect to measurement it also represents one of the most
elusive psychological constructs. An important conceptualization by Lang (1971)
addressed some of these measurement difficulties by portraying anxiety as com-
prised of three components: “cognitive” (in a person’s thoughts), “somatic” (involv-
ing physiological reactions), and “behavioral” (observed in a person’s actions).
Lang viewed each of these three components as related but independent modes of
the expression of anxiety; as such, the comprehensive assessment of anxiety
involved the measurement of each individual component. Lang included the sub-
jective feeling of anxiety as part of the cognitive component of anxiety, but more
recent efforts (Zajonc, 1980) have distinguished between the affective and cogni-
tive experiences of emotion. Koksal and Power (1990) demonstrated that the cog-
nitive and affective components of anxiety were clearly related but could be reli-
ably differentiated by self-report methods and suggested that a comprehensive
assessment of anxiety includes an assessment of four systems: affective, cognitive,
behavioral, and somatic.
The ANX scale of the PAI was designed to assess three of these components of
anxiety; the behavioral component of anxiety was not included as a subscale. Spe-
cific behaviors often serve as the basis of making differential diagnostic decisions;
for example, avoidance behavior is a critical component of the definition of a pho-
bia, whereas ritualistic behavior is a critical sign of Obsessive-Compulsive Disor-
der. Thus, in the PAI, these specific behaviors were assessed in the context of a
scale (ARD) pertaining to specific anxiety-related disorders, as described in a later
section. This exclusion makes the scale a more general, nonspecific index of anx-
iety that does not have specific ties to a particular diagnostic construct. Rather, it
relates broadly to the experience of anxiety and to how it is typically expressed.

ANX-C: Cognitive
The Cognitive subscale of ANX includes items that tap an expectation of harm,
ruminative worry, and cognitive beliefs of the type described by Beck and Emery

29
PAI Interpretive Guide

(1979) within the context of cognitive therapy of anxiety disorders. This cognitive
component involves a ruminative form of anxiety expression, people operating in
this mode of expression tend to dwell on events, running them over and over in
their minds. This is an internalizing approach to anxiety; such people tend to be
vigilant to the experience of anxiety, rather than repressing it, and these feelings of
being ill at ease will tend to have an ideational target or source. This mode of anx-
iety expression also tends to have strong trait aspects, meaning that it is both a
characteristic style of dealing with anxiety and an indication of current distress.
Elevated scores on ANX-C indicate worry and concern about current issues to
a degree that may impair the person’s ability to concentrate and attend. Such peo-
ple are likely to be overly concerned about issues and events over which they have
no control. As scores exceed 85T, the worry and negative expectations are likely
to be debilitating, and the possibility of intrusive obsessions should be investi-
gated.

ANX-A: Affective
The Affective subscale includes items that measure the feelings of tension,
apprehension, and nervousness that are characteristic of anxiety. This anxiety
tends to be free-floating rather than attached to specific objects or events. Also, the
anxiety reflected in this subscale tends to be rather persistent and trait-like; it
reflects a dispositionally low threshold for the experience of events as alarming.
High scorers on this scale experience a great deal of tension, have difficulty relax-
ing, and tend to be easily fatigued as a result of constant apprehension and high
perceived stress. Elevations on this subscale in the absence of elevations on the
remaining ANX subscales are suggestive of generalized anxiety rather than more
specific fears.

ANX-P: Physiological
The Physiological subscale of ANX includes items that assess the somatic
expression of anxiety, such as racing heart, sweaty palms, rapid breathing, and
dizziness. This subscale has a fairly different pattern of relationships to other con-
structs than ANX-C and ANX-A. For example, ANX-P correlates most highly with
the state component (as opposed to the trait component) of the STAI. However,
this may, in part, be due to the nature of that instrument, as many of its “state”
items are physiological in nature, and mode of anxiety expression may be con-
founded with duration of anxiety on the STAI (Spielberger, 1983).
Another distinction of ANX-P is that it is associated much less with indicators of
depression and much more with physical symptom expression, as compared to
ANX-C or ANX-A. This distinction captures the difference between somatization and

30
 Interpreting PAI Clinical Scale Elevations

ideation. ANX-P correlates most highly with the expression of physical symptoma-
tology. People with this pattern may not psychologically experience themselves as
anxious, but they show physiological signs that most people associate with anxi-
ety. This suggests a repressive style of dealing with stress: the person may notice
overt physical signs such as sweaty palms and shortness of breath, and still not rec-
ognize these as signs of anxiety and stress.

ANX Full Scale Interpretation

As mentioned earlier, the full scale score of ANX is a nonspecific indicator of
the degree of tension and negative affect experienced by the respondent. Average
scores on ANX (i.e., < 60T) reflect a person with few complaints of anxiety or ten-
sion. Such individuals are typically seen as calm, optimistic, and effective in deal-
ing with stress. Very low scores (i.e., < 40T) are indicative of a person reporting
fearlessness, and it is possible that this represents a reckless lack of prudence in
certain situations. Scores between 60T and 70T are indicative of a person who may
be experiencing some stress and who is worried, sensitive, and emotional. Scores
above 7OT suggest significant anxiety and tension. With scores in this range, the
respondent is probably tense much of the time and ruminative about anticipated
misfortune. These individuals may be seen as high strung, nervous, timid, and
dependent. With scores above 7OT, at least one ANX subscale is likely to be ele-
vated and such elevations should be examined to determine the typical modality
in which anxiety is expressed.
ANX scores that are markedly elevated (i.e., > 90T) will likely have elevations
on all three subscales, reflecting a generalized impairment associated with anxiety.
Such a person’s life will be seriously constricted, and the individual may not be
able to meet even minimal role expectations without feeling overwhelmed. Mild
stressors are likely to precipitate a crisis, and this repeating pattern of crises may
present difficulties for psychotherapy despite the motivating nature of the individ-
ual’s distress. Scores in this range will reflect a diagnosable anxiety disorder in most
instances; scores on ARD may suggest a specific focus for the fears, or a lack of ele-
vation on ARD may suggest that the anxiety is free-floating and generalized.

ANX Subscale Configurations

The following sections describe some of the implications of elevations on two
or more ANX subscales.

ANX-C high, ANX-A high, ANX-P high

Individuals who have all three subscales elevated are likely to be plagued by
worry to a degree that interferes with their ability to concentrate, attend, and

ol
PAI Interpretive Guide

manage stressful periods in their lives. Anxiety is experienced in all modalities,

ideationally as well as physically. Such people will ruminate about issues and
events of seemingly minor significance and over which they have no control. There
is likely to be prominent motor tension, little capacity to relax, and a general
fatigue and malaise as a result of high perceived stress.

ANX-C high, ANX-A high, ANX-P average

This pattern suggests an ideational and sensitized approach to anxiety. The ten-
dency to dwell on decisions and issues most likely interferes with their ability to
concentrate and focus on matters at hand. The respondent’ level of tension and
difficulties in relaxing are probably readily apparent to others, who are likely to per-
ceive the respondent as worrying needlessly and excessively about most matters.

ANX-C high, ANX-A average, ANX-P high

This is an unusual configuration in that the person does not report a strong
subjective experience of tension or major difficulties in relaxing, yet there appears
to be considerable worry and tension surrounding specific events or issues, and
overt physical signs of tension and stress (e.g., sweaty palms, trembling hands,
complaints of irregular heartbeats, and shortness of breath) are also present. Such
a pattern suggests some denial or lack of recognition of the degree to which gen-
eralized stress is affecting the person’s functioning.

ANX-C average, ANX-A high, ANX-P high

The primary manifestations of the respondent’ anxiety appear to be in the
affective and physiological areas. Such people feel quite tense much of the time,
have difficulty relaxing, and are likely to experience considerable fatigue and
malaise as a result of high perceived stress levels. There may be a tendency to try
to handle stress by simply not thinking about the stressful issues, as ideation does
not appear to be a prominent component of the anxiety, but it is apparent that it
is being expressed in other ways, particularly in somatic form.

Anxiety-Related Disorders (ARD)

Anxiety is typically a feature in most clinical disorders, and, as such, an anxi-
ety scale such as ANX is of limited use in identifying specific disorders in which
anxiety may be prominent. The behavioral expression of anxiety, however, varies
across different disorders, and, as such, these different diagnostic syndromes are typ-
ically defined by characteristic behaviors. The ARD scale assesses phenomena central
to three important anxiety-related disorders that, in conjunction with marked anxi-
ety as measured by ANX, can serve as a more specific indicator of these disorders.

32
 Interpreting PAI Clinical Scale Elevations

ARD-O: Obsessive-Compulsive
The Obsessive-Compulsive subscale includes items related to both the symp-
tomatic features of the disorder (e.g., fears of contamination and performance of
rituals) and the personality elements of the disorder (e.g., perfectionism and hyper-
attentiveness to detail). In DSM-IV terms, these two components represent both
Axis I (clinical syndrome) and Axis II (personality trait) aspects of the disorder. The
Axis | component involves intrusive, recurrent thoughts, images, or behaviors; the
literature suggests a number of common themes to these thoughts, such as fears of
contamination leading to characteristic avoidance behaviors (e.g., hand-washing).
The Axis II component involves a personality style that is rigid, dogmatic, and
affectively constricted. For example, if you were to visit the house of an obsessional
individual and pick up an object, the Axis I obsessional would be concerned that
you left germs on the object, whereas the Axis II obsessional would be concerned
that you did not return the object to its proper place. Although these are fairly dif-
ferent responses to the situation, both are represented on ARD-O.
The correlational pattern of ARD-O suggests that the Axis II manifestations are
most heavily represented, as the scale is less correlated with traditional markers of
anxiety and neuroticism than other ARD subscales. This pattern suggests that high
scorers are using obsessional tactics to try to control anxiety (i.e., control through
order and predictability). The relatively lower associations with ANX, for example,
point out that there are a number of individuals who are successful in these efforts
(i.e., they have little subjective experience of anxiety). Thus, with ARD-O elevated
and the full-scale of ANX low, this suggests that the obsessional tactics are reason-
ably effective. However, this control of anxiety may be achieved at a cost; other
aspects of the test may reveal pronounced interpersonal problems (e.g., low WRM,
SCZ-S, BOR-N) associated with the individual's rigidity and need for control. How-
ever, as both ANX and ARD-O elevate, this is a sign that the obsessional tactics are
failing to control the anxiety.
By comparison to most other clinical subscales, elevations on ARD-O are less
frequent in clinical samples. This suggests that these behaviors and defenses are
more unusual in clinical samples, as compared to the straightforward experience
of anxiety. Thus, relatively moderate elevations (i.e., 55T to 65T) are interpretively
significant in the clinical settings. Such people may be seen by others as being
ruminating, detail-oriented, conforming, and somewhat rigid in attitudes and
behavior. Scores ranging from 65T to 75T suggest a fairly rigid individual who fol-
lows his or her own guidelines for personal conduct in an inflexible and unyield-
ing manner. Such people ruminate about matters to the degree that they often have
difficulty in making decisions and in perceiving the larger significance of decisions

33
PAI Interpretive Guide

they do make. Changes in routine, unexpected events, and contradictory informa-

tion are likely to generate untoward stress, and such individuals will be particu-
larly wary of situations with strong affective demands. Scores at or above /5T
indicate marked rigidity and significant ruminative concerns; intrusive thoughts
are likely to be present. Such people may fear their own impulses and doubt their
own ability to control them. They are likely to be extremely indecisive, and obses-
sional defenses are probably failing to control marked anxiety.

ARD-P: Phobias
The Phobias subscale assesses several of the more common phobic fears,
including heights, enclosed places, public transportation, and social exhibition.
These fears were selected based on commonality of reporting in the research liter-
ature—commonality within clinical, rather than research, settings. For example,
snake and insect phobias are frequent objects of study in research laboratories, yet
they constitute a fairly minor proportion of presenting complaints in anxiety dis-
order clinics. Given the prevalence of social phobias, these items are heavily rep-
resented on the scale, and elevations may indicate marked social anxiety. The
ARD-P subscale correlates well with most other indicators of phobic fears as well
as with indicators of more general anxiety.
The ARD-P scale is interesting in that it also has interpretive significance at very
low scores, as the scale has a rather soft floor. Raw scores of 0 or | place a person
at roughly 35T; such scores are typically obtained in people who regard themselves
as fearless, unafraid of anything, even at times when fear is merited. In such peo-
ple, there is a possibility of recklessness because they are not likely to be inhibited
by appropriate caution; such scores are sometimes obtained in psychopathic indi-
viduals. Scores in the range from 60T to 70T suggest the possibility of specific
fears, but avoidance behaviors are not likely to be severe and probably will not pre-
clude a relatively successful level of daily functioning. As scores elevate above 7OT,
phobic behaviors are likely to interfere in some significant way, and such people
will tend to monitor their environment in an effort to avoid contact with the feared
object or situation. Marked elevations indicate the likelihood of multiple phobias
or a more pervasive phobia, such as agoraphobia, as opposed to a simple, more
circumscribed phobia.

ARD-T: Traumatic Stress

The Traumatic Stress subscale concerns phenomena related to reactions to
traumatic stressors, including nightmares, sudden anxiety reactions, and feelings
of being irreversibly changed by a traumatic event. Items were not written to detail
the nature of the traumatic event; such events might include combat experiences,

34
 Interpreting PAI Clinical Scale Elevations

rape or abuse, or some other highly stressful experience. Positive responses to the
items indicate that (a) some terrible event or events happened to this person, and
(b) these events changed the person for the worse in some way.
In light of significant elevations on this subscale, the precise nature of the event
can be determined through a follow-up inquiry. The test score can serve as a use-
ful means of broaching a topic that an individual may not be willing to disclose
during an intake interview. The PAI assessment provides an opportunity to divulge
discomforting information. The information is divulged in a “safe” forum, as it is
simply a check mark on a piece of paper; however, including it with the rest of the
items also acknowledges to the respondent that these are important issues and that
it is acceptable to discuss such issues in the context of a professional assessment.
Because this scale is commonly elevated in clinical samples, it is often an entry to
further discussion while providing the client with feedback on test results. For
example, one might say, “I notice your score is very high on the traumatic stress
scale; this usually occurs with people who have had something very bad happen
to them that really changed their life, that really affected them in a negative way.
What do you think about that?” Although this interpretation is rather unexcep-
tional given the content of the items, clients are often impressed by the extent to
which they differ from others in this regard. In addition, the acknowledgment that
the clinician understands that these are particularly important issues for the client
is generally reassuring and increases the client’s confidence in the clinician.
One aspect of ARD-T that merits mention is that it is quite frequently elevated
in clinical settings; the average score for clinical respondents is 64T, which
approaches the 90th percentile for the general population. It should be recognized
that individuals in treatment settings tend to have very high rates of traumatic
events; prevalence of a history of physical and/or sexual abuse has been estimated
as high as 70-80% in some settings. However, the frequency of this elevation also
should serve as a caution against an indeterminate use of this scale as an indication
of posttraumatic stress disorder (PTSD), which tends to have a characteristic profile
that includes other features as well as ARD-T elevations (see chapter 6). PTSD is a
syndrome that is not limited to the particular feature identified by ARD-T, although
the scale is certainly a beginning point in the identification of this syndrome.
Scores in the moderately elevated range on ARD-T (i.e., 65T to 75T) suggest
that the respondent has likely experienced a disturbing traumatic event in the past,
an event that continues to be a source of distress and to produce recurrent episodes
of anxiety. Although the item content of the PAI does not address specific causes
of traumatic stress, possible traumatic events involve victimization (e.g., rape,
abuse), combat experiences, life-threatening accidents, and natural disasters. As

35
PAI Interpretive Guide

scores become increasingly elevated, preoccupation with the trauma increases, and
scores above 9OT indicate that the trauma (single or multiple) is the overriding
focus of the person’s life and that individual views himself or herself as having been
severely damaged, perhaps irreparably, by the experience.

ARD Full Scale Interpretation

The full scale of ARD is perhaps the most difficult to interpret on the inventory,
due to its composition of three fairly diverse conditions. In general, it is a measure
of the extent of behavioral expression of anxiety. Average scores on ARD (i.e., < 60T)
reflect a person who reports little distress across many situations. Such individu-
als are typically seen as secure, adaptable, and calm under fire. Scores between
60T and 70T reflect a person who occasionally experiences, or experiences only
to a mild degree, maladaptive behavior. patterns aimed at controlling anxiety.
Such people will have some specific fears or worries and also may have little self-
confidence. Scores above 70T suggest impairment associated with fears surround-
ing a particular situation; specific subscale elevations should reveal more precisely
the nature of these fears. Such individuals may be seen as insecure and self-doubt-
ing, ruminative, and particularly uncomfortable in social situations.
ARD scores that are markedly elevated (i.e., > 90T) are likely to have elevations
on all three subscales, reflecting multiple anxiety disorder diagnoses and broad
impairment associated with anxiety. These individuals are in severe psychological
turmoil; they are faced with constant rumination and often are guilt ridden over past
transgressions, whether real or imagined. A number of maladaptive behavior pat-
terns aimed at controlling anxiety are probably present, but these patterns are hay-
ing little effect in preventing anxiety from intruding into experience and functioning.

ARD Subscale Configurations

The following sections describe some of the implications when two or more
ARD subscales are elevated in combination.

ARD-O high, ARD-P high, ARD-T high

This pattern reveals that the respondent is likely to have significant symptoms
and behaviors related to anxiety in a variety of domains, including phobic avoid-
ance, obsessive rumination, and troublesome thoughts related to a traumatic
event. The resulting avoidance behaviors are likely to interfere with social role
functioning in some significant way; such people tend to monitor their environ-
ment constantly in a vigilant manner in an effort to avoid contact with particular
situations, particularly those that evoke a disturbing traumatic event in the past.
Although phobic fears are likely, such people are more likely to have multiple

36
 Interpreting PAI Clinical Scale Elevations

phobias or a more debilitating phobia, such as agoraphobia, than to suffer from a

simple phobia.
There appears to be an attempt, apparently unsuccessful, to control these anx-
ieties through rigidity and affective constriction. Such people are often seen by
others as being perfectionistic and overly anxious about trifles. They are likely to
set and follow their personal guidelines for conduct in an inflexible and unyield-
ing manner, but they pay for this lack of flexibility by ruminating about matters
(both past and present) to the degree that decisions cannot be made. Predictability
is very important for such people, and changes in routine, unexpected life events,
and contradictory information are likely to overtax the person’s efforts at control.
They also may fear their own impulses and doubt their ability to control them.

ARD-O high, ARD-P high, ARD-T average

This subscale pattern suggests a fearful individual who attempts to manage
anxiety through rigid planning and tries to avoid affective arousal. However, anx-
iety and avoidance behaviors are likely to be interfering in some significant way in
the individual’s life, and it is probable that such individuals monitor their sur-
roundings closely to avoid unexpected disruptions in routine. Such people tend to
fear novel situations and will avoid risk-taking as much as possible. This pattern,
particularly with a concomitant elevation on DOM, suggests a person who man-
ages this fear of novelty through a rigid and inflexible need for control. However,
this need for control is complicated by the tendency to constantly ruminate about
decisions and about the unexpected consequences of any decisions that are made.
Changes in routine, unexpected events, and contradictory information are likely to
be particularly difficult to handle.

ARD-O high, ARD-P average, ARD-T high

This pattern of responses suggests an individual who ruminatively dwells on
past events in his or her life. Such people attempt to manage the discomfort gen-
erated by these past events through affective constriction and by organizing their
lives in an inflexible and unyielding manner. Although these strategies may help
in managing anxiety, they fill the person with doubt and, hence, such individuals
will have difficulty in making personal decisions and in perceiving the larger con-
sequences of decisions they do make. Such people may particularly fear their own
impulses and doubt their ability to control them should their rigid efforts at self-
control fail.

ARD-O average, ARD-P high, ARD-T high

This subscale pattern reflects individuals who have experienced a disturbing
traumatic event in the past, an event that continues to serve as a source of marked

37
PAI Interpretive Guide

distress and to produce recurrent episodes of anxiety. Such people tend to vigi-
lantly monitor their environment in an effort to avoid situations reminiscent of
past stressful events; avoidance behaviors related to these fears are likely to be suf-
ficiently severe to interfere with social role functioning. Interpersonal withdrawal
in close relationships is likely (look for low scores on WRM), and multiple pho-
bias or a more distressing phobia, such as agoraphobia, may be present.

Depression (DEP)
The measurement of depression has perhaps received more research attention
than any other construct in mental disorders. There are a host of widely used
instruments for assessing depression, including the self-report Beck Depression
Inventory, the Zung (1965) Depression Scale, and MMPI D scale, as well as
observer rating scales such as the Hamilton Rating Scale for Depression (HAM-D;
Hamilton, 1960). Despite the fact that these scales are widely used and tend to be
positively correlated, each has somewhat different characteristics (Lambert, Hatch,
Kingston, & Edwards, 1986). For example, the BDI is based on the cognitive fea-
tures of depression, such as beliefs about helplessness and negative expectations
about the future (e.g., Louks, Hayne, & Smith, 1989). In contrast, the HAM-D
addresses vegetative signs of depression more heavily than the BDI; as a result, the
two instruments have substantially different factor structures (Favarelli, Albanesi,
& Poli, 1986). However, both instruments share the characteristic of having very
low mean scores and little variance in normal samples. In contrast, the MMPI D
scale has a relatively “soft floor” with greater variability among normal respon-
dents; thus, it may be more useful for the assessment of depressive features within
the milder ranges (Hollon & Mandel, 1979). However, the MMPI items emphasize
affective features such as unhappiness and psychological discomfort, with limited
assessment of either the cognitive or the physiological features of depression.
The DEP scale of the PAI was assembled to provide an equal weighting among
the major components of the depressive syndrome and still provide items that
would prove useful across the full range of severity of symptomatology. The clini-
cal syndrome of depression is typically found to have three components: an affec-
tive component, characterized by unhappy and apathetic mood; a cognitive com-
ponent, marked by negative expectancies; and a physiological component, where
sleep and appetite disturbances and low energy are prominent (e.g., Moran &
Lambert, 1983). Thus, three DEP subscales were designed: Cognitive, to tap neg-
ative expectancies, helplessness, and cognitive errors of the type described by Beck

38
 Interpreting PAI Clinical Scale Elevations

(e.g., 1967) within the context of his theory of depression; Physiological, to assess
the vegetative and somatic features (e.g., disturbances in sleep, appetite, and sex-
ual drive) that are commonly found in depressed patients; and Affective, to mea-
sure the unhappiness, SySDROH Ry:and apathy that are universally identified with
this population.

DEP-C: Cognitive
The Cognitive component of depression involves expectancies or beliefs
regarding ones inadequacy, powerlessness, or helplessness in dealing with the
demands of the environment. According to Beck (1967, 1976) and other cogni-
tively-oriented theorists such as Abramson, Seligman, and Teasdale (1978), the
root of depressive symptomatology lies in these beliefs. Individuals with this cog-
nitive style tend to globally attribute negative events in their lives to their own
incompetence or inadequacy, whereas any positive events are minimized or attrib-
uted to some external source (e.g., good luck, assistance from others, etc.). Beck
notes a number of other characteristics of the depressive cognitive style, including
(a) a tendency to think in dichotomies, with events viewed as extremes (good or
bad, black or white); (b) making self-referential assumptions, such as believing
everyone notices if one makes a small mistake; and (c) selective abstraction of neg-
ative events.
The DEP-C scale, by tapping such cognitions, reflects an important component
of self-esteem involving a sense of personal competence or self-efficacy. Individu-
als with DEP-C elevations are likely to report feeling worthless, hopeless, and as
having failed at most important life tasks. They are likely to be quite pessimistic
and to have very little self-confidence. Concentration problems and indecisiveness
are also likely to be present. Conversely, people with very low scores on DEP-C
(i.e., < 40T) report that their abilities have few limits; such a pattern could reflect
grandiosity or narcissism.

DEP-A: Affective
The affective component of depression refers to the experience of feeling dis-
tressed, unhappy, sad, blue, and down in the dumps. Elevations on DEP-A suggest
sadness, a loss of interest in normal activities, and a loss of sense of pleasure in
things that were previously enjoyed. This scale is probably one of the most direct
measures of overall life satisfaction on the PAI. Thus, as a relatively pure measure
of distress, DEP-A can be considered a positive prognostic indicator, as it reflects
a dissatisfaction with current circumstances, and the distress can serve as a moti-
vator for change.

ao
PAI Interpretive Guide

DEP-P: Physiological
The DEP-P subscale involves what are called the vegetative signs of depression:
sleep problems, appetite problems, lack of interest, and lack of drive. Of the three
DEP subscales, DEP-P demonstrates the largest correlation with the Hamilton Rat-
ing Scale (HAM-D) for Depression (r = 75). This is informative in that the HAM-D
is the most widely used measure of depressive symptomatology in psychopharma-
cological trials of antidepressant medication; these medications tend to be
particularly effective in treating vegetative signs of depression. Therefore, the
DEP-P scale may be useful in identifying target symptoms that may be amenable
to treatment with such medications.
Elevations on DEP-P suggest that the respondent has experienced a change in
level of physical functioning. Such people are likely to show a disturbance in sleep
pattern, a decrease in energy and level of sexual interest, and a loss of appetite
and/or weight loss. Motor slowing also may be present.

DEP Full Scale Interpretation

As the sum of the three subscales, the DEP full scale score indicates the broad
spectrum of diagnostic depressive symptomatology. Because all three components
are involved in the DSM definition of a disorder, the full scale can be useful in diag-
nostic decision-making. Average scores on DEP (i.e., < 60T) reflect a person with
few complaints about unhappiness or distress. Such individuals are typically seen
as stable, self-confident, active, and relaxed. Scores between 60T and 7OT are
indicative of a person who may be unhappy and who is sensitive, pessimistic, and
self-doubting. Scores above 70T suggest prominent unhappiness and dysphoria.
With scores in this range, the respondent is probably despondent much of the time
and withdrawing from activities he or she previously enjoyed. These individuals
may be seen as guilt-ridden, moody, and dissatisfied. With scores above 7OT, at
least one subscale is likely to be elevated, and these scores should be examined to
determine the typical modality in which the depression is manifest. As scores
become elevated above 80T, there is an increasing likelihood of a diagnosis of
Major Depressive Disorder.
DEP scores that are markedly elevated (i.e., > 95T) are likely to have elevations
on all three subscales, often reflecting a diagnosis of Major Depressive Disorder.
These individuals feel hopeless, discouraged, and useless. They are socially with-
drawn and feel misunderstood by others. Typically, there is little motivation to pur-
sue interests and little energy with which to do so. Suicidal ideation is not uncom-
mon with scores in this range, and particular attention should be given to SUI
elevations when DEP is markedly elevated.

40
 Interpreting PAI Clinical Scale Elevations

DEP Subscale Configurations

The following sections describe some of the implications of different combina-
tions of elevations on the three DEP subscales.

DEP-C high, DEP-A high, DEP-P high

With all three subscales elevated, the respondent is quite likely to meet the
diagnostic criteria for a major depressive episode. Plagued by thoughts of worth-
lessness and hopelessness, such individuals are preoccupied with feelings of sad-
ness, a loss of interest in normal activities, and a loss of sense of pleasure in things
that were previously enjoyed. They are likely to show a disturbance in sleep pat-
tern, a decrease in level of energy and sexual interest, and a loss of appetite and/or
weight loss. Psychomotor slowing or retardation might also be expected.

DEP-C high, DEP-A high, DEP-P average

This subscale pattern reflects an individual who is plagued by ruminative
thoughts of worthlessness and personal failure. Such people admit openly to feel-
ings of sadness, a loss of interest in normal activities, and a loss of the sense of
pleasure in things that were previously enjoyed, and they blame themselves for
feeling this way. However, the absence of physiological signs of depression suggests
that the complete spectrum of depressive symptomatology is not present, and the
person may not meet diagnostic criteria for a major depressive episode. This pat-
tern is common in more chronic dysphoric conditions, such as those seen with
dysthymic disorder or with certain personality disorders.

DEP-C high, DEP-A average, DEP-P high

An individual with this unusual subscale pattern reports markedly low self-
esteem and numerous physiological signs of depression, yet he or she is not admit-
ting to feeling unhappy or distressed. This suggests that the individual might not
recognize the aforementioned symptoms as signs of dysphoria and stress or might
be repressing the experience of unhappiness to some extent. Alternatively, the per-
son may not be willing to admit to personal unhappiness, viewing it as a sign of
weakness. Regardless, it is likely that the person is unhappy at some level and will
be vulnerable to future episodes of depression during times of stress.

DEP-C average, DEP-A high, DEP-P high

Although such people do not appear to feel hopeless and their self-esteem is
largely intact, they are manifesting affective and physiological signs of depression.
Such a pattern would appear to contraindicate a more cognitively-based interven-
tion for the depression and, instead, may underscore the importance of managing

a|
PAI Interpretive Guide

the physical symptoms, perhaps with antidepressant medication. The relatively

lower score on DEP-C suggests that external circumstances, rather than internal
shortcomings, may be blamed for the person’s current unhappiness.

Mania (MAN )
By definition, mania is a disorder with a fluctuating presentation of sympto-
matology, and this fluctuation presents a measurement challenge for traditional
assessment methods. Within a particular manic episode, symptoms can vary
widely; for example, mood can be alternatively elevated, irritable, or depressed
within a brief time span. Over the past few decades, an empirical literature has
emerged that documents the symptomatic complexity of patients presenting dur-
ing a manic episode. Goodwin and Jamison (1990), in a comprehensive descrip-
tion of the manic-depressive syndrome, reviewed the results of a number of
these studies of symptomatology in an attempt to identify the most salient diag-
nostic features of mania. They divided symptoms into four broad areas: (a) mood,
(b) cognitive, (c) activity and behavior, and (d) psychotic symptoms. By collapsing
results across several studies, Goodwin and Jamison were able to calculate a
weighted mean representing the diagnostic sensitivity of different signs and symp-
toms within each of the four areas. With respect to mood symptoms, the most
commonly observed were irritability (80% of patients), followed by depression
(72%), and euphoria (71%); among cognitive symptoms, grandiosity (78%), rac-
ing thoughts (71%), and poor concentration (71%) were most common; and
among behavioral symptoms, hyperactivity (87%), typically involving pressured
speech (98%), and decreased sleep (81%) were often observed. However, psy-
chotic symptoms such as delusions (48%) or hallucinations (15%) were much less
frequently observed.
The MAN scale of the PAI was designed to assess prototypic signs of a manic
episode. Consistent with the findings of Goodwin and Jamison (1990), disruptions
in mood, cognition, and behavior were each assessed via different subscales;
because of the low sensitivity of psychotic symptomatology and because such
symptoms are often of limited utility in making a differential diagnosis from other
psychotic disorders (Carlson & Goodwin, 1973), assessment of psychotic features
received relatively little weight in the final scale. Thus, three MAN subscales were
designed: Activity Level, with items addressing pressured speech, decreased sleep,
increased motor activity, and extravagance; Grandiosity, including inflated self-
esteem, overvalued ideas, and interpersonal overconfidence:; and Irritability, par-
ticularly involving impatience and demandingness with others.

42
 Interpreting PAI Clinical Scale Elevations

MAN-A: Activity Level

The primary feature of manic behavior is that it is elevated: individuals in a
manic episode engage in more behaviors than most people. The activity level is
heightened with respect to ideational as well as behavioral activity, so ideas flow
as
rapidly as behaviors (i.e., flight of ideas). However, this increase in quantity of
behavior is accompanied by a decrease in quality; both the ideation and the overt
activity become pressured and disorganized. Thus, high scorers on the scale are
not merely involved in many activities; instead, they are overinvolved and ineffec-
tive at managing all of their commitments.
The MAN-A subscale has one of the “softest floors” of the PAI clinical scales,
meaning that it is possible to obtain very low scores. Scores in this range (i.e.,
< 30T) represent very low activity levels and marked apathy and indifference that
often characterize severely depressed individuals. Scores in the moderate range
(i.e., 55T to 65T) suggest an activity level somewhat higher than normal; in the
upper end of this range, the person may be overcommitted to a wide variety of
activities, but not necessarily in a disorganized fashion. Scores between 65T and
75T represent an activity level that is perceptibly high to most observers. Such
people tend to be involved in a wide variety of activities in a somewhat disorga-
nized manner and to experience accelerated thought processes. As scores exceed
75T, this acceleration renders the person confused and difficult to understand;
scores in this range are unusual, as such people often have difficulty focusing their
attention for the time required to complete the PAI.

MAN-G: Grandiosity
The grandiosity component of mania involves an overevaluated self-image, an
overestimation of one’s talents and capabilities. Hence, MAN-G items inquire about
the person's self-evaluation of many talents and abilities. Grandiose individuals
tend to believe they are good at almost anything, and, thus, they obtain elevated
scores. In milder forms, this may merely reflect an optimism and an unwillingness
to be hampered by one’s limitations. In more extreme forms, this represents an
incapacity to recognize one’s limitations and an inability to think clearly about
one’s own capabilities.
The MAN-G subscale, like MAN-A, is interpretively useful at the lower end.
Because the scale has a major component of self-evaluation, it can be useful in
identifying persons with low self-esteem who are not necessarily depressed. Very
low scores on MAN-G can render an individual vulnerable to depression, as such
people tend to feel rather inadequate and to be unwilling to accept or acknowl-
edge their own positive aspects. Conversely, when DEP is elevated and MAN-G is

43
PAI Interpretive Guide

not suppressed, this may indicate that blame for the current circumstances is being
externalized. Thus, for example, a paranoid individual may be pessimistic about
his or her ability to deal with external forces, yet the self-esteem will remain intact.
So, although they may have an elevated DEP-C, suggesting that they doubt their
ability to succeed against external forces, their self-esteem is unimpaired because
they simply project the blame outward. Thus, even more than DEP-C, the MAN-G
score may reflect the extent to which a low self-concept has been internalized.
Scores on MAN-G that are in the moderately elevated range (i.e., 60T to 70T)
represent an optimistic and, perhaps, driven type of individual. Content of
thought is likely to be marked by an element of expansiveness and self-confidence,
with a focus on strategies for success or achievement. Toward the upper end of this
range, the possibility of inflated self-esteem increases. As scores exceed 7OT, the
likelihood of grandiosity must be considered, as scores in this range are unusual
in clinical settings. Such elements may range from beliefs of having exceptionally
high levels of common skills to beliefs that border on delusional in terms of hay-
ing special and unique talents that will lead to fame and fortune. Others may view
such people as self-centered and narcissistic.

MAN-I: Irritability
Although elevated mood is one of the more striking affective features of mania,
it is actually not as characteristic of mania as might be expected. More typical of
manic affect is volatility; the mood can change rather abruptly, particularly in
response to frustration. Thus, MAN-I items tap a frustration-responsive irritability
that is typical of manic patients. There tend to be two aspects to these items, one
involving a certain degree of ambition and the other involving low frustration tol-
erance. It is this combination of features that makes the scale reasonably specific,
rather than a more general marker of trait hostility, a characteristic that may be
more directly addressed by some of the PAR subscales.
Low scores on MAN-I (.e., S 40T) reflect an individual who portrays himself
or herself as very patient and rather immune to frustrations. Milder elevations (i.e.,
60T to 70T) suggest a person who is impatient, and individuals with scores in the
upper end of this range may be seen by others as demanding. Such people may
have difficulty with others who do not cooperate with them or who do not keep up
with their plans and schedule of activities. As scores exceed 70T, relationships with
others are probably under stress due to the demanding presentation of the respon-
dent. Such people are easily frustrated by lack of ability or cooperation in other
people, and these other people will tend to be blamed for the respondents failures

44
 Interpreting PAI Clinical Scale Elevations

and to be accused of attempting to thwart the respondents possibly unrealistic

plans for success and achievement. With scores above 80T, the person is quite
volatile in response to frustration, and his or her judgment in such situations may
be poor. The quality of mood state in such people can change very rapidly, and
they are prone to lash out at people they view as the source of their frustrations.

MAN Full Scale Interpretation

Elevations on the full scale of MAN tend to be rarer in clinical settings than any
of the other clinical scales of the PAI. Indeed, the average scores for clinical and
community respondents are nearly identical, which is certainly not the case with
any other PAI clinical scale. As such, the “psychological threshold” for identifying
MAN scores as problematic should be lowered in most clinical settings.
Average scores on MAN (i.e., < 55T) reflect a person with few features of mania
or hypomania. Although depressed individuals are rarely grandiose and do not
have heightened activity levels, they are often quite irritable; hence, depression
will not invariably be associated with very low MAN scores. Scores between 55T
and 65T are indicative of a person who may be seen as active, outgoing, ambitious,
and self-confident; however, toward the upper end of this range such individuals
also may be rather impatient, hostile and quick-tempered. Scores in the 65T to
75T range are associated with increasing restlessness, impulsivity, and high energy
levels. Other people are likely to perceive such individuals as unsympathetic and
hot-headed.
MAN scores that are markedly elevated (i.e., > 75T) are typically associated
with disorders such as mania, hypomania, or cyclothymia. These individuals take
on more than they can handle and react in a hostile manner to suggestions that
they reduce their activities. They are typically quite impulsive and have little abil-
ity to delay gratification; their lack of judgment in such situations is likely to lead
to significant impairment in role functioning. They may be experience flights of
ideas, and their grandiosity may be delusional in proportion. Their interactions
with others are likely to be problematic, as their self-importance, hostility, and nar-
cissism impede their ability to be empathic in relationships.

MAN Subscale Configurations

The following sections describe some of the implications of different combina-
tions of evaluations on the three MAN subscales. Because MAN subscale elevations
tend to be unusual in clinical settings, subscale scores greater than 65T should be
considered “high” in interpreting configurations using the following paragraphs.

45
PAI Interpretive Guide

MAN-A high, MAN-G high, MAN-I high

This pattern of subscale scores suggests a clinical picture with numerous ele-
ments of mania. Such people will have an activity level that is perceptibly high to
most observers. They are probably involved in these activities in an overcommit-
ted and disorganized manner, and they may experience their thought processes as
being accelerated, although they may not recognize the extent of their disorgani-
zation. In part, they are active in many areas because they feel that they have spe-
cial talents in many areas; content of thought is likely to be marked by overvalued
ideas, inflated self-esteem, or grandiosity. They may believe that they have excep-
tionally high levels of common skills, and they possibly harbor delusional beliefs
of having special and unique talents that will lead to fame and fortune. Relation-
ships with others are probably under stress, due to a frustration with the inability
or unwillingness of these other people to keep up with overvalued plans and
possibly unrealistic ideas. At its extreme, this irritability may result in accusations
that significant others are attempting to thwart these plans for success and achieve-
ment, particularly when there is an accompanying elevation on PAR.

MAN-A high, MAN-G high, MAN-I average

This pattern of responses represents a very active person who is probably
involved in his or her activities in an enthusiastic, overcommitted, and disorga-
nized manner. The significance or importance of these activities may be overval-
ued, as may the person's self-perception of his or her talents and abilities. The lack
of any elevation of MAN-I is a favorable sign (i.e., it suggests greater perseverance
in these behaviors than might be found otherwise), and this increases the possi-
bility that some of the individual's energy and enthusiasm can be translated into
effective action.

MAN-A high, MAN-G average, MAN-I high

This pattern suggests that the clinical picture is characterized by heightened
energy levels and irritability. This combination suggests a great emphasis on action
and activity, perhaps at the expense of relationships and feelings. Other people
probably view the respondent as driven, impatient, and demanding, and the
respondent is easily frustrated by any inability or unwillingness of these other peo-
ple to keep up with the agenda and accompanying (possibly unrealistic) expecta-
tions. At its extreme, this irritability may result in resentment that significant oth-
ers are attempting to thwart the respondent's plans for success and achievement.

MAN-A average, MAN-G high, MAN-I high

This pattern of scores suggests an individual with inflated self-esteem and
overvalued ideas, who has little tolerance for others who fail to recognize his or

46
 Interpreting PAI Clinical Scale Elevations

her special talents and unique abilities. Others are likely to view the respondent as
demanding, impatient, and arrogant. The self-esteem may be particularly vulnera-
ble to insult (particularly if BOR-I is elevated), and, when it is threatened, such
individuals may lash out in frustration at those around them. Relationships with
others are probably strained, as such people will repeatedly clash with anyone who
differs from them or their agenda. However, they probably do not view themselves
as hostile, but rather as acting in a manner merited by the strength and importance
of their ideas and convictions.

Paranoia (PAR)
As is the case with anxiety in milder conditions, symptoms of paranoia are
found in a variety of diverse and more severe psychopathologic conditions. The
manifestations can range from characterological suspiciousness (e.g., that found in
Paranoid Personality Disorder) to the frank persecutory delusions that character-
ize paranoid psychosis. However, paranoid symptoms are not specific to these syn-
dromes; these beliefs are often encountered in schizophrenia, mania, other per-
sonality disorders such as antisocial and borderline personality, and certain organic
conditions. Regardless of the nature of the primary diagnosis, paranoid symptoms
present a difficult assessment challenge because the respondent is, by definition,
defensive and suspicious of diagnostic and treatment efforts. In identifying the rel-
evant components of the paranoia construct for the PAI, a decision was made to
place an emphasis on the phenomenology of the disorder, rather than on the more
overt symptomatology, in an effort to reduce the impact of defensiveness on scale
performance.
The PAR scale was designed to identify the personological elements of para-
noia, as well as the more symptomatic elements. One of the three PAR subscales,
Persecution (PAR-P), includes items consistent with the typical delusional beliefs
associated with severe paranoia. The items for the remaining two subscales were
written to capture the experience of the paranoid in a manner that might be less
affected by the typically guarded posture of the paranoid respondent. The Hyper-
vigilance (PAR-H) subscale indicates an attitude of preparedness, sensitivity, and
wariness in interactions with others. The Resentment (PAR-R) subscale involves
somewhat bitter and envious feelings toward others, along with a sense of being
treated unfairly by others.

PAR-H: Hypervigilance
The paranoid individual carries the predisposition to distrust people that he or
she does not know well. As a result, such individuals tend to be vigilant and

ay
PAI Interpretive Guide

guarded in their interactions with others, looking for warning signs that the per-
son with whom they are dealing is not completely trustworthy. This tendency is
more of an interpersonal set, a way of relating to others, than it is a specific belief;
therefore, elevations should not be interpreted as indicative of a delusional system.
Rather, there is a wariness in interactions with others and a reluctance to let one’s
guard down in relationships.
PAR-H has a reasonably soft floor and very low scores are possible. When
scores below 40T are obtained, this suggests a person who reports being exceed-
ingly trusting and open in relationships. If this self-report is accurate, such people
are vulnerable to interpersonal exploitation, particularly if DOM is low. However,
such scores may also be obtained by individuals who are motivated to appear as
trusting. Moderate elevations (i.e., 60T to 70T) suggest individuals who are prag-
matic and skeptical in relationships with others; such people may be difficult to
know well and may keep casual acquaintances at arm's length. Scores above 70T
indicate a person who spends a great deal of time monitoring the environment for
evidence that others are not trustworthy and may be trying to harm or discredit
the individual in some way. Others will view such people as hypersensitive and
easily insulted in their interactions. Such people will question and mistrust the
motives of those around them as a matter of course, despite the nature or history
of the relationships. As a result, working relationships with others are likely to be
strained and may require an unusual degree of support and assistance in order to
succeed.

PAR-P: Persecution
The items on the Persecution subscale directly address beliefs that others are
attempting to obstruct or impede the respondents efforts. These beliefs can range
from mild feelings of jealousy to delusional beliefs of conspiracy and intrigue. Of
the three PAR subscales, PAR-P is most closely tied to Axis I manifestations of delu-
sional disorders involving paranoia.
Because item content on PAR-P is unusual, raw scores tend to be low in the
general population and the standard deviation tends to be small. Hence, the scale
can elevate rapidly even if relatively few items are answered in the positive direc-
tion. Elevated scores suggest an individual who is quick to feel that he or she is
being treated inequitably and easily believes that there is a concerted effort among
others to undermine his or her best interests. Working and social relationships are
likely to be very strained, despite any efforts by others to demonstrate support and
assistance. As scores increase above 85T, the possibility of delusional beliefs
should be investigated, particularly if SCZ-P is also elevated.

48
 Interpreting PAI Clinical Scale Elevations

PAR-R: Resentment
The third PAR subscale captures the hostility and bitterness of the paranoid
character, the tendency to approach life with a “chip on the shoulder.” The
obstructions provided by others (reflected in the scores on the other subscales) are
a source of lingering resentment for such individuals. These people feel that they
have not treated fairly in life, and they nurse grudges against all who have trans-
gressed against them in the past. Blame for any failure is projected outwards, and
forgiveness from the respondent is not likely. Indeed, “getting even” with the
objects of this resentment may be a major preoccupation for such people.
Scores on PAR-R that are moderately elevated (i.e., 60T to 70T) suggest a sen-
sitive person who is easily insulted or slighted and responds by holding grudges
toward the offending party. As scores elevate above 7OT, the respondents are
increasingly inclined to attribute their misfortunes to the neglect of others and to
discredit the successes of others as being the result of luck or favoritism. They are
likely to be envious of others and disinclined to assist others in achieving their
goals and successes. As scores exceed 80T, the person may dwell on past slights
by others and may be preoccupied with evening the score. Examination of scores
on DOM and AGG may suggest whether this hostility is likely to be expressed
directly or in more passive-aggressive form.

PAR Full Scale Elevations

The PAR scale measures the characteristic phenomenology of the paranoid
individual with respect to both symptomatology and personality elements. The
item content addresses a vigilance in monitoring the environment for potential
harm, a tendency to be resentful and to hold grudges, and a readiness to spot
inequities in the way the respondent has been treated by others. At the full scale
level, PAR represents a direct measure of interpersonal mistrust and hostility.
Average scores on PAR (i.e., < 60T) reflect a person who reports being open
and forgiving in relationships with others. Scores between 60T and 70T are indica-
tive of a person who may be seen as sensitive, tough-minded, and skeptical.
Toward the upper end of this range, individuals may also be rather wary and cau-
tious in their interpersonal relationships. With scores above 7OT, the person is
likely to be overtly suspicious and hostile. Such a person tends to be distrustful of
close interpersonal relationships and probably has few close friends.
PAR scores that are markedly elevated (i.e., > 84T) are typically associated
with paranoia of potentially delusional proportions. These individuals are bitter
and resentful of the the way they have been treated by others, and they expect

49
PAI Interpretive Guide

that others will attempt to exploit them. Any close relationships that may exist are
probably troubled by jealousy and accusations. Ideas of reference and delusions of
persecution or grandiosity are not uncommon when scores are in this range.

PAR Subscale Configurations

The following sections describe some of the implications of different combina-
tions of elevations on the three PAR subscales.

PAR-H high, PAR-P high, PAR-R high

This pattern suggests a hypersensitive and hypervigilant individual who often
questions and mistrusts the motives of others. Such people are extremely touchy
in interactions with others and tend to harbor strong feelings of resentment as a
result of perceived slights and insults. When circumstances fail to go their way,
they are quick to feel that they are being treated inequitably and often holds
grudges against others, even if the perceived affront is unintentional. Consistent
with the constellation of hypervigilance, suspiciousness, and resentrnent, such
people are seen by others as being quite hostile. Working relationships with others
are likely to be very strained, despite any efforts by others to demonstrate support,
reassurance, and assistance.

PAR-H high, PAR-P high, PAR-R average

This type of individual feels that he or she has been taken advantage of in the
past and is on guard to prevent similar circumstances from happening again. Such
individuals approach relationships in a hypervigilant fashion and easily mistrust
the motives of others. They are very sensitive to any perceived affronts and will
withdrawal quickly from individuals who are perceived as anything less than
totally supportive. Casual relationships are likely to be quite distant and strained,
and even efforts by others to demonstrate support and assistance may be viewed
with skepticism by the respondent.

PAR-H high, PAR-P average, PAR-R high

This patterns suggests a characterologically suspicious individual who is pre-
disposed to question and mistrust the motives of others. Such people are vigilant
to any signs that they are being treated unfairly, and they will harbor strong and
lingering feelings of resentment following any perceived slights and insults.
Although they may not view themselves as unduly suspicious, others are likely to
see such people as hostile and unforgiving. Establishing close relationships with
such people tends to be quite difficult because of the lack of trust and the suspi-
cion of any efforts to render assistance.

50
 Interpreting PAI Clinical Scale Elevations

PAR-H average, PAR-P high, PAR-R high

This pattern suggests a person who feels that life has treated him or her
unfairly. Such people are bitter about their perceived mistreatment, and they feel
they have been victimized in some manner through the neglect or active interfer-
ence of others. They tend to be envious of others and to denigrate their accom-
plishments, and they are not likely to support or cooperate with the efforts of oth-
ers. They are very slow to forgive transgressions and may ruminate about past
slights and insults at the hands of others. Such people are prone to attribute the
causes for any untoward circumstances externally, and they often feel as if they
have very little control over the outcomes in their lives, seeing themselves as the
pawn of various malevolent forces. They place a very high premium on loyalty in
the people around them, but their high expectations in this regard are often impos-
sible to meet.

Schizophrenia (SCZ)
Schizophrenia is one of the most heterogeneous of all clinical syndromes, and
this heterogeneity poses a number of problems for assessment. Historically, there
have been many schemes for subtyping schizophrenia, with the number of sub-
types ranging from the three originally described by Kraepelin (i.e., paranoid, cata-
tonic, hebephrenic) to the dozens of subtypes described by Leonhard (e.g., Ban,
1982). The distinction between “positive” and “negative” symptoms in schizo-
phrenia has received considerable research support in recent years. Positive symp-
toms involve the presence of features that are normally not present in individuals;
they include phenomena such as hallucinations, delusions, and bizarre behavior.
Negative symptoms represent the absence of features that normally are present in
individuals, such as social behavior and affective responsiveness (Andraesen,
1985). The clinical import of the distinction can be found in a wide variety of
areas; for example, patients with predominantly negative symptoms often show lit-
tle response to neuroleptic medication and have poorer prognoses (Angrist,
Rotrosen, & Gershon, 1980).
However, thought disorder is an important diagnostic feature of schizophrenia
that does not fit neatly into the positive-negative distinction. Some features of
thought disorder (e.g., tangential speech) are considered positive symptoms,
whereas others (e.g., thought blocking, attentional problems) are sometimes char-
acterized as negative symptoms. Confirmatory factor analyses have demonstrated
that features of thought disorder tend not to group well with either symptom
group (Lenzenweger, Dworkin, & Wethington, 1980), and some analyses have

at
PAI Interpretive Guide

suggested that thought disorder should be considered a third, relatively indepen-

dent pattern of impairment in schizophrenia.
The SCZ scale of the PAI was designed to assess these three aspects of schizo-
phrenia. Positive symptoms, negative symptoms, and thought disorder were each
assessed via different subscales. The Psychotic Experiences subscale emphasizes
the positive symptoms of schizophrenia, such as delusions and hallucinations, that
are central to the DSM definition of the disorder. The Social Detachment subscale
focuses on the most characteristic negative symptom of schizophrenia, social with-
drawal and poor rapport. Finally, the Thought Disorder subscale includes items
assessing experiences such as thought blocking, confusion, distractibility, and con-
centration problems.

SCZ-P: Psychotic Experiences

Positive symptoms of schizophrenia involve delusions and hallucinations, as
well as characteristic bizarre thought content. The positive symptoms tend to have
a rather distinct course, with episodic exacerbations and often complete remis-
sions, and persons with predominantly positive symptomatology do not tend to
demonstrate intellectual impairments. These symptoms also tend to respond favor-
ably to antipsychotic medications.
The SCZ-P items tap various positive symptoms of schizophrenia that vary in
severity from unusual perceptions and magical thinking to the characteristic first-
rank psychotic symptoms of schizophrenia. In keeping with efforts to maintain
discriminant validity, the features are designed to be relatively specific to schizo-
phrenia rather than more broadly defined, nonspecific symptoms that might be
found in other syndromes (e.g., delusions of grandeur or nihilistic delusions).
Scores that are moderately elevated (i.e., 60T to 70T) suggest that the respondent
may entertain some ideas that others tend to find unconventional or unusual:
toward the upper end of this range, the person may strike others as peculiar and
eccentric. Scores above 70T indicate the experience of unusual perceptual or sen-
sory events and/or unusual ideas that may involve delusional beliefs. Scores
exceeding 85T are often associated with an active psychotic episode, with poor
judgment and breakdown in reality testing as hallmark features; full blown hallu-
cinations or delusions are probable.

SCZ-S: Social Detachment

The negative symptoms of schizophrenia involve behavioral deficits such as
poor interpersonal rapport, flattening of affect, and poverty of communication.
Such individuals are apathetically indifferent to others, usually speaking to others
only when necessary and avoiding interpersonal contact whenever possible. In

52
 Interpreting PAI Clinical Scale Elevations

schizophrenia, the course of these negative symptoms tends to be endurin

g, as
opposed to episodic, and they are less responsive than positive symptoms to
phar-
macologic interventions. This pattern of behaviors is also consistent with the fea-
tures of schizoid personality, which may simply be an alternative name for the
same phenomenon.
The SCZ-S items focus upon the features of social disinterest and lack of affec-
tive responsivity. Moderate scores (i.e., 60T to 70T) suggest a quiet, impassive indi-
vidual who exhibits little interest in the lives of other people. Toward the upper
end of this range, scores may indicate a lack of ability to interpret the normal
nuances of interpersonal behavior that provide the meaning to personal relation-
ships. Scores above 7OT reflect a person who neither desires nor enjoys close rela-
tionships; social isolation and detachment may serve to decrease the sense of dis-
comfort fostered by interpersonal contact. Their lack of interest in others is
mirrored in a lack of self-interest; they are generally indifferent to how others
view them and are disinterested in introspection. They are made particularly
uncomfortable by strong emotions, which they themselves tend not to experience
and which they do not understand in others.

SCZ-T: Thought Disorder

Schizophrenia is characterized by disruptions in thought process that do not
seem to covary with either positive or negative symptoms. At the extreme, a
thought disorder can render the patient incoherent and unable to string together
an intelligible sentence. In its milder forms, difficulties in concentration, decision-
making, and memory will occur. It should be recognized that these milder features
tend to be nonspecific, associated with severe affective disorders in particular.
Thus, SCZ-T elevations are commonly observed in severe major depression, with-
out accompanying elevations on SCZ-P.
The SCZ-T items sample across the range of clarity and freedom from confu-
sion in thought processes. Moderate elevations (i.e., 60T to 701) suggest problems
in concentration and decision-making; such scores would not be unexpected
among depressed or anxious individuals. However, toward the upper end of this
range, there will be increasing likelihood of confusion and perplexity in addition
to the more benign cognitive inefficiencies. Scores above 7OT reflect a loosening of
associations and increased difficulties in self-expression and communication.
However, in the absence of a clinical elevation of the full SCZ scale, this finding
can reflect various causes other than schizophrenic disorder. Severe depression or
mania, the sequelae of brain injury or disease, the effects of medication, and the
consequences of drug or alcohol abuse should all be explored as potential causes
of elevations on this subscale.

53
PAI Interpretive Guide

SCZ Full Scale Elevations

The SCZ scale was designed to measure a number of the different facets of
schizophrenia; this multifaceted approach is necessary, because the disorder is one
of the most heterogeneous of all clinical groups. Hence, elevations on the full scale
could result from a number of causes: unusual beliefs and perceptions; poor social
competence and social anhedonia; or inefficiency and disturbances in attention,
concentration, and associational processes. Average scores on SCZ (i.e., < 60T)
reflect a person who reports being effective in social relationships and has no trou-
ble with attention or concentration problems. Scores between 60T and 7OT are
indicative of a person who may be seen as withdrawn, aloof, and unconventional.
Toward the upper end of this range, individuals may be quite cautious and hostile
in their few interpersonal relationships. With scores above 70T, the person is likely
to be isolated and to feel misunderstood and alienated from others. Some difficul-
ties in thinking, concentration, and decision-making are probable with scores in
this range. Specific subscale elevations may reveal the presence of unusual per-
ceptions or beliefs that may be psychotic in nature.
SCZ scores that are markedly elevated (i.e., > 9OT) are typically associated with
an active schizophrenic episode. These individuals are confused, withdrawn, sus-
picious, and tend to have poor judgment and reality testing. Prominent psychotic
symptomatology is likely with scores in this range, and specific elevations on other
scales may be helpful in identifying the precise nature of such symptoms. For
example, concomitant elevations on PAR may indicate the presence of delusions of
persecution. With increasing T-score elevations, delusions of thought broadcast-
ing, thought insertion, thought withdrawal, and thought control become more
likely. These individuals may require referral to evaluate the need for psychotropic
medications.

SCZ Subscale Configurations

The following sections describe some of the implications of different combina-
tions of elevations on the three SCZ subscales.

SCZ-P high, SCZ-S high, SCZ-T high

This pattern indicates prominent features from across the schizophrenic spec-
trum. It is likely that the respondent experiences unusual perceptual events or full-
blown hallucinations as well as unusual ideas that may include magical thinking
or delusional beliefs. However, because such people are quiet and avoid interac-
tions with others, this unusual thought content may not be readily apparent. Such

at
 Interpreting PAI Clinical Scale Elevations

people are likely to be socially isolated, with few interpersonal relationships

that
could be described as being close and warm. In addition to having limited social
skills, the person’s thought processes are likely to be marked by confusion, dis-
tractibility, and difficulties in concentration; such individuals may experience their
thoughts as blocked, withdrawn, or somehow influenced by others.

SCZ-P high, SCZ-S high, SCZ-T average

This pattern represents a person who reports unusual thought content with no
disruptions in thought process. The thought content may involve unusual percep-
tual or sensory events (perhaps including full-blown hallucinations) and/or as
unusual ideas that may include magical thinking or delusional beliefs. If PAR-P is
markedly elevated, these ideas may involve persecutory beliefs that may be part of
a well integrated delusional system. Such a finding would also explain the person's
presentation as being a socially isolated individual with few, if any, close relation-
ships.

SCZ-P high, SCZ-S average, SCZ-T high

This pattern suggests an individual presenting with acute psychotic sympto-
matology, involving unusual perceptual or sensory events (perhaps including full-
blown hallucinations) as well as unusual ideas that may include magical thinking
or delusional beliefs. The person’s thought processes are likely to be marked by
confusion, distractibility, and difficulties in concentration, and he or she may expe-
rience thoughts as being blocked, withdrawn, or somehow influenced by others.
The relative absence of negative symptoms may be a favorable prognostic sign for
eventual remission of these symptoms.

SCZ-P average, SCZ-S high, SCZ-T high

This pattern suggests a socially isolated individual who has few interpersonal
relationships that could be described as being close and warm. Such people tend
to have limited social skills, with particular difficulty in interpreting the normal
nuances of interpersonal behavior that provide the meaning to personal relation-
ships. Generally apathetic and disinterested in other people and their emotional
state, such individuals may withdraw from social interaction to decrease the sense
of confusion fostered by interpersonal contact. Thought processes are likely to be
inefficient and marked by distractibility and concentration problems. Such indi-
viduals are likely to have difficulty communicating effectively, and others who suc-
ceed in getting to know them (probably a difficult task) may see them as strange
and peculiar.

55
PAI Interpretive Guide

Borderline Features (BOR)

The BOR scale assesses a number of elements related to severe personality dis-
order; although all of these elements are part of the borderline syndrome, individ-
ually they are also common to numerous other disorders. This scale is the only PAI
scale that has four subscales, largely due to the complexity of the construct as it
has been represented in the literature. Part of the reason for this complexity is that
this is inherently a more nebulous construct than some that have been recognized
for a much longer time (e.g., depression or schizophrenia). The borderline concept
has always been thought of as reflecting a “boundary,” presumably representing
some border, but the nature of the border has never been exactly clear. Initially,
borderline personality represented the border of analyzability (i.e., patients who
were marginally able to be treated with psychoanalysis). Over time, this came to
be synonymous with the boundary between neurosis and psychosis, with a neu-
rotic level of adaptation presumably reflecting problems in the Oedipal stage char-
acterized primarily by difficulties with anxiety, and with psychosis reflecting more
primitive issues involving breaks with reality. In this framework, borderline indi-
viduals fell somewhere in the middle. It was thought that much of the time the
borderline individual superficially would appear to be at a neurotic level of adap-
tation, but that, under stress, and particularly in more unstructured situations,
such individuals would deteriorate and appear psychotic.
The actual incorporation of borderline personality into the diagnostic literature
occurred in the DSM-III (1980). The formulation of the current construct grew out
of work conducted by Robert Spitzer, the chair of the DSM-III Task Force, who
identified two types of individuals who were being identified as borderline: One
type who appeared to lie at the boundary of psychosis or the boundary of schizo-
phrenia and another type who were affectively and behaviorally unstable and
erratic. The “unstable” variant was eventually renamed “borderline”; the other type
was named Schizotypal Personality Disorder and was considered to represent a
schizophrenia spectrum disorder. The resulting borderline criteria, reflecting an
erratic and inconsistent group of individuals, were quite factorially complex, but
subsumed personality features useful in understanding a variety of different and
severe personality disorders.
Over the years, a number of investigators have examined the borderline con-
struct using factor-analytic or cluster-analytic studies (e.g., Grinker, Werble, &
Drye,1968; Hurt & Clarkin, 1990; Morey, 1989). These studies have provided con-
vergence in identifying the major facets of the borderline construct, and each facet
represents a theoretically important etiological mechanism. The four BOR sub-
scales of the PAI were designed to reflect these facets.

56
 Interpreting PAI Clinical Scale Elevations

BOR-A: Affective Instability

Individuals with borderline personality present with emotions that fluctuate
impressively, leading some theorists to propose that the disorder may represent a
variant of bipolar affective disorder (e.g., Akiskal, Yerevanian, & Davis, 1985).
However, the mood changes in borderline patients tend to differ in many ways
from the mood changes in bipolar patients. First, the mood changes in borderline
individuals are not regular. Instead, they tend to be very sudden, without any
rhythmicity. Also, borderline patients rarely, if ever, return to a period of normal
affect; there are few days where there is not some dramatic affective change in such
individuals. Furthermore, studies of family histories yield smaller estimates of relat-
edness between the disorders than would be expected if borderline was a bipolar
spectrum disorder.
Nonetheless, affective instability in the form of sudden emotional change is
one of the hallmark characteristics of borderline personality. These affects are not
a polarity between happiness and sadness, however. Rather, for borderline patients
affective instability involves a propensity to rapidly become anxious, angry,
depressed, or irritable. The BOR-A subscale reflects this rapidity of mood shift. Ele-
vations could, for example, represent an individual with a bad temper (which can
be confirmed by an examination of the AGG-A subscale), or it might indicate a per-
son who becomes anxious easily (a conclusion that might be supported from
inspecting the ANX-A or ARD-P subscales). The unique contribution of the BOR-A
subscale is in ascertaining the suddenness of the affective change.
Thus, high scorers on BOR-A are highly responsive emotionally, typically man-
ifesting rapid and extreme mood swings, rather than the more cyclic mood
changes seen in affective disorders. In the highest ranges (i.e., roughly > 80T) all
affects are likely to be involved, including episodes of poorly controlled anger. In
the range from 70T to 80T, a propensity to experience a particular negative affect
may be responsible, and investigation of other scales may determine whether anx-
iety (ANX-A or ARD-P), depression (DEP-A), or anger (AGG-A) is the typical
response. On the other hand, unusually low scores (i.e., < 40T) reflect individuals
who describe themselves as fairly unresponsive emotionally and who may appear
to others as affectively constricted.
BOR-I: Identity Problems
Theoretically, the notion of issues surrounding identity are central to Kern-
berg’s (1975) view of borderline personality. Kernberg describes this facet as “iden-
tity diffusion,” meaning that borderline patients have a difficult time maintaining
a constant representation of who they are, where they are headed in life, and what
they value. As a result of this diffuse sense of self, such individuals tend to rely on

57
PAI Interpretive Guide

others to help them formulate an identity, thus defining themselves primarily in

relationship to other people. Theoretically, this involves a developmental failure to
establish an autonomous identity independent of the primary caregivers, leading
to similar difficulties in adulthood. In a sense, this involves being dependent upon
others, as illustrated in DSM criteria such as “fears of abandonment.” Although
there is certainly substantial diagnostic overlap between borderline and dependent
personality disorder (Morey, 1988), there is a qualitative difference in the nature of
these behaviors. Borderline individuals do not really want the assistance of others
to make sure that they perform their jobs effectively or make good decisions;
rather, they have a profound need for others to help them define for themselves
who they are. In the absence of these important others, borderline individuals may
initiate very desperate and frantic efforts to try to reestablish this needed contact,
not out of fear that they will be unable to do their jobs effectively, but because they
are afraid they will cease to exist.
Because borderline individuals may desperately cling to the people who are
most important and central to defining them, it is at times assumed that this rep-
resents a form of “over-idealization” of others; indeed, this notion is incorporated
into the DSM criteria. However, to some extent this description misinterprets the
behavior of the borderline individual, confusing a profound need for others with
an idealization of those others. Many times, this need will not necessarily be man-
ifest in idealization; in fact, borderline individuals are likely to have constant con-
flicts with the people closest to them. Nonetheless, even through this conflict with
important others, the borderline individual can continue to maintain an identity
as an extension of these others, as a spouse, a friend, an offspring, or even as an
enemy. It is the urgent necessity of these relationships, rather than their idealized
quality, that is characteristic of the borderline individual.
One implication of the problems in identity and sense of self reflected by BOR-I
is that the self-concept is unstable and inconsistent. At any particular moment, a
borderline individual may have an overriding life ambition that he or she can
describe with great earnestness, but by the next week, the ambitions are likely to
be totally different. No matter how deep an attachment to some particular course
of action may appear, within a short period of time a design of equal intensity may
emerge in an entirely different direction. Individuals with elevations on BOR-I are
likely to be prone to these sudden shifts in ambitions and goals.
In sum, scores above 7OT represent uncertainty about major life issues and dif-
ficulties in developing and maintaining a sense of purpose. Such uncertainty is
more common in younger adults, and BOR-I is correlated with age: The average

58
 Interpreting PAI Clinical Scale Elevations

score for persons 18 to 29 years of age is 55T, whereas it is 46T for those above
age 60. Nonetheless, scores above 7OT are reflective of identity issues beyond
what is expected during adulthood, regardless of age. With more extreme scores
(i.e., > 80T) this may involve quite sudden and unpredictable reversals in life
plans and directions; more modest elevations suggest feelings of emptiness, lack of
fulfillment, and boredom. Elevations also suggest a fair degree of anxiety around
identity issues and disruption or dysfunction within the family of origin is a pos-
sibility to be explored. Scores at the low end of BOR-I (i.e., < 45T) suggest a more
stable and fixed self-concept. In many cases, this represents a strength, but it can
also involve a therapeutic challenge if there are strongly fixed negative elements to
the person's identity.

BOR-N: Negative Relationships

The concept of “negative relationships” involves the interpersonal presentation
of borderline personality: a tendency to repeatedly become involved in relation-
ships that are very intense and chaotic. High scores on BOR-N are an indication
that the person’s closest attachment relationships are likely to be stormy; these
relationships might include one’s family, spouse or partner, or therapist. Part of the
storminess revolves around the borderline individual's experience that important
other people have not met his or her needs. They approach such relationships with
a great deal of longing and hope (which may be where the supposed “idealization”
originates); invariably, however, the borderline individual eventually comes away
feeling not just disappointed, but betrayed and exploited. To some extent, this
stems from the general affective reactivity of the borderline personality described
earlier (i.e., a fairly small slight can generate a very catastrophic response). How-
ever, the research literature indicates that borderline patients have extremely high
rates of physical and sexual abuse during childhood (Herman, Perry, & Van der
Kolk, 1989). With this background, it is easy to understand the borderline indi-
vidual’s fear that the people who are closest are likely to exploit him or her. The
BOR-N items tap this perception of betrayal in past relationships, as well as a dis-
trust and pessimism surrounding future relationships.
Considered in isolation, the BOR-N scale reflects a history of involvement in
ambivalent, intense, and unstable relationships. At extreme scores (i.e., > 80T), the
person is quite bitter and resentful about the way past relationships have gone,
feeling betrayed by the people who were once closest and preoccupied with fears
of abandonment or rejection by those who are currently important to him or her.
Scores between 7OT and 80T suggest numerous problems and failures in past
attachment relationships, although intense feelings of past exploitation are less
likely in this range than in higher scores.

59
PAI Interpretive Guide

The concept of “ambivalence” is often raised in discussions of borderline per-

sonality, usually in reference to the putative defense of “splitting.” In splitting, the
person is presumably unable to integrate the positive and negative elements of
another person, and this results in alternating periods of extreme idealization and
devaluation of important others. However, as discussed in relation to BOR-I, the
ambivalence that seems more central to the borderline personality is not one
between good and bad, but one between need and fear. The BOR-I and BOR-N
scales together capture this latter, fundamental ambivalence in borderline individ-
uals: the profound need for others in order to establish who they themselves are,
a tremendous distrust of these critically important people, and an expectation that
they are going to be exploited or abused. Obviously, a person entering a relation-
ship with this set of expectations is likely to experience problems, both in non-
clinical interpersonal relationships and in therapeutic ones. If both BOR-I and
BOR-N are elevated upon entering therapy, the treating clinician is likely to be
taken aback by both the intensity of the client’s need for the therapist and his or
her readiness to perceive in the clinician signs of rejection, disinterest, or abuse,
including any efforts the therapist might make to set limits on the relationship.

BOR-S: Self-Harm
The final borderline subscale reflects a tendency to act impulsively without
much attention to the consequences of those acts. Such acts will thus be viewed
by others as self-damaging or self-destructive (e.g., substance abuse, sexual reck-
lessness, or quitting a job suddenly with no future job prospects). BOR-S is some-
times mistaken for a direct indicator of suicidal behaviors or self-mutilation.
Although a person with a high score on BOR-S would be expected to more at risk
for such behaviors than someone with a low score, the scale is more directly reflec-
tive of impulsivity than of either suicide risk or self-mutilation. Although a sample
of self-mutilators did yield elevated BOR-S scores (Morey, 1991), not all elevations
on BOR-S will involve self-mutilation. Similarly, whereas persons currently on sui-
cide precautions scored above the mean on BOR-S, their average scores were only
around 60T (Morey, 1991). Because many completed suicides are quite premedi-
tated and are not impulsive acts, BOR-S is probably neither sensitive nor specific
if used in isolation as a suicide indicator.
Extreme elevations on BOR-S (e.g., above 85T) reflect hazardous levels of
impulsivity and recklessness. These individuals are impulsive in areas that have
high potential for negative consequences (e.g., spending money, sex, substance
abuse). Such behavior has typically interfered repeatedly with effective social or
occupational performance, or both. High scorers may also be at increased risk for

60
 Interpreting PAI Clinical Scale Elevations

self-mutilation and suicidal behavior, and accompanying SUI elevations may indi-
cate a risk for impulsive suicide gestures.

BOR Full Scale Interpretation

The configuration of the BOR subscales is critical in assigning DSM-based diag-
noses of borderline personality; if three or four of the subscales are elevated, the
person is likely to meet the criteria for the disorder. However, a similar conclusion
should not be drawn from elevations of the full scale. The full scale score is prob-
ably better considered in line with Kernberg’s (1975) view of borderline personal-
ity as a level of personality organization or adaptation that ranges somewhere
between neurosis and psychosis. Thus, low scorers will tend to be fairly healthy
with respect to personality issues, whereas high scorers will present with fairly
primitive concerns, perhaps across many different variants of personality disorder
as they are categorized in the DSM manuals. Diagnostically, if the full BOR scale is
elevated, it is a sign of problems in the personality realm, whereas the configura-
tion of the subscales can confirm whether the problems are classically borderline
(i.e., elevations on three or four subscales) or circumscribed problems associated
with other issues (e.g., BOR-N reflecting relationship problems stemming from
posttraumatic stress disorder).

Average scores on BOR (i.e., < 60T) reflect a person who reports being emo-
tionally stable and who also has stable relationships. Scores between 60T and 70T
are indicative of a person who may be seen as moody, sensitive, and having some
uncertainty about life goals; scores in this range are not uncommon in young
adults. Toward the upper end of this range, individuals may be increasingly angry
and dissatisfied with their interpersonal relationships. Individuals with scores
above 7OT are likely to be impulsive and emotionally labile; they tend to feel mis-
understood by others (who often perceive them as egocentric) and find it difficult
to sustain close relationships. They tend to be angry and suspicious and, at the
same time, anxious and needy, making them quite ambivalent about interactions
with others. However, scores in this range do not necessarily suggest a diagno-
sis of borderline personality disorder unless there are prominent elevations on
each of the four BOR subscales, because individual features are common to other
disorders.
BOR scores that are markedly elevated (i.e., > 90T) are typically associated with
personality functioning within the borderline range. These individuals typically
present in a state of crisis, often regarding difficulties in their relationships. With
elevations in this range, respondents are invariably hostile and feel angry and
betrayed by the people around them. Symptomatically, they often report being

61
PAI Interpretive Guide

very depressed and anxious in response to their circumstances. They are impulsive
and will act in ways that appear to others to be quite self-destructive; for example,
they seem to sabotage their own best intentions with acting-out behaviors. These
behaviors can include alcohol or drug abuse, suicidal gestures, or aggressive out-
bursts; scores on ALC, DRG, SUI and AGG should be consulted to identify poten-
tial problem areas of this type.

BOR Subscale Configurations

The following sections describe the implications of particular combinations
where two or more BOR subscales are elevated together.

BOR-A high, BOR-I high, BOR-N high, BOR-S high

This pattern of scale elevations suggests difficulties in numerous areas: emo-
tional instability, volatile interpersonal relationships, anger, identity disturbance,
and impulsivity. The respondent is likely to be quite emotionally labile, manifest-
ing fairly rapid and extreme mood swings, and, in particular, is probably quick to
display intense and poorly controlled anger. There is also uncertainty about major
life issues, with little sense of direction or purpose in life. A history of involvement
in intense and volatile relationships is likely, as well as preoccupations with fears
of being abandoned or rejected by those important to the respondent. The indi-
viduals response to these perceived interpersonal rejections is likely to involve
impulsive acts that are likely to be self-harmful or self-destructive (e.g., spending
money, sex, substance abuse). This pattern of behaviors is consistent with a diag-
nosis of Borderline Personality Disorder.

BOR-A high, BOR-I high, BOR-N high, BOR-S average

This subscale configuration suggests difficulties with emotional instability,
volatile interpersonal relationships, underlying anger, and identity issues. Such
individuals are likely to be quite emotionally labile, manifesting fairly rapid and
extreme mood swings and reactive anger. There is likely to be much uncertainty
and ambivalence surrounding major life issues, goals, values, and close relation-
ships. These latter relationships are likely to be intense and volatile, with rumina-
tive fears of abandonment, rejection, or exploitation. The comparatively lower
score on BOR-S is a positive sign in that it suggests that, although such people may
respond dramatically to affectively arousing situations, this response does not typ-
ically involve self-destructive impulsive acts.

BOR-A high, BOR-I high, BOR-N average, BOR-S high

This pattern suggests difficulties with emotional instability, anger, identity dis-
turbance, and impulsivity. Such individuals are likely to be quite emotionally

62
 Interpreting PAI Clinical Scale Elevations

labile, manifesting fairly rapid and extreme mood swings; in particular,

they tend
to experience episodes of poorly controlled anger. There also appears to
be uncer-
tainty about major life issues and little sense of direction or purpose in life at
this
time. They are also quite impulsive and prone to behaviors that are likely to
be
self-harmful or self-destructive (e.g., spending money, sex, substance abuse): there
also may be increased risk for self-mutilation or suicidal behavior, and scores
on
SUI should be examined. The relatively lower score on BOR-N suggests that these
individuals may be devoting considerable efforts to maintaining their relationships
in the face of their anger and impulsivity; they may experience considerable guilt
following impulsive or angry acts, and their contrition may serve to sustain rela-
tionships that would otherwise crumble.

BOR-A high, BOR-I average, BOR-N high, BOR-S high

This configuration of BOR subscales suggests difficulties in emotional control,
volatile interpersonal relationships, and notable impulsivity. Such individuals are
likely to be quite emotionally labile, manifesting fairly rapid and extreme mood
swings and, in particular, episodes of poorly controlled anger during which they
lash out at the persons closest to them. It is likely that, as a result, they have a
history of involvement in intense and volatile relationships that may lead to a pre-
occupation with fears of being abandoned or rejected by important others. They
also are likely to be impulsive in other areas, prone to behaviors that are likely to
be self-harmful or self-destructive (e.g., spending money, sex, substance abuse);
they may also be at increased risk for self-mutilation or suicidal behavior, and the
score on SUI should be examined. The comparative lack of elevation on BOR-I may
suggest that this is a relatively fixed (as opposed to a reactive) pattern of behavior
and, thus, may be quite difficult to change.

BOR-A average, BOR-I high, BOR-N high, BOR-S high

This pattern suggests a history of involvement in volatile interpersonal rela-
tionships, a poorly formed personal identity, and noteworthy impulsivity. The his-
tory of relationship problems may have left such individuals preoccupied with
consistent fears of being abandoned or rejected by those around them. Such pre-
occupations are worsened by an uncertainty about major life issues and a lack of
sense of direction or purpose indicated by the elevation on BOR-I, such individu-
als are not certain what to do without important others to guide them. The pattern
also includes marked impulsivity, suggesting a tendency to display behaviors likely
to be self-harmful or self-destructive (e.g., spending money, sex, substance abuse).
These behaviors are likely to be most prominent following disruptions or crises in
close interpersonal relationships and may reflect “acting-out” as a way of warding
off the experience of unpleasant affects.

63
PAI Interpretive Guide

BOR-A high, BOR-I high, BOR-N average, BOR-S average

This configuration suggests that issues of affect control and identity formation
are important personality problems for such individuals. They are likely to be
quite emotionally labile, manifesting fairly rapid and extreme mood swings; in par-
ticular, their anger may be poorly controlled. There is also an uncertainty about
major life issues, and their sense of direction or purpose in life probably vacillates
in relation to their mood. The comparative lack of impulsivity or severe interper-
sonal disruption may indicate that much of their anger is directed internally, rather
than impulsively expressed toward the people around them.

BOR-A high, BOR-I average, BOR-N high, BOR-S average

This subscale configuration suggests that the person is likely to be quite emo-
tionally labile and moody, with anger management likely to be an issue. This emo-
tionality and hostility probably have contributed to an apparent history of involve-
ment in stormy and volatile relationships. The comparative lack of impulsivity
may suggest that, during most normal situations, a fair amount of effort to control
emotions is being expended. However, during times of stress, particularly during
heated interpersonal conflict, the person is likely to react with sudden emotional
outbursts.

BOR-A average, BOR-I high, BOR-N high, BOR-S average

This configuration represents the BOR “splitting” duo of interpersonal need
(represented by BOR-I) and interpersonal conflict and distrust (represented by
BOR-N). There is uncertainty about major life issues and a lack of direction or pur-
pose in life. This uncertainty is likely to extend to the arena of interpersonal rela-
tionships, as such individuals may have a very unstable sense of what they desire
from these interactions. As a result, a history of involvement in intense, needy, and
short-lived relationships is likely, and they tend to be preoccupied with consistent
fears of being abandoned or rejected in these relationships. As a result, they are
quick to perceive in others any sign of real or imagined rejection, disinterest, or
abuse, often including any efforts another person might make to put limits on the
relationship.

BOR-A high, BOR-I average, BOR-N average, BOR-S high

This pattern indicates emotional lability, with fairly rapid and extreme mood
swings and, in particular, episodes of poorly controlled and impulsively expressed
anger. However, rather than being directed at others, the anger may be self-
directed, resulting in behaviors likely to be self-harmful or self-destructive (e.g.,
irresponsible spending, sex, substance abuse). Any angry gestures that are out-
wardly directed may be followed by considerable guilt, and their contrition may

64
 Interpreting PAI Clinical Scale Elevations

serve to sustain relationships that would otherwise suffer. Such individuals

may be
at increased risk for self-mutilation or suicidal behavior during times of affective
turmoil, and the score on SUI should be examined.

BOR-A average, BOR-I high, BOR-N average, BOR-S high

This configuration indicates pronounced uncertainty about major life issues
and a lack of direction or purpose in life as it currently stands. This is likely to be
exacerbated by impulsivity, such that frequent and impulsive changes of direction
in vocational interests, hobbies, religion, or other social roles may be the norm.
There may also be some more overtly self-harmful or self-destructive behaviors
(e.g., spending money, sex, substance abuse).
BOR-A average, BOR-I average, BOR-N high, BOR-S high
This subscale pattern suggests a history of involvement in intense and short-
lived relationships. These relationships may be impulsively ended, or they may
dissolve due to the respondent’ tendency to engage in behaviors likely to be self-
harmful or self-destructive (e.g., spending money, sex, substance abuse). This pat-
tern makes such individuals pessimistic about relationships, and they may be pre-
occupied with fears or expectations of being abandoned or rejected. The pattern
of impulsivity and volatile relationships may place such individuals at increased
risk for self-mutilation or suicidal behavior, particularly during times of marked
conflict in relationships, and scores on SUI should be examined.

Antisocial Features (ANT)

The ANT scale is the second of the two scales (BOR being the other) that specifi-
cally assess character pathology. These two constructs were selected for the PAI
because, together, they account for nearly all empirical research that has been con-
ducted on personality disorders. However, it is important to note that the repre-
sentation of antisocial personality on ANT departs more than the BOR scale from
the DSM conceptualization of the disorder.
The history of this construct is an interesting one. The origins of the concept
of Antisocial Personality Disorder are generally traced back to Pinel’s notion of
manie sans delire (madness without delirium) described at the turn of the 19th cen-
tury. This concept was one of the first to describe a mental disorder that did not
include a defect in reasoning; for this reason, Pinel’s concept has been described as
the forerunner of all modern theory on personality disorders (Mack, 1975). Grad-
ually the concept acquired an element of defects in morality, and it eventually
evolved into a notion resembling one of the “born criminal”; Koch (1891) selected

65
PAI Interpretive Guide

Table 2-1
Cleckley’s (1941) 16 Diagnostic Indicators of Psychopathy
Superficial charm and good “intelligence”
Absence of delusions and other signs of irrational thinking
Absence of “nervousness” or psychoneurotic manifestations
Unreliability
Untruthfulness and insincerity
Lack of remorse or shame
Inadequately motivated antisocial behavior
Poor judgment and failure to learn by experience
Pathologic egocentricity and incapacity for love
—enGeneral poverty in major affective reactions
SL
eh
Coke
ey
ae
EEN
Be
= _ . Specific loss of insight
Unresponsiveness in general interpersonal relations
. Fantastic and inviting behavior with drink and sometimes without
. Suicide rarely carried out
. Sex life impersonal, trivial, and poorly integrated
ee
ee Oar
wD
. Failure to follow any life plan

the term psychopathic inferiority for this condition to emphasize its purported con-
stitutional basis, and this term served as the foundation of the term “psychopath.”
Perhaps the most influential development in the evolution of this concept was
the publication of The Mask of Sanity by Cleckley (1941). This book made explicit
the personological features that set the psychopathic personality apart from crim-
inality. Among the features Cleckley stressed as pathognomonic of this personality
constellation were a lack of guilt, a general absence of anxiety or depression, and
a seeming inability to learn from experience. For assistance in diagnosis, Cleckley
described 16 signs that have become firmly embedded in the clinical lore sur-
rounding this syndrome; these 16 features are presented in Table 2-1.
The DSM-III (1980) conceptualization of Antisocial Personality Disorder rep-
resented a substantial departure from the notion of psychopathy. The DSM-III def-
inition was based extensively on a history of delinquent or antisocial behavior, in
contrast to the personality elements described by Cleckley and others. To a large
extent, these behaviors were derived from the well known study by Robins (1966),
which attempted to establish the adolescent antecedents of antisocial behavior in
adults. However, these criteria seem to tap a somewhat different population than
did the older “psychopathic personality” concept.

66
 Interpreting PAI Clinical Scale Elevations

One difficulty with the representation of this construct in the DSM-III (1980)
and its successors is that, in failing to include the more personological elements of
the construct, it misses critical motivational differences for antisocial behavior.
Some have criticized the DSM definition as being practically synonymous with
criminal behavior; for example, at least half (if not more) of inmates will meet such
criteria for the disorder (Hart & Hare, 1989). Others have expressed the concern
that the DSM focus on delinquent behaviors leads to an overapplication of the
diagnosis to lower socioeconomic groups, missing “white-collar” variants of the
disorder. Finally, there is some support for the conclusion that the concept of psy-
chopathy may be more valid than the DSM representation of this disorder. For
example, some studies (e.g., Hart, Kropp, & Hare, 1988; Serin, Peters, & Barba-
ree, 1990) indicate psychopathy ratings are more useful than the DSM concept of
antisocial personality in predicting recidivism in prisoners.
Hare’s approach (Hare et al., 1988) to the representation of psychopathy has
been found to have two different components or factors. One of these is a behav-
ioral component that involves a variety of antisocial acts; this factor corresponds
reasonably closely to the DSM-III (1980) conceptualization. However, the second
factor involves a component of psychopathy that incorporates personality traits,
such as tendencies to be unempathic, callous, or egocentric. The inclusion of such
traits in the conceptualization of the disorder increases predictive validity; for this
reason, the PAI was constructed to assess each of these facets. The final version of
the ANT scale included a total of three facets, one (ANT-A) assessing antisocial
behaviors, and the remaining two (ANT-E, ANT-S) tapping antisocial traits.

ANT-A: Antisocial Behaviors

The items comprising the ANT-A subscale inquire about antisocial acts during
both adolescence and adulthood. High scorers are likely to have manifested a con-
duct disorder during adolescence, and during adulthood they may have been
involved in illegal occupations or engaged in criminal acts involving theft, destruc-
tion of property, and physical aggression toward others. This subscale of ANT is
the one that corresponds most closely to the more behavioral DSM-III (1980) and
DSM-III-R (1987) definition of the disorder, as it reflects an individual who commits
antisocial acts. The subscale in isolation does not, however, indicate psychological
attributes underlying these acts. Such behaviors could arise from impulsivity,
from egocentricity or entitlement, from environmental presses, or from anger-
management problems. Inspection of other PAI scales and subscales can shed light
on each of these potential sources.
Scores above 70T on ANT-A reflect a history of difficulties with both authority
and social convention. A pattern of antisocial behavior was probably first evident

67
PAI Interpretive Guide

in adolescence, and, with scores in this range, it is likely that the pattern has con-
tinued into adulthood. Scores in the moderate range (i.e., 60 to 69T) may be more
likely than more elevated scores to reflect historical problems. However, because
many of the questions on the subscale are historical in nature, a past history of
such acts can lead to elevations that may not reflect current functioning. For
example, the item “I’ve done some things that weren't exactly legal” might be
referring to behaviors that occurred 30 years earlier. Scores that are very low
(i.e., < 40T) could indicate a very conforming, perhaps moralistic individual, or
perhaps, a person motivated to deny any history of mischievous behavior what-
soever.

ANT-E: Egocentricity
The items comprising the ANT-E subscale tap a callousness and lack of empa-
thy in interactions with others. It is this personological component that is proba-
bly closest to the classic definition of the “psychopath,” yet, in isolation, this scale
does not imply psychopathy. Instead, it suggests a certain self-centeredness that
also could be suggestive of a histrionic or narcissistic personality pattern. However,
in combination with acting-out behavior (ANT-A) and anger-management prob-
lems (AGG), the likelihood of psychopathy as opposed to other issues increases
considerably. It should also be recognized that higher scores are obtained in
younger people; the average score for individuals 18 to 29 years of age is 56T.
High scorers on ANTE (i.e., 2 70T) tend to be seen as egocentric, with little
regard for others or for the opinions of the society around them. In their desire to
satisfy their own goals and impulses, they may take advantage of others, even those
who are closest to them. They feel little responsibility for the welfare of others and
have little loyalty to their acquaintances. Such individuals would be expected to
place little importance in their social role obligations (e.g., as a spouse, parent, or
employee). Although they may describe feelings of guilt over past transgressions,
they are not likely to feel much remorse of any lasting nature, as their inflated
sense of self and their feelings of entitlement would make them unlikely to believe
that they were in the wrong. Such people may be perceived by others as hostile,
but, aside from irritability, there may be little affective involvement in their inter-
actions with others. More marked anger and hostility, if present, will be identified
by elevations on AGG and PAR, rather than on ANT-E.
Moderate elevations on ANT-E (i.e., 60T to 69T) suggest a person who tends to
be self-centered and pragmatic in interactions with others. Such people feel rela-
tively little social anxiety or guilt, and, therefore, they may be quite effective in
superficial social contacts. However, long-lasting relationships may be less suc-
cessful, as these individuals rarely will place others’ needs before their own. In

68
 Interpreting PAI Clinical Scale Elevations

contrast, scores that are very low (i.e., < 40T) suggest a person who may repeat-
edly place others’ needs first and, as such, have difficulty getting his or her own
needs met. In combination with below-average scores on MAN-G, this suggests a
humility that is driven by low self-esteem.

ANT-S: Stimulus Seeking

The ANT-S items tap a personality component associated with a willingness to
take risks and a desire for novelty. Although individuals with antisocial personal-
ity score considerably above the average on most sensation-seeking scales, this trait
is certainly not specific to this diagnostic group, nor is it in isolation a pathologi-
cal, or even an undesirable, characteristic. However, in combination with other
traits (e.g., lack of empathy, poor impulse control, or anger management prob-
lems), this characteristic can lead to a variety of problem behaviors, because the
inhibiting effects of anxiety are minimized. Thus, in relation to other PAI scales,
ANTS has a disinhibition component that might heighten the impact of elevations
beyond what might be expected otherwise. As is true of the other ANT subscales,
ANT-S scores tend to be higher in younger individuals (i.e., the average score is
56T in 18- to 29-year-olds), perhaps lending empirical support to the notion of
“the recklessness of youth.”
High scorers on ANT-S (i.e., 2 70T) are likely to manifest behavior that is reck-
less and potentially dangerous to themselves and/or those around them. They
crave novelty and stimulation; easily bored by routine and convention, they may
act impulsively in an effort to stir up excitement. Their desire for new experiences
may lead to periods of nomadic wandering and make any long-term commitments
unlikely. They also tend to be less anxious than most people, even in situations
where anxiety should be expected. More moderate elevations (i.e., 60T to 69T)
suggest a more controlled, but still potentially reckless, individual. In this range,
however, the trait may not have led to difficulties. However, accompanying eleva-
tions on ANT-A, AGG, BOR-S, ALC, or DRG are all signs that novelty is being
sought in self-destructive, acting-out ways.
Very low scores on ANT-S (i.e., $ 40T) suggest a person who is very timid and
avoidant of novelty. These people are likely to feel uneasy over disruptions in routine,
and ARD-P should be examined for the possibility of phobic avoidance behaviors.

ANT Full Scale Interpretation

At the full scale level, the ANT scale provides an assessment of personality and
behavioral features relevant to the constructs of antisocial personality and psy-
chopathy. As noted earlier, ANT item content ranges from indicators of egocen-
tricity, adventuresomeness, and poor empathy to items addressing antisocial

69
PAI Interpretive Guide

attitudes and behaviors; as a result, individuals with average-to-moderate eleva-

tions can have quite different constellations of features. It is the conjunction of ele-
vations on the three subscales that is suggestive of the psychopath; however, a per-
son with antisocial behaviors, but without psychopathic personality features, may
achieve a full scale elevation on ANT solely through the elevated ANT-A.
Average scores on ANT (i.e., < 607) reflect individuals who report being con-
siderate and warm in their relationships with others; these individuals also typi-
cally exhibit reasonable control over impulses and behavior. Scores between 60T
and 70T are indicative of individuals who may be seen as somewhat impulsive and
risk-taking; scores in this range are fairly common in young adults, particularly in
young men (i.e., the average T score for such a group approaches 60T). Toward the
upper end of this range, individuals may be increasingly self-centered, disinhib-
ited, skeptical of other's intentions, and unsentimental in their interpersonal rela-
tionships. With scores above 7OT, the respondent is likely to be impulsive and
hostile, and there may be a history of reckless or antisocial acts. Such individuals
may be seen by others as callous in their relationships, and long-lasting friendships
tend to be the exception to the rule.
When ANT scores are markedly elevated (i.e., > 82T) individuals typically dis-
play the prominent features of antisocial personality disorder. They are likely to be
unreliable and irresponsible and probably have had little sustained success in
either social or occupational realms. They tend to have a coldly pragmatic
approach to relationships and will exploit such interactions to suit their own
needs. Such people tend to be impulsive in their approach to life and have a his-
tory of conflicts with authority figures.

ANT Subscale Configurations

The following sections describe the implications of particular combinations
where two or more ANT subscales are elevated simultaneously.

ANT-A high, ANT-E high, ANT-S high

This triumvirate represents the pattern associated with the classic formulation
of the psychopath. There is a history of antisocial behavior that likely began dur-
ing adolescence, and, given the personality attributes of egocentricity and sensa-
tion seeking, this pattern has probably persisted to the present time. Such people
tend to have little regard for others or for the opinions of society. In order to sat-
isfy their own impulses, they will take advantage of others, and there is likely to
be little sense of loyalty, even to those who are closest to them. Such people
approach life in a reckless manner, entertaining risks that are poorly motivated

70
 Interpreting PAI Clinical Scale Elevations

and potentially dangerous to themselves and to those around them. Social-role

responsibilities are likely to be neglected in favor of pursuing novelty and excite-
ment; old occupations and old relationships lose their appeal quickly for such
individuals. Although feelings of guilt over past transgressions may be reported, it
is unlikely that there is real remorse of any lasting nature.

ANT-A high, ANT-E high, ANT-S average

This subscale pattern suggests a history of antisocial behavior reflecting more
of a callous disregard for others than a desire for excitement. Conduct problems
probably date back to adolescence, and are most likely to represent a pattern of
illegal occupations or acts motivated by personal gain. Such acts may involve plan-
ful exploitation, rather than impulsive acting-out, particularly if BOR-S reveals no
elevation. People with this pattern are likely to be egocentric, with little regard for
others or for the opinions or conventions of the society. Substantial feelings of loy-
alty or remorse are unlikely, and responsibility for the history of behavioral diffi-
culties is likely to be projected outward, especially with above-average scores on
PAR-R.

ANT-A high, ANT-E average, ANT-S high

This pattern suggests a history of impulsive and poorly motivated antisocial
acts and behaviors, likely beginning with a conduct disorder during adolescence.
Such people display reckless and risky behaviors that are potentially dangerous to
themselves and to those around them. Some of these behaviors may have involved
destruction of property, and physical aggression toward others may have been
part of the picture (i.e., inspect for elevated scores on AGG-P). As many of the
acts may have been impulsive, rather than premeditated, respondents may expe-
rience genuine remorse for their behavior, but feel unable to control or prevent
repeat occurrences.

ANT-A average, ANT-E high, ANT-S high

This pattern suggests an individual who may appear successful and effective,
but who is ultimately likely to be self-centered and irresponsible in dealing with
social and vocational obligations. Although the individual may be able to conform
to social convention in order to avoid negative consequences, this pattern reflects
a lack of empathy or respect for others. In their desire to satisfy their own impulses
or needs, such people may exploit others, regardless of the closeness of the rela-
tionship. For this reason, relationships with others are predictably short lived due
to the predatory and manipulative behavior that characterizes such people.
Although guilt over past transgressions may be professed, it is unlikely that there
is remorse of any lasting nature. Dangerous risks may be taken, resulting from the

i
PAI Interpretive Guide

desire for personal gain as well as the sheer excitement of the danger, and such
people may not hesitate to expose others to similar risks.

Substance Abuse Scales

The PAI includes two scales pertinent to substance abuse, one measuring alco-
hol problems (ALC) and one related to drug use and abuse (DRG). Alcohol and
drug problems are common among patients with mental disorders, but, at times,
these problems are overlooked when more dramatic psychological problems are
evident. The frequency of such problems merited the inclusion of this scale within
a broad-band diagnostic instrument. As is true of the other clinical scales, items for
ALC and DRG vary along a continuum of severity. The measurement model for
ALC and DRG was patterned after the approach taken by Edwards and Gross
(1976), who emphasized two facets of alcohol problems: core features of alcohol
dependence, such as withdrawal symptoms and loss of control over drinking, and
alcohol-related disabilities, such as social or legal consequences of drinking. Sub-
sequent work (e.g., Edwards, Arif, @ Hodgson, 1982) suggested that a similar pat-
tern could be found in the drug abuse area. Because of the high interrelationship
between dependence and disability, ALC and DRG were designed as unitary scales
without subscales; however, this is not meant to imply that alcohol-related prob-
lems are either unitary or homogeneous. The ALC and DRG scale items were writ-
ten to identify the presence and severity of alcohol and drug-related problems.
Once such a problem has been identified, a more specialized assessment device
(i.e., one predicated on the assumption that the respondent has problems with
substance abuse) may be used to further pinpoint the nature and pattern of alco-
hol or substance use.
The ALC and DRG scales share certain features that are critical in evaluating
respondents’ scores on these scales. First, a good deal of the information gathered
on these scales is historical (i.e., inquiries are made about events that may have
happened in the past). These historical items reflect major milestones or major
markers that exist in the development of a substance abuse behavior pattern (e.g.,
Jellinek, 1960), and it is these markers that are critical in assigning diagnoses
under most widely used diagnostic systems, including the DSM. As such, ALC or
DRG, or both, can be elevated in people who have had a substance abuse problem
in the past, but who are not currently drinking or using drugs. An individual who
has a current substance abuse problem will tend to have scores that are quite ele-
vated. However, it is certainly possible for a person to score in the vicinity of 70T
on either scale largely through historical information. A “recovering” alcoholic who
has been abstinent for 10 years still might obtain an elevated score on ALC if, for

72
 Interpreting PAI Clinical Scale Elevations

example, he or she has lost jobs or has experienced withdrawal symptoms

during
past episodes of heavy drinking. Thus, moderate elevations on these scales should
be followed up with some inquiry about current or recent substance consumption
patterns.
In rare instances, respondents may refuse to answer ALC items and, particu-
larly, DRG items, claiming the items are not relevant because they do not use alco-
hol or drugs. This has been most commonly observed in individuals who approach
the test in a suspicious or legalistic manner; for instance, such responses are some-
times found in preemployment screening applications of the test. For example,
such people will not answer an item such as, “My drug use has never caused prob-
lems for me,” because they feel this would be admitting to using drugs. In such
instances, it is recommended that the respondent be asked to consider all types of
drugs, not just illegal or street drugs: prescription medication, over the counter
preparations, and so forth. A refusal to respond to these items is most likely not
to indicate hidden substance abuse; rather, it suggests that the test is being
approached in a very careful and guarded manner, and this may be of use in eval-
uating the test results.

Examining Substance Abuse Denial

A feature shared by the ALC and DRG scales is that both address substance use
and problems directly related to substance use. In other words, the item content is
not subtle; hence, the scales are susceptible to denial, a problem of concern to
many professionals in the substance abuse field. This direct method of inquiry is
potentially problematic in a population noted for denial and dishonesty, and a
number of writers have questioned the validity of such self-reports (Fuller, Lee, &
Gordis, 1988). However, the general results of studies support the direct ques-
tioning method used in the PAI. For example, Sobell and Sobell (1975) found that
the self-report of alcoholics about information (later verified through contact with
agencies such as the FBI, the Department of Motor Vehicles, and state and county
hospitals) was quite accurate, and that overestimates of problems by the patients
were more frequent than underestimates. Another study of this issue (Hesselbrock
et al., 1983) found that self-reported drinking estimates were supported by collat-
eral informants and also were good predictors of post-discharge drinking. Fur-
thermore, strategies that rely on a covert assessment of substance abuse tend to
have dubious validity. Physiological markers of alcoholism (e.g., use of various
liver function tests) generally have much lower sensitivity and specificity than self-
report measures (Bernadt et al., 1982; Skinner et al., 1986). Indirect psychologi-
cal markers of substance abuse have also been of limited utility. For example, the
MacAndrew scale (1965), which was designed to covertly identify alcohol use

73
PAI Interpretive Guide

from the MMPI item pool, has been found to correctly identify only 25% of alco-
holics in inpatient treatment programs (Colligan et al., 1990). Given such findings,
the direct content-based approach was taken in the PAI. However, if a person is
motivated to deny substance use or the problems associated with such use, this
will affect scores on these scales. It is easy to imagine why, in certain contexts,
someone would deny use of illegal drugs, and the test user must be aware of this
potential factor.

To some extent, the problems in identifying denial of alcohol and/or drug

abuse are similar to those of defensiveness in general. As such, the general strate-
gies for identifying defensive responding on the PAI (described in chapter 5) can
be useful within the specific domain of substance use. For example, Fals-Stewart
(1996) evaluated the ability of the PIM score to identify individuals attempting to
deny substance abuse problems. He compared patients receiving treatment for
drug abuse and normal controls with two “questionable responding” groups, one
a group of drug abuse patients instructed to respond defensively, the second a
group of respondents receiving the PAI as part of a forensic assessment, who were
referred by the criminal justice system and who had positive urinalysis testing for
recent drug ingestion but had denied drug use during the past 6 months. Fals-
Stewart (1996) found that the optimal cutting score for PIM (T > 56) described in
the PAI Professional Manual (Morey, 1991) successfully identified 88% of the indi-
viduals in the “questionable responding” groups while incorrectly identifying 20%
of controls (both patients and nonclinical respondents) as “questionable.” In other
words, individuals motivated to deny substance abuse problems were more than
four times as likely to score above 56T on PIM than individuals without such moti-
vation. This result demonstrates that PIM is a useful starting point in evaluating
substance abuse denial; chapter 5 provides a more detailed discussion of this scale
and other strategies for identifying general defensiveness that may also be of use in
identifying such individuals.
However, individuals may be specifically motivated to deny alcohol or drug
abuse (for example, in the context of pre-employment screening) although not
necessarily being defensive in describing other domains of their lives. Such indi-
viduals will tend to obtain very low raw scores on ALC and DRG (e:cMtO Ores
reporting that they are teetotalers, that they neither drink nor use drugs of any
sort. Although persons motivated to deny substance use will obtain scores in this
range, so will large numbers of adults in the community, and thus, in most
instances, such low scores are accurate reflections of their use of substances. How-
ever, these low scores should be regarded with some suspicion if the person has
other characteristics that would lead one to expect the person to have at least
experimented with alcohol or controlled substances. Although this approach has

7
 Interpreting PAI Clinical Scale Elevations

limitations (e.g., witness the limited efficacy of the MacAndrew scale),

to a certain
extent these characteristics may be inferred from PAI scale scores. In particular, five
scales demonstrate substantial correlations with both ALC and DRG; these scales
are BOR-S (indicating impulsivity), ANT-S (sensation-seeking), ANT-A (history of
antisocial behavior), ANT-E (interpersonal callousness), and AGG-P (history of
physical aggression). If these five scales are elevated, one would expect ALC or
DRG, or both, to also be elevated, as such behaviors are part of this constellation.
These features represent a personality style that is particularly prone to use of alco-
hol or other substances, and ALC and DRG scores that are markedly low in such
individuals are rare.
To systematize this possibility, simple linear regression estimates of predicted
scores on ALC and DRG using the sum of T scores from these five scales were
derived from the clinical normative data (n = 1,246). The following regression
equations were obtained:

Estimated ALC T score = [0.162184 x (sum of BOR-A, ANT-A, ANT-E, ANT-S, AGG-P)] + 14.39

Estimated DRG T score = [0.199293 x (sum of BOR-A, ANT-A, ANT-E, ANT-S, AGG-P)] + 3.07

For convenience, the predicted estimates for ALC and DRG scores based upon
the sum of these five scales are presented in Table 2-2; this sum correlates at .46
with the ALC scale and .59 with the DRG scale. Obtained scores on the substance
abuse scales that are markedly lower than the estimates provided in Table 2-2 raise
the possibility that some denial of substance problems may be operating. For
example, Figure 2-1 presents the mean PAI profiles of the two “questionable
responding” groups from the study by Fals-Stewart (1996)! described earlier.
There were two such groups in that study. One was a “forensic” group consisting
of 59 individuals referred for evaluation by the criminal justice system; these indi-
viduals (a) reported no illicit drug use or alcohol abuse during the 6 months prior
to the evaluation; (b) expressly refused treatment for substance abuse; and (c) tested
positive on urine assays or breath tests conducted at the time of evaluation, sug-
gesting that one or more psychoactive substances had been recently ingested. The
second group was a “positive dissimulation” group of 59 patients in treatment for
substance abuse problems who had been instructed to deny substance abuse prob-
lems in responding to the PAI. A variety of scenarios were presented to these
patients, such as child custody evaluation, applying for a job, avoiding unwanted

'The author would like to thank Dr. W. Fals-Stewart (personal communication) for providing the
complete PAI means for all scales from the Fals-Stewart (1996) article.

75
 PAI Interpretive Guide

Table 2-2
Predicted ALC and DAG Scores From the Sum of
BOR-S, ANT-A, ANT-E, ANT-S, and AGG-P

Sum of 5 predictor Expected ALC Expected DRG

scales (T scores) T score T score
0 14 3
25 18 8
50 22 13
75 27 18
100 31 23
125 35 28
150 39 33
175 43 38
200 47 43
225 51 48
250 55 53
275 59 58
300 63 63
325 67 68
350 71 73
375 75 78
400 79 83
425 83 88
450 87 93
475 91 98

500 95 103

substance abuse treatment, or undergoing a court-ordered presentencing evalua-

tion. The “positive dissimulation” patients were offered movie passes if they could
avoid detection as having engaged in positive dissimulation and of having prob-
lems with substance use.
The characteristics of the profiles in Figure 2-1 confirm many of the observa-
tions noted in the preceding paragraphs. For example, the PIM elevation in these
groups should immediately raise questions of defensiveness. Also, as will be seen
in chapter 5, the prominent RXR scores seen in these profiles are also an indicator
of generally defensive responding. More specifically, however, this figure demon-
strates that the five substance predictor scales all display some relative elevations
in these groups. Table 2-3 provides a summary of the actual and estimated ALC

76
 Interpreting PAI Clinical Scale Elevations

PROFILE FORM FOR ADULTS-SIDEA 4 2 3 4 5 6 7 8 9 10 11 A B ¢ D E Y z

e—— Forensic
@—-—=-8 Positive
dissimulation

&

® F4

mscn

Jirved

y
\\

ai=)

°
a

bane
lero
Vovselocrebersebecre
4 2 3 4 5 6 ie 8 9 10 a4 A 8 c re) E y z
ICN INF NIM PIM SOM = ANX ARD. OEP = MAN PAR SCZ BOR ANT ALC DRG AGG sul STR NON RXR DOM WRM

PROFILE FORM FOR ADULTS - SIDE B

o—— Forensic
@—=—- Positive
dissimulation
rrbaverdecerl

Vioraebouroda

a
ee
are)
ce

a ro)

a So

nyi) ee
ee
ee
SOM SOM-S SOMH ANXC ANXA ANXP ARD-O ARD-P ARD-T DEP-C DEPA DEP-P MANA MANG MANI PARH PARP PARR SCZP SCZS SCZT BORA BORI BORN BORS ANTA ANTE ANTS AGGA AGGV AGGP
CONV SOMA HEAL COG AFF PHYS OBS PHOB TRAU COG AFF PHYS ACT GRND IRRI HYPE PERS RSNT PSYC SOC THGT AFF ID NEG SELF ANT EGO STIM AGG VERB PHYS
SYMP ZATN CONC SYMP SYMP SYMP COMP IAS STRS SYMP SYMP SYMP LEVL IOSY BLTY VIG CUTN MENT EXP DET DIS INST PROB REL HARM BEH CEN SEEK ATT AGG AGG

Figure 2-1. Mean profiles for groups denying substance abuse problems, adapted from Fals-Stewart
(1996).

1
PAI Interpretive Guide

Table 2-3
Observed and Estimated ALC and DRG Scores for
Groups From the Fals-Stewart (1996) Study
Sum of
mean scores
for BOR-S, Est Obs Est Obs
ANT-A, ANT-E, ALC ALC DRG DRG
Group N ANT-S, AGG-P T-score T-score T-score T-score

Forensic 59 324.8 67 55 68 54
Positive
dissimulation 59 299.3 63 50 63 51
Substance abuse
patients 59 349.2 71 74 73 85

Note. Est = Estimated; Obs = Observed.

and DRG scores for these two groups, as well as for the “standard instruction” sub-
stance abuse treatment group from the Fals-Stewart (1996) study, using the regres-
sion estimates described earlier. For both of the “questionable responding” groups,
the estimated scores on the substance abuse scales exceeded the observed scores
by a considerable margin. The group of substance abuse patients who completed
the test under standard instructions obtained ALC and DRG scores equal to or
above their predicted scores.
The results of these analyses support the conclusion that in instances where the
estimated substance abuse score from Table 2-2 exceeds the observed score by 10T
or more, there is reason to suspect that some denial of substance use may be oper-
ating. When this occurs, discussing substance use with some type of collateral
informant (e.g., a spouse or family member) might be worthwhile. It should be
recognized that any indirect method of ascertaining substance abuse has limited
ability to circumvent denial issues, and asking directly about use of substances is
the most straightforward and most accurate means of obtaining such information
in most cases. Nonetheless, there are situations that provide powerful motiva-
tion to deny such problems, as in forensic situations, custody evaluations, or pre-
employment screenings. In such circumstances, an overall evaluation of of the pro-
file for defensiveness (as discussed in chapter 5) followed by a specific evaluation of
the possibility of substance abuse denial (as described earlier) should be conducted.

Alcohol Problems (ALC)

The ALC scale provides an assessment of behaviors and consequences related
to alcohol use, abuse, and dependence. The item content ranges from statements

78
 Interpreting PAI Clinical Scale Elevations

of total abstinence through frequent use to the severe consequences of drinking,

loss of control, and alcohol-related cravings. Questions inquire directly about the
use of alcohol; thus, prominent denial of alcohol problems can suppress scores on
the scale. If ALC raw scores are very low and there are elevated scores on the five
predictor scales mentioned earlier, some follow-up inquiry about alcohol use
might be appropriate. However, in general, direct inquiry about alcohol use will
usually provide more accurate data than making inferences from indirect sources
of information.
Average scores on ALC (i.e., < 60T) reflect a person who reports a moderate
alcohol intake and few adverse consequences related to drinking. Scores between
60T and 7OT are indicative of a person who may drink regularly and who may
have experienced some adverse consequences as a result. Toward the upper end of
this range, there is increasing likelihood that alcohol has caused or is causing prob-
lems for the person. With scores above 70T, the respondent is likely to meet cri-
teria for alcohol abuse. Such a score indicates that use of alcohol has had a negative
impact on the respondents life. Alcohol-related problems are likely, including diffi-
culties in interpersonal relationships, difficulties on the job, and possible health
complications; the respondent’ current functioning is probably compromised.
ALC scores that are markedly elevated (i.e., above 84T, which is the average
score for individuals in alcoholism treatment centers), are typically associated with
severe alcohol dependence. Such a score indicates that alcohol use has resulted in
a number of adverse consequences for the individual. Numerous alcohol-related
problems are likely, including difficulties in interpersonal relationships, difficulties
on the job, and possible health complications. Such individuals are likely to be
unable to cut down on their drinking despite repeated attempts at sobriety. They
typically feel quite guilty about their drinking, but report little ability to control
the effect it has on their lives. They probably have a history of social and occupa-
tional failures that were related to drinking and have had episodes when they were
intoxicated for prolonged periods. Blackouts and physiological signs of depen-
dence and withdrawal are probable with scores in this range.

Drug Problems (DRG)

The DRG scale provides an assessment of behaviors and consequences related
to drug use, abuse, and dependence. The item content ranges from statements of
total abstinence through frequent use to the severe consequences of drug use.
Questions inquire directly about the use of drugs (both prescription and illicit);
thus, prominent denial of drug use can suppress scores on the scale. As with ALC,
if DRG raw scores are very low and there are elevated scores on the five predictor

19
PAI Interpretive Guide

scales described earlier, some follow-up inquiry about drug use might be appro-
priate. However, in general, direct inquiry about a history of drug use will usually
provide reasonably accurate data in the absence of strong situational pressures
(e.g., in forensic settings or pre-employment screenings) to deny drug use.
Average scores on DRG (ie., < 60T) reflect a person who reports using drugs
infrequently, if at all. Scores between 60T and 70T are indicative of a person who
may use drugs on a fairly regular basis and who may have experienced some
adverse consequences as a result. Toward the upper end of this range there is
increasing likelihood that drug use has caused, or is causing, problems for the per-
son. With scores above 7OT, the respondent is likely to meet criteria for drug
abuse. It is likely that drug use has caused difficulties in interpersonal relationships
or in work performance, and the individual’ current functioning is probably com-
promised.
DRG scores that are markedly elevated (i.e., > 80T, which is the average score
for individuals in treatment for drug abuse) are typically associated with drug
dependence. Such individuals are likely to be unable to cut down on drug use
despite repeated attempts and have little ability to control the effect that the desire
for drugs has on their lives. They probably have a history of social and occupa-
tional failures related to drug use. Depending on the primary substance of abuse,
physiological signs of dependence and withdrawal are probable with scores in this
range.

80
 CHAPTER 3
Two-POINT CODETYPES IN
PROFILE INTERPRETATION
The use of two-point codes in profile interpretation has become somewhat of
a tradition in the assessment field. Although two-point codes provide a starting
point for the configural interpretation of the PAI profile, it is important to note that
such a code provides a severely limited summary of the information contained in
the profile. First, the two-point code obviously ignores the wealth of information
provided by the other test scales. Second, because of the subscale structure of the
PAI scales, meaningful differences on even the two scales that comprise the code
can be observed between individuals who have identical codes. Finally, the relia-
bility of the small differences that can determine a two-point code on any psycho-
logical instrument is often suspect. For example, consider a profile where DEP is
at 85T, ANX is at 82T, and BOR is at 81T. Although this is nominally a DEP-ANX
two-point code, the difference between ANX and BOR is considerably less than one
standard error of measurement, and that difference is not interpretively significant.
Yet, the DEP-BOR codetype has different implications than the DEP-ANX code-
type. Given these limitations, it is best to (a) consider the following descriptions of
codetypes as a rough beginning to interpretation, and (b) examine all relevant
descriptions (e.g., DEP-ANX, DEP-BOR, ANX-BOR in the present example) when
scales determining the codetype fall within one standard error of each other.
The following sections describe the major features and interpretive significance
of the 55 possible PAI two-point codes. Inclusion in one of these codetypes is
based upon the two highest scores on the 11 PAI clinical scales, with each of the
2 scales involving scores of at least 70T. No distinction is provided in these sec-
tions with respect to order of the scales within the code: For example, the DEP-
ANX codetype applies to all profiles for which DEP and ANX have the two highest
clinical scale scores, regardless of which is higher, with both at least 70T. Reported
frequencies and diagnostic correlates of these profiles were derived from Appendix
A of the PAI Professional Manual (Morey, 1991).

81
 PAI Interpretive Guide

SOM-ANX
This pattern suggests a person who is reporting marked distress, with particu-
lar concerns about physical functioning. Such individuals see their lives as severely
disrupted by a variety of physical problems, some of which may be stress-related.
These problems render them tense, unhappy, and probably impaired in their abil-
ity both to concentrate on and to perform important life tasks. The somatic con-
cerns may have led to friction in close relationships, and other people often per-
ceive these individuals as complaining and demanding. Secondary elevations on
ARD and DEP are often observed with this codetype, and the level of STR can be
informative in ascertaining the degree of life disruption associated with the somatic
concerns. This is a relatively common profile configuration, observed in 1.1% of
clinical respondents. Common diagnostic correlates include somatoform disor-
ders, posttraumatic stress, adjustment reactions, and major depression. Interest-
ingly, this codetype also is observed disproportionately in schizophrenia, perhaps
reflecting the onset of somatic delusions.

SOM-ARD
This configuration of the clinical scales suggests a person who has ruminative
concerns about physical functioning. Such people see their lives as disrupted by a
variety of physical problems, some of which may be related to marked stressors;
an inspection of the relative elevations of ARD-T and STR may reveal whether such
stressors involve recent or more long-term events. These problems have left them
tense and worried, and this may have led to disruption in close relationships.
Secondary elevations on ANX and DEP are often observed with this codetype;
elevations in other areas, however, are unusual. This pattern, observed in 0.9% of
clinical respondents, tends to be seen more often in anxiety disorders (including
posttraumatic stress) than in more purely somatoform disorders.

SOM-DEP
This configuration of the clinical scales suggests a person who is reporting sig-
nificant distress, with particular concerns about physical functioning. Such people
see their lives as severely disrupted by a variety of physical problems. These prob-
lems have left them unhappy, with little energy or enthusiasm for concentrating on
important life tasks and little hope for improvement in the future. Performance in
important social roles has probably suffered as a result, and lack of success in
these roles will serve as an additional source of stress. Secondary elevations on
ANX are frequent, and SUI is often elevated; this pattern suggests that some probe
of suicidal ideation is merited when the SOM-DEP codetype is observed. This is
a relatively common profile, observed in 2.8% of clinical respondents. Common

82
 Iwo-Point Codetypes in Profile Interpretation

diagnostic correlates include somatoform disorders, organic mental disorders, and

major depression.

SOM-MAN
This configuration of the clinical scales is rather unusual, as it suggests a per-
son who is reporting significant problems in physical functioning accompanied by
heightened activity levels and irritability. The somatic concerns and emotionally
labile style are likely to have led to some friction in close relationships, and others
may see such people as complaining and demanding. Secondary elevations on
ARD, BOR, and STR are often observed with this codetype, suggesting that both
situational and characterological factors should be considered in evaluating the
somatic concerns. Inspection of DRG also is warranted, as abuse of prescription
drugs may be a risk for this type of individual. This is a relatively uncommon pat-
tern, seen in only 0.2% of clinical respondents. This pattern was observed with
some frequency in patients diagnosed with Schizoaffective Disorder.

SOM-PAR
This configuration of the clinical scales is unusual. It suggests a person with
prominent hostility and suspiciousness who is also reporting significant problems
in physical functioning. Such respondents perceive others as unsympathetic to
their somatic concerns and unsupportive of their perceived limitations. Their hos-
tility has probably led to some friction in close relationships: other people may see
them as complaining and demanding, but the respondents probably attribute the
source of these conflicts to the way that they are treated by others. If presenting for
treatment, they are unlikely to be receptive to examining any psychological factors
that might be associated with their physical complaints, and they probably will be
resistant to psychological interventions. Secondary elevations on NON are often
observed with this codetype, underscoring the resentment they experience toward
the perceived lack of support they receive from family and friends regarding their
health concerns. This is a rare configuration, with only 0.1% of patients in the clin-
ical standardization sample displaying this pattern. Among these patients, individ-
uals with organic mental disorders were disproportionately represented.

SOM-SCZ
This configuration of the clinical scales suggests a person with significant think-
ing and concentration problems accompanied by marked concerns about health
and physical functioning. The somatic complaints may be highly unusual, and, in
some circumstances, can involve somatic delusions. The reported combination

83
PAI Interpretive Guide

of physical limitations and social discomfort severely limits the extent of their
social interactions; whatever few close relationships there are may revolve around
somatic preoccupations. Secondary elevations on DEP are often observed with this
codetype, as are pronounced elevations on NIM, suggesting that the possibility of
symptom exaggeration should be evaluated. This pattern was observed in 0.6% of
clinical respondents. Common diagnostic correlates include schizophrenia and
bipolar disorder, manic episode.

SOM-BOR
This pattern on the clinical scales suggests a person reporting significant prob-
lems in physical functioning who is also hostile and emotionally labile. Such peo-
ple are likely to harbor some bitterness toward important others, who may be
viewed as unsympathetic to the respondents’ somatic concerns and unsupportive
of their perceived limitations. This hostility and emotionality has probably been
the source of friction in close relationships; others are likely to see such people as
complaining and demanding, and these others may view the somatic complaints
as a manipulative means through which the respondent can control the relation-
ship. Secondary elevations on DEP and SUI often are observed with this codetype,
underscoring the distress of such people; the intensity of associated bitterness and
anger is often revealed with elevations on PAR and AGG. This is a relatively rare
profile, observed in 0.3% of clinical respondents. Common diagnostic correlates
include somatoform disorders, posttraumatic stress, and antisocial personality,

SOM-ANT
This configuration of the clinical scales is quite unusual. It suggests a person
who is self-centered and preoccupied with his or her somatic problems to the
exclusion of concern or caring for other people. Others are likely to view such peo-
ple as complaining, self-centered, and demanding, and these others may view the
somatic complaints as a manipulative means through which the respondent can
control the relationship. If in treatment, such clients tend to be very difficult to
work with, as they are typically unreceptive to examining psychological factors
associated with their physical complaints and resistant to psychological interven-
tions. ALC, and particularly DRG, should be examined to determine whether sub-
stance abuse may be contributing to the health issues, or, alternatively, the health
issues May serve as a means to obtain prescription medication. This is a very
uncommon profile pattern, as it was never obtained in the clinical standardization
sample.

84
 Two-Point Codetypes in Profile Interpretation

SOM-ALC
This configuration of the clinical scales suggests a person with a history of
drinking problems who is experiencing a number of physiological difficulties that
may be partially related to the drinking. These somatic problems might involve
withdrawal symptoms, or they might be medical complications of alcohol abuse
(e.g., problems associated with the central nervous system sequelae of alcoholism),
The combination of alcohol use and physical symptomatology is probably caus-
ing severe disruptions in relationships and work, and these difficulties are most
likely serving as additional sources of stress; secondary elevations on STR are
often observed with this codetype. Seen in 1.3% of clinical respondents, the most
common diagnostic correlates include alcohol dependence and organic mental
disorders.

SOM-DRG
This configuration of the clinical scales suggests a person with a history of drug
abuse who is experiencing a number of physiological problems that may be par-
tially related to the use of drugs. These somatic problems might involve with-
drawal symptoms, or they might be medical complications of drug abuse. The
combination of substance use and physical symptomatology is probably causing
severe disruptions in social-role functioning, and these difficulties typically will
serve as additional sources of stress. Secondary elevations on DEP and BOR are
often observed with this codetype; elevations on ANT and BOR may raise the pos-
sibility that the person is at risk for abusing prescription medication associated
with the somatic condition. This profile is observed in 0.6% of clinical respon-
dents, with somatoform disorders predominating.

ANX-ARD
This clinical scale configuration suggests a person with marked anxiety and
tension. Such people may be particularly uneasy and ruminative about their per-
sonal relationships, some of which are probably not going well; these relationships
may be an important source of current distress, and such people tend to respond
to their circumstances by becoming socially withdrawn or passively dependent.
The disruptions in their lives often leave them questioning their goals and priori-
ties and tense and fearful about what the future may hold. Secondary elevations on
DEP and SUI are often observed with this codetype and are prominent as the dis-
tress becomes more debilitating. This is a fairly common profile, observed in 1.9%
of clinical respondents. Common diagnostic correlates include various types of
anxiety disorders as well as major depression.

85
 PAI Interpretive Guide

ANX-DEP
This configuration of the clinical scales suggests a person with significant
unhappiness, moodiness, and tension. Although such people are quite distressed
and acutely aware of their need for help, their low energy level, passivity, and
withdrawal may make them difficult to engage in treatment. Typically, self-esteem is
quite low, and they view themselves as ineffectual and powerless to change their life
direction. Often accompanied by elevations on STR, life disruptions can leave such
people uncertain about goals and priorities and tense and pessimistic about what
the future may hold. They are likely to have difficulties in concentrating and mak-
ing decisions, and the combination of hopelessness, agitation, confusion, and
stress apparent in these scores may place such people at increased risk for self-
harm; secondary elevations on SUI are often observed with this codetype. This pat-
tern is observed in 1.3% of clinical respondents, and it is associated with diagnoses of
Dysthymic Disorder, Major Depressive Disorder, and Borderline Personality Disorder.

ANX-MAN
This is an unusual configuration of the clinical scales. It suggests a person who
is agitated, irritable, and affectively labile. Such people may have a high activity
level that has left them stretched thin and thus hindered in their ability to perform
any of their roles effectively. The resulting strain has left the respondent tense and
feeling overwhelmed by self-imposed demands. Close relationships may have suf-
fered particular strain from the moody and often demanding presentation charac-
teristic of these individuals. Secondary elevations on BOR and STR often are
observed with this codetype. DOM also is often elevated, indicative of the respon-
dent's strong need for control; anxiety is likely to ensue when this control must be
relinquished. This is an uncommon profile pattern, observed in only 0.2% of clin-
ical respondents.

ANX-PAR
This is a relatively unusual configuration of the clinical scales, suggesting a per-
son with prominent hostility and suspiciousness who is also acutely anxious, sen-
sitive, and tense. These individuals tend to demonstrate heightened sensitivity in
social interactions that probably serves as a formidable obstacle to the develop-
ment of close relationships. Although such people may harbor considerable anger
and resentment, the degree of anxiety may lessen the likelihood that this anger is
expressed directly. Secondary elevations on ARD and BOR are often observed with
this codetype. SUI can be quite elevated in these individuals, and any such ideation
should be carefully evaluated given the extent of the hostility and anxiety

86
 Two-Point Codetypes in Profile Interpretation

suggested by the profile. This pattern is uncommon, observed in 0.1% of clinical

respondents.

ANX-SCZ
This configuration of the clinical scales suggests a person with significant
thinking and concentration problems, accompanied by prominent agitation and
distress. Such individuals are likely to be withdrawn and isolated, with few if any
close interpersonal relationships, and may become quite anxious and threatened
by such relationships. Social judgment is probably fairly poor, and such people
tend to have marked difficulty in making decisions, even about matters of little
apparent significance. Secondary elevations on DEP are often observed with this
codetype, which further underscores the extent of the distress and cognitive inef-
ficiency. Seen in 0.9% of clinical respondents, this pattern is most frequently asso-
ciated with diagnoses of Schizoaffective Disorder, Schizophrenia, and Posttrau-
matic Stress Disorder.

ANX-BOR
This configuration of the clinical scales suggests a person who is tense, angry,
unhappy, and emotionally labile. Such people often present in a state of crisis and
marked distress that may be associated with difficulties or rejection (perceived or
actual) in interpersonal relationships. This may be part of a more general pattern
of anxious ambivalence in close relationships, marked by bitterness and resent-
ment on the one hand, and by dependency and marked anxiety about possible
rejection on the other. Secondary elevations are often observed on DEP (suggest-
ing the primacy of the distress) and AGG (when present pointing to significant
underlying anger). This profile is observed in 0.9% of clinical respondents. Common
diagnostic correlates include borderline personality and somatoform disorders.

ANX-ANT
This configuration of the clinical scales is quite unusual. It suggests a person who
is impulsive and self-centered, yet is experiencing considerable anxiety and tension.
Because these two personality elements are so inversely correlated, such respondents
are likely to fluctuate between these disparate elements, with periods of impulsive
acting-out followed by worry and anxiety regarding the consequences of their impul-
sive behavior. They may view themselves as incapable of controlling their reactions
to stressful circumstances; however, this pattern of impulsivity will tend to recur and
lead others to doubt the sincerity of their concern and desire to alter their behavior.
This pattern was never observed in the clinical standardization sample.

87
 PAI Interpretive Guide

ANX-ALC
This configuration of the clinical scales suggests a person with a history of
drinking problems who is experiencing prominent anxiety. The anxiety and alco-
hol use may be related in a number of different ways; for example, alcohol use may
be serving a functional role of tension reduction. The person is also likely to be
anxious and guilty about the impairment in social role performance that has
resulted from drinking; the alcohol use is probably causing severe disruptions in
relationships and work, with these difficulties serving as additional sources of
stress and, perhaps, further aggravating the drinking problems. Secondary eleva-
tions on STR are often observed with this codetype, further supporting the possi-
bility that alcohol is serving a stress-reduction function. This profile pattern,
observed in 1.0% of clinical respondents, is associated with diagnoses of Alcohol
Dependence, Major Depressive Disorder, and Dysthymic Disorder.

ANX-DRG
This clinical scale configuration suggests a person with a history of substance-
abuse problems who is experiencing prominent anxiety. This anxiety and the sub-
stance use may be related in a number of different ways: for example, the drug use
may be serving a functional role of tension reduction, or the impairments associ-
ated with the drug use may be heightening subjective distress. Such people tend
to be anxious and guilty about these impairments in social-role performance,
including relationships and work; such difficulties serve as additional sources of
stress and, perhaps, further aggravate the tendency to abuse drugs. Secondary ele-
vations on DEP and SUI are often observed with this codetype. Also, it is not
uncommon to see RXR in a range that suggests limited motivation for treatment,
perhaps associated with a reliance on drugs to solve the individual’s problems. This
profile configuration, observed in 1.0% of clinical respondents, is actually rela-
tively uncommon in substance-abusing samples, but it is seen with some fre-
quency in individuals who have psychotic symptoms.

ARD-DEP
This configuration of the clinical scales suggests a person with significant ten-
sion, unhappiness, and pessimism. Although such people are quite distressed and
acutely aware of their need for help, their low energy level, tension, and with-
drawal may make them difficult to engage in treatment. Various stressors, both
past and present, have adversely affected self-esteem, and they tend to view them-
selves as ineffectual and powerless to change their life direction. The life disrup-
tions have left them feeling uncertain about goals and priorities, and tense and

88
 Two-Point Codetypes in Profile Interpretation

pessimistic about what the future may hold. They are likely to have difficulties in
concentrating and in making decisions, and the combination of hopelessness,
anxiety, and stress apparent in these scores places a person at increased risk for
self-harm. Secondary elevations on ANX and SUI are often observed with this
codetype. This is a relatively common profile, observed in 2.5% of clinical respon-
dents. Common diagnostic correlates include posttraumatic stress and other anx-
iety disorders, major depression and dysthymic disorder, borderline personality,
and schizoaffective disorder.

ARD-MAN
This combination of the clinical scales is quite unusual. It suggests a person
who is fearful, irritable, and affectively labile. Such people may see themselves as
overextended and vulnerable, with goals and expectations that are beyond their
capacity, leaving them stretched thin and hindering their ability to perform any
roles effectively. The resulting strain has probably left the respondent tense and
feeling overwhelmed by these demands. Close relationships may have suffered
particular strain from the moody and often demanding presentation of the respon-
dent. This profile pattern is quite rare, and it was never observed in the clinical
standardization sample.

ARD-PAR
This is an unusual configuration of the clinical scales; it suggests a person with
prominent hostility and suspiciousness who is acutely tense, fearful, and hyper-
sensitive. The respondent’s heightened sensitivity in social interactions probably
serves as a formidable obstacle to the development of close relationships, and
those relationships that are established are probably a source of ruminative worry.
Although the pattern hints at considerable anger and resentment, the degree of
anxiety concerning social interaction may lessen the likelihood that this anger is
directly expressed. Secondary elevations on BOR are often observed with this code-
type, as are elevations on NIM, raising the quesiion of symptom exaggeration. This
is a rare profile, observed in only 0.1% of clinical respondents.

ARD-SCZ
This configuration of the clinical scales suggests a person with significant
problems in thinking and concentration, accompanied by prominent distress and
ruminative worry. Such respondents are likely to be withdrawn and isolated, feel-
ing estranged from others. As a result, they probably have few, if any, close inter-
personal relationships, and they tend to become quite anxious and threatened by

89
PAI Interpretive Guide

such relationships. Their social judgment tends to be fairly poor, and they are often
confused about their goals and pessimistic about what the future may hold. Sec-
ondary elevations on SUI are often observed with this codetype, and the combi-
nation of marked anxiety and clouded judgment heightens concerns in this area.
This profile is seen in 0.4% of clinical respondents, with diagnoses of Posttrau-
matic Stress Disorder and Bipolar Disorder disproportionately represented.

ARD-BOR
This pattern suggests a person who is uncomfortable, impulsive, angry, and
resentful. People with this type of profile often are presenting in a state of crisis
and marked distress. Such crises often are associated with difficulties or rejection
(perceived or actual) in interpersonal relationships; these respondents often feel
betrayed or abandoned by others who are close to them. This may be part of a
more general pattern of anxious ambivalence in close relationships, marked by bit-
terness and resentment, on the one hand, and by dependency and anxiety about
possible rejection on the other. Various stressors, both past and present, may have
both contributed to and maintained this pattern of interpersonal turmoil; ARD-T
and STR may yield information about the relative importance of recent, as opposed
to more distant, stressors. Regardless of the temporal progression, the disruptions
in their lives leave such people uncertain about goals and priorities, and tense and
cynical about future prospects. Secondary elevations on DEP and SUI are often
observed with this codetype, which is obtained in 1.6% of clinical respondents.
Common diagnostic correlates include borderline personality, major depression,
and dysthymic disorder.

ARD-ANT
This configuration of the clinical scales is unusual. It suggests a person who is
anxious, tense, and ruminative, combined with impulsivity and the potential for
acting-out behaviors. Such people are likely to fluctuate between these seemingly
disparate personality elements, with periods of impulsive acts followed by worry
and rumination regarding the consequences of their behavior. They may view
themselves as victims of their impulsivity, incapable of controlling their reactions
to stressful circumstances. However, this pattern of impulsivity will tend to recur,
and it may lead others to conclude that they are hostile and to doubt the sincerity
of any expressed concern and desire to alter their behavior. Secondary elevations
on MAN and PAR are often observed with this codetype. This is a rare profile,
observed in 0.1% of clinical respondents.

90
 Two-Point Codetypes in Profile Interpretation

ARD-ALC
This configuration of the clinical scales suggests a person with a history of
drinking problems who is experiencing prominent stress and anxiety. The anxiety
and alcohol use may be related in a number of different ways. Alcohol use could
be serving a functional role of reducing tension, as it may be seen as relieving the
impact of stressors past and present. The respondent is likely to ruminate about
life circumstances, and the urge to drink may be at the center of many of these
ruminations. Such individuals are likely to be quite anxious and guilty about the
impairment in social-role performance that has resulted from their drinking: the
alcohol use is probably causing severe disruptions in relationships and work, with
these difficulties serving as additional sources of stress and perhaps further aggra-
vating the drinking problems. Secondary elevations on BOR and DEP are com-
monly seen with this codetype. This profile, observed in 0.6% of clinical respon-
dents, is most commonly associated with a diagnosis of Alcohol Abuse or Alcohol
Dependence.

ARD-DRG
This configuration of the clinical scales suggests a person with a history of sub-
stance abuse problems who is experiencing prominent stress and anxiety. The anx-
iety and drug use may be related in a number of different ways: for example, the
use of drugs could be serving a functional role of reducing tension, and the per-
son may use it to relieve the impact of past and present stressors. Such people tend
to ruminate about life circumstances, and the desire and craving for drugs may be
at the center of these ruminations. They tend to be quite anxious and guilty about
the impairment in social-role performance that has resulted from the substance
abuse; the drug use is probably causing severe disruptions in their relationships
and work, with these difficulties serving as additional sources of stress and, per-
haps, further aggravating the drug problems. Secondary elevations on BOR and
ANT are often observed with this codetype, suggesting characterological problems.
This is a relatively uncommon profile, observed in 0.2% of clinical respondents.
Common diagnostic correlates include substance abuse, major depression, and
acting-out personality disorders (e.g., borderline and antisocial).

DEP-MAN
This configuration of the clinical scales is quite unusual, as it suggests a person
who is experiencing severe distress, irritability, and unhappiness, as well as periods
of heightened activity and energy. Typically, people with this pattern fluctuate

91
PAI Interpretive Guide

between these disparate mood states, with periods of dysphoria followed by

episodes of elevated mood. Hypomanic features may be masking a severe depres-
sion that may not be apparent from overt behaviors. Such people tend to be rather
preoccupied and self-absorbed by their mood state and to be regarded by others
as irritable and, perhaps, self-centered. Even individuals with marked mood
swings tend not to obtain this profile; instead, whatever mood state they are cur-
rently experiencing tends to dominate the profile configuration. This pattern was
never observed in the clinical standardization sample.

DEP-PAR
This configuration of the clinical scales suggests a person with prominent
depression and hostility. People displaying such a pattern may be experiencing an
embittered pessimism, attributing many of the negative circumstances occurring in
their lives to the shortcomings of others, and they see little hope that they can
change these circumstances. Their heightened sensitivity in social interactions has
most likely led to significant withdrawal, and it probably serves as a formidable
obstacle to the development of close and trusting relationships. Although such
people appear to harbor considerable anger and resentment, this anger is as much
directed at themselves as it is directed at others. Profiles with this configuration
often have a number of secondary elevations, both on the neurotic side of the spec-
trum (e.g., ANX, SOM, ARD) and on the more impaired side (e.g., SCZ, BOR, SUI).
Elevations on NON are also observed with this codetype, and this underscores the
degree of resentment directed at significant others. This profile was observed in
0.9% of clinical respondents. Common diagnostic correlates include schizoaffec-
tive disorder, major depression, and posttraumatic stress disorder.

DEP-SCZ
This configuration of the clinical scales suggests a person with significant prob-
lems in thinking and concentration problems, accompanied by prominent distress
and dysphoria. Such people are likely to be quite withdrawn and isolated, feeling
estranged from the people around them. These current difficulties have probably
placed a strain on the few close interpersonal relationships that the person does
have. Such people see little hope that their circumstances will improve to any
significant degree, and this hopelessness and pessimism, combined with the
likelihood of impaired judgment, may place them at increased risk for self-harm:
secondary elevations on SUI are often observed with this codetype. This config-
uration is reasonably common in clinical settings and was observed in 2.4% of
respondents in the clinical standardization sample. Common diagnostic correlates

92
 Two-Point Codetypes in Profile Interpretation

include schizoaffective disorder, posttraumatic stress or other severe anxiety

dis-
orders, borderline personality, major depression, and schizophrenia.

DEP-BOR
This pattern suggests a person who is unhappy, emotionally labile, and proba-
bly quite angry at some level. Clients with such profiles are typically presenting in
a state of crisis with marked distress and depression. The current distress may be
associated with difficulties or rejection, perceived or actual, in interpersonal rela-
tionships. Individuals with such profiles often feel betrayed or abandoned by those
close to them, and this compounds their feelings of helplessness and hopelessness.
For the respondent, this may be part of a more general pattern of anxious ambiva-
lence in close relationships, marked by bitterness and resentment on the one hand,
and by dependency and anxiety about possible rejection on the other. The under-
lying anger can cause such people to lash out impulsively at those closest to them:
however, the anger seems as much self-directed as it is directed at others. Life dis-
ruptions leave these individuals quite uncertain and ambivalent about goals and
priorities and tense and pessimistic about what the future may hold. The combi-
nation of hopelessness, resentment, and impulsivity may place such people at
increased risk for self-harm, and SUI and STR are typically elevated with this code-
type. This is a relatively common profile, observed in 2.5% of clinical respondents.
Common diagnostic correlates include borderline personality, major depression,
and adjustment disorders.

DEP-ANT
This configuration of the clinical scales is rather unusual. It suggests a person
who is dysphoric and pessimistic, combined with impulsivity and the potential for
acting-out behaviors. Such people are likely to fluctuate between these seemingly
disparate personality elements, with periods of impulsive acts followed by worry
and guilt regarding the consequences of their behavior. They may see themselves
as incapable of controlling their acting-out behavior, viewing it as a reaction to
stressful external circumstances. However, this pattern of impulsivity will tend to
recur, and it may lead others to view them as hostile and unreliable and to doubt
the sincerity of any displayed remorse or stated desire to alter their behavior. Typ-
ically, this configuration has few other neurotic elements, suggesting that most of
the distress arises from external, rather than internal, precipitants; secondary ele-
vations on PAR are also common, indicating an outward projection of blame for
current sources of stress. This is an unusual profile type, observed in only 0.2% of
clinical respondents.

93
 PAI Interpretive Guide

DEP-ALC
This configuration of the clinical scales suggests a person with a history of
drinking problems who is quite unhappy and pessimistic. For such individuals,
alcohol problems probably have led to severe impairment in the ability to main-
tain their social-role expectations, and this behavior has most likely alienated
many of the people who were once central in their lives. Such setbacks have prob-
ably led to significant guilt and rumination about their life circumstances, and the
urge to drink may be at the center of many of these ruminations. The depression
and the alcohol use may be related in a number of different ways: for example, the
depression could be driving the alcohol use, or it could be a consequence of the
social disruption associated with alcohol use. Regardless of whether the depression
is primary or secondary, the respondent may well be desperate for help, but cyni-
cal about the prospects for change or improvement. Secondary elevations on SUI
are often observed with this codetype and, when present, heighten concerns about
the possibility of self-harm, given the potential for disinhibition associated with
alcohol use. Observed in 1.7% of clinical respondents, this configuration is com-
monly associated with diagnoses of Alcohol Dependence, Major Depressive Disor-
der, and Posttraumatic Stress Disorder.

DEP-DRG
This configuration of the clinical scales suggests a person with a history of
substance-abuse problems who is quite unhappy and pessimistic. The drug use
has probably led to severe impairment in the ability to maintain social-role expec-
tations concerning relationships and employment, and the drug-related behaviors
have likely alienated many of the people who were once close to the respondent.
The configuration indicates significant guilt and distress about current life cir-
cumstances. The depression and drug use may be related in a number of different
ways: for example, the depression could be driving the use of drugs, or it could be
a consequence of the disruption associated with substance abuse. Regardless of
whether the depression is primary or secondary, it has probably left the person quite
pessimistic about the prospects for change or improvement. Secondary elevations
on SUI are often observed with this codetype, and this should be monitored closely,
given the potential for disinhibition associated with drug misuse. This configura-
tion is found in 1.0% of clinical respondents, and is commonly associated with
diagnoses of Borderline Personality, Major Depressive Disorder, and disorders
involving drug abuse and/or dependence.

94
 Two-Point Codetypes in Profile Interpretation

MAN-PAR
This combination of clinical scales is rather rare. It suggests a person with
expansive mood and heightened activity accompanied by prominent hostility and
irritability. Such people tend to see themselves as having had their plans thwarted
by the neglect or obstruction of others; however, they are probably more impeded
by an activity level that includes self-expectations that are beyond their actual
capacity. The sensitivity in social interactions and perhaps unrealistic self-appraisal
probably serve as formidable obstacles to the development of close relationships,
and those close relationships that are established have most likely suffered parti-
cular strain from their moody and often demanding presentation. The combina-
tion of impulsivity, resentment, and high energy levels could cause such individ-
uals to lash out impulsively at those whom they feel have slighted them in some
way. Secondary elevations on NON are often observed with this codetype, perhaps
pointing to the potentially unrealistic demands that such individuals place on oth-
ers. This is an uncommon profile, observed in only 0.1% of clinical respondents.

MAN-SCZ
This configuration of the clinical scales suggests a person with significant prob-
lems in thinking and concentration, accompanied by heightened activity levels and
irritable and expansive mood. Such people are likely to be agitated and confused,
feeling both irritated with and estranged from the people around them. Their
social judgment is probably quite poor, and those close relationships that have
been maintained are probably strained by their moody, disorganized, and often
demanding style of relating to others. Secondary elevations on STR and, occasion-
ally, SOM may be seen with this codetype. This is a relatively uncommon profile
pattern, observed in only 0.1% of clinical respondents.

MAN-BOR
This pattern on the clinical scales suggests a person with very labile mood,
impulsivity, and heightened activity levels, accompanied by prominent hostility
and irritability. For such people, interpersonal relationships are likely to be fairly
stormy, and even those close relationships that have been maintained have most
likely suffered particular strain from their moody, unpredictable, and often
demanding presentation. The combination of impulsivity, resentment, and high
energy levels could cause such people to overreact to minor events and to be at
risk for lashing out impulsively at those closest to them. These same traits also

95
PAI Interpretive Guide

place the respondent at increased risk for acting-out behaviors. Secondary eleva-
tions on NON are often observed with this codetype; elevations on DOM can reveal
very strong needs for control in relationships that are seldom met. Observed in
0.5% of clinical respondents, this configuration appears to be more strongly
related to bipolar disorder than to borderline personality.

MAN-ANT
This configuration of the clinical scales suggests a person who is impulsive,
hostile, impatient, and unempathic. Interpersonal relationships are likely to be
fairly stormy and rather short-lived; even those close relationships that are main-
tained most likely will have suffered strain from their hostile, self-centered, and
often demanding presentation. The combination of impulsivity, resentment, and
high energy levels could cause such people to have little consideration for the
needs of others and to lash out impulsively at those around them when crossed.
These same traits also place such people at increased risk for acting-out behaviors,
and it is likely that these behaviors have led to impairment in their ability to main-
tain social-role expectations in both formal (e.g., work) and informal settings. Sec-
ondary elevations on STR are common; elevations on AGG also are frequent and,
when present in the context of this codetype, represent a particular concern. This
is not a particularly common configuration, seen in 0.4% of clinical respondents,
but observed with some frequency in individuals with bipolar disorder, antisocial
personality, and drug abuse or dependence.

MAN-ALC
This configuration of the clinical scales suggests a person with a history of
drinking problems who is emotionally labile and impulsive. The alcohol problems
have likely led to fairly severe impairment in their ability to maintain social role
expectations, and their general recklessness has probably alienated their friends
and family. Such people are likely to be particularly disinhibited under the influ-
ence of alcohol and to display exceptionally poor judgment and demonstrate other
acting-out behaviors while intoxicated. It is also unlikely that there is much last-
ing remorse associated with any such behaviors. Secondary elevations on STR and
AGG often are observed with this codetype. This is a relatively rare profile,
observed in only 0.1% of clinical respondents, and it is unusual to find this pro-
file in the absence of alcohol dependence.

96
 Two-Point Codetypes in Profile Interpretation

MAN-DRG
This clinical scale configuration suggests an emotionally labile and impulsive
person with a history of substance abuse problems. The drug problems probably
have led to fairly severe impairment in the ability to maintain social role expecta-
tions, and their generally reckless approach to life has probably alienated many of
the people around them. They are likely to be particularly disinhibited under the
influence of drugs and may display markedly poor judgment and demonstrate
other acting-out behaviors while intoxicated. Secondary elevations on ANT and
NON are often observed with this codetype, indicating the externalization of blame
for the respondent’ current difficulties. This profile type is uncommon, observed in
only 0.1% of clinical respondents.

PAR-SCZ
This configuration of the clinical scales suggests a person with significant prob-
lems in thinking and concentration, accompanied by prominent hostility, resent-
ment, and suspiciousness. Sensitivity in social interactions probably serves as a
formidable obstacle to the development of close relationships, and, thus, they are
likely to be cautious, withdrawn, and isolated, feeling both estranged from and
mistreated by the people around them. Their judgment is probably fairly poor, and
they are likely to be chronically tense and apprehensive about what the future may
hold. If such a person presents for treatment, establishing a therapeutic relation-
ship may be challenging because such people tend to become quite anxious and
threatened by the offer of a close interpersonal relationship. Secondary elevations
on NIM are often observed with this codetype, raising the possibility that symp-
tom exaggeration may be driving up the scores on PAR and SCZ. This is a relatively
common profile in inpatient settings, observed in 2.4% of respondents in the clin-
ical standardization sample. Common diagnostic correlates include schizophrenia,
schizoaffective disorder, antisocial personality, and drug dependence.

PAR-BOR
This configuration of the clinical scales represents a person who is angry,
resentful, impulsive, and emotionally labile. Such people are likely to be
extremely sensitive in social interactions and very quick to perceive rejection, real
or imagined, by others; they are likely to feel that they are repeatedly betrayed by
those close to them. This is likely to be part of a more general pattern of chronic

97
PAI Interpretive Guide

maladjustment in social relationships, marked by anxious ambivalence between

bitterness and resentment, on the one hand, and by dependency and fear of pos-
sible rejection on the other. The bitterness is likely to surface readily, and they
may react impulsively when they feel others have slighted them in some way. The
combination of impulsivity, anger, and dysphoria could place the respondent at
increased risk for self-harm or acting-out behaviors. Secondary elevations on AGG
and DEP are often observed with this codetype, and, when present, may indicate
that angry outbursts tend to be followed by rumination over the interpersonal con-
sequences of these outbursts. This profile, observed in 0.4% of clinical respon-
dents, is more typically observed in antisocial personality disorder than in border-
line personality.

PAR-ANT
This configuration of the clinical scales suggests a person who is impulsive,
hostile, bitter, and unempathic. Interpersonal relationships are likely to be short-
lived and to be characterized by marked conflict; even those close relationships
that are maintained most likely will have suffered strain from an irritable and seli-
centered style. The combination of impulsivity, egocentricity, and anger could
cause such individuals to lash out angrily at those who are perceived as having
impeded them in some way. These same traits also place them at increased risk for
acting-out behaviors, and it is likely that these behaviors have led to marked
impairment in the respondent's capacity to work effectively with others. Secondary
elevations on AGG and SUI often are observed with this codetype, and these ele-
vations heighten concerns about managing these individuals in treatment. This
profile is seen in 0.4% of clinical respondents and is found with relative frequency
among individuals with a diagnosis of Antisocial Personality Disorder.

PAR-ALC
This configuration of the clinical scales suggests a person with a history of
drinking problems who is embittered and angry. A general sensitivity and hostility
in social interactions probably serve as formidable obstacles to the development of
close relationships, and, thus, such people are likely to be withdrawn and isolated.
Alcohol may be playing a functional role in helping them withdraw from such rela-
tionships or in reducing the anxiety and threat that they pose. The respondents are
likely to ruminate about their life circumstances, but they may not fully acknowl-
edge the severity of their drinking problems. It is likely that there is significant
impairment in social-role performance that has resulted from drinking; however,

98
 Two-Point Codetypes in Profile Interpretation

such people are more likely to attribute such problems to external factors than to
admit the connection to their drinking. This is a relatively rare profile configura-
tion, observed in 0.2% of clinical respondents, and it is most commonly seen with
diagnoses of drug dependence or Antisocial Personality Disorder, or both.

PAR-DRG
This pattern on the clinical scales suggests a person with a history of
substance-abuse problems who is embittered, suspicious, and angry. Sensitivity
and hostility in social interactions probably serves as a formidable obstacle to the
development of close relationships, and, thus, such individuals are likely to be with-
drawn and isolated. The drugs may be playing a functional role in helping them
withdraw from such relationships or in reducing the anxiety and threat that they
pose, but the drug use may also be contributing to the suspicion and mistrust with
which they view others. The respondent is likely to ruminate about life circum-
stances, and the urge and craving for drugs may be at the center of many of these
ruminations. It is likely that there is significant impairment in social role perfor-
mance that has resulted from the substance abuse; however, the respondent is more
likely to attribute such problems to external factors than to admit the connection to
drug use. Secondary elevations on BOR and STR are often observed with this code-
type. This pattern is found in 0.3% of clinical patients and is most commonly asso-
ciated with diagnoses involving abuse and/or dependence on substances.

SCZ-BOR
This configuration of the clinical scales suggests a person who is confused,
emotionally labile, and angry. The respondent is reporting marked interpersonal
dysfunction and significant problems in thinking and concentration; it is possible
that bitterness and constant preoccupations with relationships impairs their abil-
ity to think clearly. Individuals with this profile are typically presenting in a state
of crisis and marked distress, often related to interpersonal disruption. During
such crisis periods, the respondents may experience brief episodes during which
judgment and reality testing deteriorates markedly. Because of their unhappiness,
resentment, impulsivity, and poor judgment, these individuals may be at increased
risk for self-harm or acting-out behaviors. Secondary elevations on DEP and AGG
are often observed with this codetype, and the relative positioning of these latter
scales may reveal whether anger will be directed outward or inward. This profile,
observed in 0.6% of clinical respondents, is particularly associated with diagnoses
of Bipolar Disorder and Posttraumatic Stress Disorder.

99
PAI Interpretive Guide

SCZ-ANT
This combination of clinical scales is quite unusual. It suggests a person with
significant problems in thinking and concentration, accompanied by impulsivity
and the potential for acting-out behaviors. Given such a pattern, social judgment
is probably quite poor, and those few social relationships that have been main-
tained are probably strained by an unempathic and self-centered approach. The
combination of impulsivity and poor judgment contributes to a propensity for
antisocial behaviors, and such people may view themselves as incapable of con-
trolling these behaviors, seeing them as reactions to external circumstances. This
is a rare configuration, and none of the patients in the clinical standardization sam-
ple demonstrated this pattern.

SCZ-ALC
This configuration of the clinical scales suggests a person with a history of
drinking problems who is confused and socially isolated. A general discomfort in
social interactions probably serves as a formidable obstacle to the development of
close relationships, and, thus, such people are likely to be withdrawn and isolated
and to feel estranged from the people around them. Alcohol may be playing a func-
tional role in helping them distance themselves from such relationships or in
reducing the anxiety and threat posed by such relationships. Their judgment is
probably fairly poor, and they are generally both apprehensive about what the
future may hold and cynical about the prospects for change. Secondary elevations
on NIM are often observed with this codetype, raising the possibility that symp-
tom exaggeration may be contributing to the SCZ elevation. This codetype is not
uncommon, as it is seen in 1.0% of clinical respondents. Alcohol dependence is
the most common diagnostic correlate of this pattern.

SCZ-DRG
This clinical scale configuration suggests a person with a history of substance-
abuse problems who is confused and socially isolated: Discomfort in social inter-
actions probably impedes the development of close relationships, and, thus, such
individuals are likely to be withdrawn and isolated and to feel estranged from the
people around them. Drugs may be playing a functional role in helping them with-
draw from such relationships or in reducing the anxiety and threat posed by such
relationships, but the mistrust and exploitativeness that characterizes this lifestyle
is likely to simply exacerbate such problems. In most areas, their judgment is
probably fairly poor, and they are likely to be pessimistic and cynical about long-
term plans for change. Secondary elevations on BOR and SUI are often observed

100
 Iwo-Point Codetypes in Profile Interpretation

with this codetype and, when present, heighten concerns about the individual’s
capacity for self-destruction. This profile, observed in 0.6% of clinical respon-
dents, is seen with relative frequency in individuals with drug dependence, as well
as with antisocial personality.

BOR-ANT
This configuration of the clinical scales suggests a person who is impulsive,
emotionally labile, and unempathic. For such people, interpersonal relationships
are likely to be short-lived and to be characterized by marked conflict; even those
close relationships that have been maintained most likely will have suffered strain
from their hostile, self-centered, and perhaps manipulative style. The combination
of impulsivity, egocentricity, and anger could cause them to lash out at those whom
they feel have slighted them in some way. These same traits also place them at
increased risk for acting-out behaviors, and it is likely that these behaviors have
led to severe impairment in their ability to maintain employment. They may view
themselves as incapable of controlling such acting-out behavior, seeing themselves
as victims of unfair and stressful circumstances. However, this pattern of impul-
sivity will tend to be recurrent, leading others to view them as irresponsible and
unreliable and to doubt the sincerity of any remorse or desire to alter their behav-
ior. Secondary elevations on AGG and SUI are often observed with this codetype
and, when present, may point to a worrisome state intensification of the charactero-
logical issues. This profile, observed in 0.9% of clinical respondents, is particularly
common in borderline personality, but also is seen with antisocial personality.

BOR-ALC
This configuration of the clinical scales suggests a person with a history of
drinking problems who is impulsive and affectively labile. Drinking may be part
of a more general pattern of self-destructive behavior. Interpersonal relationships
are likely to be volatile and to be characterized by marked conflict; even those
close relationships that have been maintained will have suffered some strain from
an impulsive, unpredictable, and probably hostile style of interaction. These rela-
tionships likely will have deteriorated even further as a consequence of the drink-
ing. Such people are likely to be particularly disinhibited under the influence of
alcohol, and they may display remarkably poor judgment and demonstrate other
acting-out behaviors while intoxicated, perhaps blaming the alcohol for their own
unacceptable behavior. Secondary elevations on STR and low scores on RXR are
often observed with this codetype and, when present, suggest a desperate recog-
nition of the need for help. This is a relatively common profile, observed in 1.3%

101
PAI Interpretive Guide

of clinical respondents. Common diagnostic correlates include alcohol dependence,

and antisocial as well as borderline personality.

BOR-DRG
This configuration of the clinical scales suggests a person with a history of
substance-abuse problems who is impulsive and affectively labile. The drug use
may be part of a more general pattern of self-destructive behavior, and it probably
exacerbates an already erratic approach to life. Interpersonal relationships are
likely to be volatile and to be characterized by marked conflict; even those close
relationships that have been maintained will have suffered some strain from the
unpredictable and hostile style of interaction. These relationships likely will have
deteriorated even further as a consequence of the drug abuse. Such people are
likely to be particularly disinhibited under the influence of drugs, and they will
tend to display particularly poor judgment and to demonstrate other acting-out
behaviors while intoxicated. Secondary elevations on AGG are often observed with
this codetype. This profile, observed in 1.1% of clinical respondents, is most com-
monly associated with drug abuse and/or dependence diagnoses and Borderline
Personality.

ANT-ALC
This configuration of the clinical scales suggests a person with a history of
acting-out behavior, most notably in the area of alcohol abuse, but probably
involving other behaviors as well. The impulsivity and drinking problems likely
have led to severe impairment in the ability to maintain social-role expectations,
and their reckless approach to life has probably alienated most of the people who
were once close to them. Generally impulsive and thrill-seeking, the alcohol use
probably further impairs their already suspect judgment. Interpersonal relation-
ships are likely to be volatile and short-lived; even those relationships that have
been maintained will have suffered some strain from the egocentricity and from
the consequences of drinking. Secondary elevations on DRG often are observed
with this codetype, and very low raw scores on this scale may reflect denial. This
profile, observed in 0.7% of clinical respondents, is observed frequently among
polysubstance abusers.

ANT-DRG
This configuration of the clinical scales suggests a person with a history of
acting-out behavior, most notably in the area of substance abuse, but probably
involving other behaviors as well. Impulsivity and drug use have likely led to

102
 Iwo-Point Codetypes in Profile Interpretation

severe impairment in their ability to maintain stable employment; their reckless-

ness has probably alienated most of their family and friends. Generally impulsive
and thrill-seeking, the use of drugs is likely to further impair their already suspect
judgment. Interpersonal relationships are likely to be superficial, volatile, and
short-lived; even those relationships that have been maintained will have suffered
some strain from the respondents’ egocentricity and from the consequences of
their drug use. Secondary elevations on AGG are often observed with this codetype
and, when present, suggest one possible result of the disinhibition associated with
drug use. This is a relatively common profile, observed in 2.1% of clinical respon-
dents. This pattern is common in groups with drug dependence diagnoses, Bor-
derline Personality, or both.

ALC-DRG
This pattern on the clinical scales suggests a person with a history of polysub-
stance abuse, including alcohol as well as other drugs. When disinhibited by the
substance use, other acting-out behaviors may become apparent as well. The
substance abuse is probably causing severe disruptions in social relationships and
work performance, with these difficulties serving as additional sources of stress
and, perhaps, further aggravating the tendency to drink and use drugs. Secondary
elevations on STR are often observed with this codetype. This profile pattern is
quite common, observed in 9.0% of clinical respondents, and it characterizes
roughly one fourth of the individuals in alcohol or drug treatment. Other common
diagnostic correlates include antisocial personality and bipolar disorder.

103
 ‘f

, fi

i ' \.

iii q
lw 9 701 te &

¥ ; Dey .
tr asic © laptrits is siiaty eet wit
ms $j Ths 3 edtew te JG?fiolp nko cute vor
’ sats moe aN / Oren UUeeeTrT9finy Loken
Otte syideciedst lee 1 eee ate dene racy leek tad aapeeatio
CANE Sy ANG iwinleobs, oy te asitioralilth seks dstw SIS BW -

Punbatrioae emirahssear Larne iit or nobis adi grit egaycadian) 2qeiag, ae ee

shareuine cling aT sqeehos adpapiethgnceda ovis sieN02 no Sonera e
eetias Mails Lag nel dt, Inotustfp.0 Ds8 fede '
eS A age “ih
logingyiame ae Aatopalaing at

ee an) a a } Gmert ieee eiptnrts =

7 * 7 rage,we a a - m

die
 CHAPTER 4
NEGATIVE DISTORTION: RANDOM
RESPONDING AND MALINGERING

One of the difficulties that have beset the field of psychological assessment
since its inception concerns the accuracy of self-reported information as an indi-
cation of psychological status. Myriad reasons have been offered as to why self-
report might be distorted. One source of distortion may arise from an intention to
deceive the recipient of the information; such examinees may attempt to distort
their responses in order to appear either better adjusted or more poorly adjusted
than is actually the case. A second source of distortion may arise from limited
insight or self-deception. Such examinees may genuinely believe that they are
doing quite well or quite poorly, but this belief might be at odds with the impres-
sions of objective observers. A third source of distortion might arise from careless-
ness or indifference in taking a test; examinees who answer questions with little
reflection (or even randomly) may yield results that do not accurately mirror their
experiences.

Such concerns have led many test developers to create scales that provide mea-
sures of these sources of distortion. The PAI offers four validity scales that are
designed to provide an assessment of factors that could distort the results of a self-
report questionnaire, as well as indices constructed to supplement these scales.
Elevated scores on any of these scales suggests that other scales should be viewed
with caution and that any interpretation of results should be tentative. In general,
if a subject obtains a score that is more than 2 standard deviations above the mean
of the representative clinical sample on any of these scales or indices, the profile is
likely to be seriously distorted by some test-taking response style. This result casts
serious doubt on all other information derived from the test, and, under such con-
ditions, the need to consider the PAI protocol in light of information derived from
other sources becomes particularly critical.
The specific distortions to be considered in this chapter involve those that
might lead the interpreter to draw a more negative conclusion from the data than
might otherwise be warranted. Two particular sources of such distortion will be
considered: first, test protocols that were completed carelessly, randomly, or

105
PAI Interpretive Guide

idiosyncratically by the respondent; and, second, the effect of efforts to malinger

or simulate mental disorder.

Detecting Careless or Idiosyncratic Responding

When clinical scales are elevated on a self-report test, several reasons for this
elevation must be considered. One possibility involves careless responding, where
the person answers the items more or less randomly because of confusion, disin-
terest, resistance, or clerical errors in test-taking or scoring. Because many items
on the PAI reflect severe psychopathology and have very low endorsement rates,
such individuals will obtain marked elevations (i.e., many of these items will be
endorsed). Two of the PAI validity scales were designed to identify this source of
distortion.

Infrequency (INF)
The INF scale is useful in the identification of people who complete the PAI in
an atypical way because of carelessness, confusion, reading difficulties, or other
sources of random responding. The scale consists of items that were designed to
be answered similarly by all respondents, regardless of clinical status; half of these
items are expected to be answered Totally False (e.g., “My favorite poet is Raymond
Kertezc.”), whereas the other half should be answered Very True (e.g., “Most peo-
ple would rather win than lose.”). The INF items are placed evenly throughout the
PAI to identify potentially problematic responding at any point within the test-
taking. There is no thematic connection between the content of different items on
the scale. The items were selected on the basis of very low endorsement frequen-
cies in both normal and clinical respondents; this contrasts with scales such as the
MMPI F scale, where items were selected on the basis of infrequency in the nor-
mative sample. Such scales often yield elevations in clinical samples. This is
because the item content is confounded with psychopathological symptoms that
are infrequent in normative samples, but may reflect valid responding in a clinical
respondent.
INF scale items were written to provide item content that would be infrequent,
yet would not sound bizarre (e.g., “I have never seen a building.”). For example,
one question asks if the respondent’ two favorite hobbies are archery and stamp
collecting. Interest in these hobbies actually turns out to be inversely related, and,
as such, there are few people for whom both these hobbies are primary interests.
However, the combination is not implausible; it is merely uncommon. Because

106
 Negative Distortion: Random Responding and Malingering

each individual item on the scale is uncommon, an individual who endorses

more
than a few of these items is a unique individual indeed.
The Infrequency scale is primarily a measure of carelessness in responding.
However, there is another potential element underlying INF elevations, reflecting
a tendency to answer the PAI items in a very idiosyncratic way. A quick inquiry
about INF items can easily distinguish between these two sources of elevation.
Individuals who respond idiosyncratically to the inventory will have an explana-
tion for their endorsement, albeit an explanation that suggests that the test items
may have been interpreted in an unusual way. Because such people are not
approaching the test in the way that most people do, the results of the self-report
test should not be interpreted as if they were. For example, one respondent who
obtained an INF score of 85T was questioned about his responses, and his com-
ments were revealing about the nature of the elevation. For the item “My favorite
sporting event on television is the high jump,” he answered Very True. He was
asked about the source of this interest and whether he had been a high-jumper in
high school. He responded that he didn’t think he had actually ever seen the high
jump on television, but he said, “Well, I really like to watch sports on TV, and I
think that is a big part of my personality, so I wanted to make sure that you knew
that, and I didn’t see any other opportunity to tell you that on here, so I was going
to tell you with that item.” This response reveals both that the respondent was try-
ing to answer the test honestly, and that he was probably not being careless. How-
ever, if a person makes idiosyncratic inferences about items that do not reflect the
actual content of the item, or if he or she begins to respond to items figuratively
rather than literally, the results will not be interpretable in any straightforward way.
The distribution of INF is similar for both normal and clinical respondents;
both distributions are quite dissimilar from the distribution derived by simulating
random responding. Generally, low scores (i.e., < 60T) suggest that the respondent
did attend appropriately to item content in responding to the PAI items. Moderate
elevations (i.e., 60T to 75T) indicate some unusual responses to INF items, and, at
the higher end of this range, one should consider potential sources such as read-
ing difficulties, random responding, confusion, errors in scoring, idiosyncratic
item interpretation, or failure to follow the test instructions. Any interpretive
hypotheses based on the PAI should be reviewed with caution if INF is in this
range, and some inquiry about INF responses would be useful before clinical scale
results are interpreted.
High scores on INF (i.e., 2 75T) suggest that the respondent did not attend
appropriately to item content in responding to the PAI items; a completely random

107
PAI Interpretive Guide

completion of the PAI would result in an average INF score of 86T. There are
several potential reasons for scores in this range, including reading difficulties, ran-
dom responding, confusion, errors in scoring, or failure to follow the test instruc-
tions. Regardless of the cause, however, the test results are best assumed to be
invalid and no clinical interpretation of the PAI is recommended. However, an
examination of specific INF items may yield useful information. For example, if the
endorsed INF items are all from the second half of the test, the subject may have
completed the initial half of the instrument appropriately and may have begun
responding haphazardly at a later point. In this instance, score estimates for most
PAI scales may be extrapolated from the responses to the first 160 items, as
described in chapter 11 of the PAI Professional Manual (Morey, 1991).

Inconsistency (ICN)
The ICN scale is an empirically derived scale that reflects the consistency with
which the respondent completed items with similar content. The scale is com-
prised of 10 pairs of items with related content; 5 of these pairs should be
answered similarly and 5 of the pairs are psychologically opposite. The items on
ICN were the pairs that were empirically found to be the most similar during the
development of the PAI. Although each pair of items is similar in content, the pairs
differ from one another; thus, the scale does not reflect any particular construct
other than response consistency.
Because ICN theoretically reflects measurement error, it tends to have low cor-
relations with most other measures. The largest correlation of ICN appears to be
with the Marlowe-Crowne (1957) Social Desirability scale (r = —.24). Although
low, this correlation is informative, as it suggests that people who tend to respond
in a socially desirable direction also tend to answer questions consistently. Thus, if
social desirability is considered to have any sort of an impression-management
component, a person trying to manage his or her self-presentation appears to do
so with some care. This suggests that ICN elevations are probably not the result of
efforts at impression management, although, at times, such scales are interpreted
in this manner (i.e., people who tell an inconsistent story are not telling the truth).
However, if someone is consciously trying to distort in a given direction, ICN often
does not elevate at all. Rather, ICN is much more likely to reflect carelessness or
confusion in responding.

One commonly observed problem that can cause elevations on ICN is a failure
to attend to negated items (i.e., item statements that contain the word not). Although
there are relatively few such items on the PAI, these items are overrepresented

108
 Negative Distortion: Random Responding and Malingering

on ICN to examine how such items were interpreted. Respondents who are not
attending closely may misinterpret the question that reads, “I have no trouble
falling asleep,” instead reading it as, “I have trouble falling asleep.” This pattern
alerts the interpreter that the respondent may not have been reading the items
carefully when completing the inventory.
The distribution of ICN is fairly similar for both normal and clinical respon-
dents, although clinical respondents tend to score slightly higher (i.e., respond
slightly less consistently) than normal respondents. The distributions from clinical
and normal respondents are quite dissimilar from that derived by simulating ran-
dom responding. Generally, low scores on ICN (ie., < 64T) suggest that the
respondent did respond consistently and probably attended appropriately to item
content in responding to the PAI items. Moderate elevations (i.e., 64T to 73T)
indicate some inconsistency in responses to similar items, which could arise from
a variety of sources ranging from carelessness or confusion to attempts at impres-
sion management. Interpretive hypotheses based on other PAI scales should be
reviewed with caution if ICN is in this range.
High scores on ICN (i.e., 2 737) suggest that the respondent did not attend
consistently or appropriately to item content in responding to the PAI items; a
completely random completion of the PAI would result in an average ICN score of
approximately 73T. There are several potential reasons for scores in this range,
including carelessness, reading difficulties, confusion, errors in scoring, or failure
to follow the test instructions. Regardless of the cause, however, the test results are
best assumed to be invalid, and no clinical interpretation of the PAI is recom-
mended when ICN scores are in this range.

Random Responding: Scale Configuration

The mean profile for a group of 1,000 protocols that were generated using
random-responding simulations is presented in Figure 4-1. The most prominent
characteristic of this profile is that both INF and ICN fall above the thresholds for
profile validity described in previous sections. This result is quite rare in actual
protocols; only 0.2% of respondents in both the community and clinical norma-
tive samples had both INF and ICN above the recommended cutoffs. The NIM
scale is also elevated in this profile, although not to the extent that occurs in malin-
gering simulation samples. In general, if both INF and NIM are elevated and the
scores are comparable (i.e., within 10T of one another), then random responding
is suggested. Malingered protocols tend to lead to profiles where NIM greatly
exceeds INF, typically by 20T or more.

109
PAI Interpretive Guide

1 2 3 4 5 6 ie 8 9 10 44 A B c D E y Zz
PROFILE FORM FOR ADULTS-SIDEA
a ] e—se Random
| response

w an

oOsaoog
t]
°

Porvalovveala

1 2 3 4 5 6 if 8 9 10 aa A B c D E Y z
ICN INF NIM PIM SOM = ANX ARD DEP = =MAN PAR sez BOR ANT ALC ORG AGG SUI STR NON RXR DOM WRM

PROFILE FORM FOR ADULTS - SIDE B

o—e Random
response

100 —

80 —

4
8 = 7)

2oe == §3
60 =

one

40 —

30 3 oO
=e oO
- -
o-
SOM-C SOM-S SOMH ANXC ANXA ANXP ARD-O ARD-P ARD-T DEP-C DEP-A DEP-P MANA MANG MAN PAR-H PARP PARR SCZP SCZS SCZT BORA BORI BORN BORS ANT-A ANT-E ANT-S AGG-A AGG-V AGGP
CONV SOMA HEAL COG AFF PHYS OBS PHOB TRAU COG AFF PHYS ACT GRND JIRRI HYPE PERS RSNT PSYC SOC THGT AFF ID NEG SELF ANT EGO STIM AGG VERB PHYS
SYMP ZATN CONC SYMP SYMP SYMP COMP IAS STRS SYMP SYMP SYMP LEVL JOSY BLTY VIG CUTN MENT EXP DET DIS INST PROB REL HARM BEH CEN SEEK ATT AGG AGG

Figure 4-1. PAI profile for 1,000 protocols (Morey, 1991) using random-responding simulations.

110
 Negative Distortion: Random Responding and Malingering

Because random responding will include many positive responses to unusual

items, the overall profile is elevated. However, these elevations are not as marked
as tends to be seen in malingered protocols; for example, fewer than half of the
subscales are elevated above 70T in the random response profile. Another note-
worthy feature of the random response profile is the lack of differentiation among
the clinical scales, with most falling between 65T and 75T. Because of the empha-
sis on discriminant validity in the construction of the PAI, a relatively “flat” profile
in the elevated range tends to be usual, as it involves clinical features that are not
commonly seen in the same person (e.g., anxiety and antisocial features, or
depression and increased self-esteem). In general, malingered profiles tend to be
more elevated than random profiles, because responses are consistently patho-
logical, rather than randomly either pathological or healthy. However, the malin-
gered profile also has sharper differentiations, with some scales (e.g., SCZ) likely
to be markedly elevated and others (e.g., MAN) to be influenced less consistently
when pathology is simulated.

Detecting Malingering
Negative Impression (NIM)
Generally, the starting point in the evaluation of malingering is the NIM scale,
although it must be emphasized that NIM is not a malingering scale. The NIM scale
was designed to alert the interpreter to the possibility that the test results may por-
tray a more negative impression of the individual than might otherwise be mer-
ited. In other words, the self-report of a high scorer on NIM is probably more
pathological than an objective observer would report. The items were selected on
the basis of low endorsement frequencies in both normal and clinical respondents,
although NIM items are endorsed with greater frequency in clinical patients than
in normal adults. Individuals with clear-cut and severe emotional problems can
and will obtain elevated scores on NIM, and more disturbed populations obtain
higher scores than those who are less impaired. For example, the mean for the out-
patient mental health patients in the PAI clinical normative sample on NIM was
59T (i.e., nearly 1 SD above the mean of the community sample), whereas the cor-
responding value for inpatients was 65T (i.e., 1.5 SD above the community mean).
If NIM is a measure of a response style, rather than a measure of psycho-
pathology, why should there be such a relationship between psychopathology and
NIM elevations? The answer lies in the association between certain forms of
mental disorder and the characteristic perceptual and cognitive features that can
lead to negative response styles. Several different types of mental disorders lead

111
PAI Interpretive Guide

individuals to perceive themselves, other people, or situations in a more negative

manner than might be warranted in the eyes of an objective observer. A depressed
patient views himself or herself as worthless, incompetent, and inadequate, whereas
others may view the patient as capable and highly effective. Interpersonal rela-
tionships that appear solid to others may be suspect in the mind of the paranoid
individual. A situation that may appear relatively benign to the clinician may be
perceived as an insurmountable crisis by the borderline patient. In these cases,
patients are likely to portray themselves or their circumstances, or both, in a more
negative manner than appears objectively warranted. However, these individuals
are not malingering; in fact, they do have significant, and perhaps quite severe,
forms of mental disorder. Nonetheless, in interpreting the PAI profile, the clinician
must be cognizant of the influence of these perceptual styles on the obtained pat-
tern of responses.
The NIM scale includes two types of items: some present an exaggerated or dis-
torted impression of the self and the present circumstances, and some represent
extremely bizarre and unlikely symptoms. Each of these tendencies may cause dis-
tortion of a self-report in a negative direction. Individuals who tend to exaggerate
the negative aspects of their lives can provide self-reports that appear quite patho-
logical. However, this tendency does not necessarily reflect malingering, which is
a conscious attempt to simulate psychopathology. Rather, this response style can
actually represent a prominent component of many psychopathological syndromes.
For example, depression, which is a prominent affect in the majority of clinical
respondents, lends a negative coloration to most events and circumstances. People
who are depressed often can find the grey cloud around every silver lining. A
depressed person may describe his or her childhood in very bleak terms, yet the
same person might report a happy childhood following the remission of the
depressive episode. In fact, both the DEP scale and the clinical diagnosis of major
depression are positively correlated with NIM. This does not mean that people
who are depressed are malingering. Rather, it may mean that, when such people
are asked questions about negative events in their lives, they tend to magnify the
negative, and, given the opportunity to describe their circumstances, they tend to
portray them very bleakly. Hence, it becomes an issue of discriminant validity,
However, this does not mean that NIM is a depression scale, either; depressed peo-
ple may score very high on NIM, whereas others can score quite low. The NIM
scale can reveal the degree to which this negative perceptual style is operating
within the profile of a particular depressed individual and how much distortion
may be resulting from this style. At times, there may be so much distortion that
the test results are rendered invalid, although this does not mean that the person
was malingering. Rather, the test is invalid in the sense that there are serious

112
 Negative Distortion: Random Responding and Malingering

distortions, and extreme caution must be exercised in interpreting the test results
at face value. Nonetheless, the result may accurately depict the way such individ-
uals feel about themselves and their circumstances.
A second example of the operation of cognitive style as an influence on self-
report may be found in borderline personality. Individuals with Borderline Per-
sonality Disorder tend to have an extremely negative evaluation of everything in
their environment that is not uniformly positive, whereas depression leads to neg-
ative self-evaluations. This tendency is sometimes referred to as splitting, where
self and others are divided into good and bad, with no middle ground; but the true
split is between what the individual perceives as uniformly positive and everything
else, which is evaluated in an intensely negative manner. Because self-report tests
such as the PAI repeatedly ask for an evaluation of present life circumstances, oper-
ation of this cognitive style can lead to distorted test results that portray events as
much more negative than they would seem to an objective observer.
It is true, however, that all patient groups score considerably lower on NIM
than research respondents instructed to simulate the responses of a mentally dis-
ordered patient, and, as such, the scale serves as a useful beginning point in the
detection of malingering. This is because another group of NIM items is more
closely related to malingering. These items were written to sound as if they repre-
sented pathological symptoms, but they were, in fact, extremely rare or nonexis-
tent in clinical populations. The item content is varied, but the items share the fea-
ture that they are dramatic sounding and play to stereotypes of mental disorder. In
fact, a few of the items are dissociative in nature, and it has been observed that
individuals with severe dissociative disorders sometimes obtain marked elevations
on NIM. Idiosyncratic responses to item content also can result in NIM elevations,
although in these instances INF also tends to be elevated. Regardless of the con-
text, some inquiry about the nature of positive responses to these NIM items is
merited.
Generally, low scores (i.e., < 73T) on NIM suggest that there is little distortion
in a negative direction on the clinical scales; the respondent probably did not
attempt to present a more negative impression than the clinical picture would war-
rant. Moderate elevations (i.e., 73T to 841) suggest an element of exaggeration of
complaints and problems. Any interpretive hypotheses based on clinical scale ele-
vations should be considered with caution, because there is some possibility that
the hypotheses will overrepresent the extent and degree of significant test findings.
The likelihood of distortion increases in the range from 84T to 92T. Elevations in
this range may be indicative of a “cry for help” or an extremely negative evaluation
of both self and life; some deliberate distortion of the clinical picture also may be

113
PAI Interpretive Guide

present. The cutoff of 84T has been found to optimally discriminate malingerers
from actual patients when the a priori probability of malingering is 50%.
High scores on NIM (i.e., =>92T) suggest that the respondent attempted to por-
tray himself or herself in an especially negative manner. The item content suggests
the strong possibility of careless responding, extremely negative self-presentation,
or malingering; research respondents who were instructed to malinger obtained an
average NIM score of 117T; a completely random completion of the PAI would
result in an average NIM score of 96T. Regardless of the cause, however, the test
results are best assumed to be invalid, and no clinical interpretation of other PAI
scales is recommended when scores are in this range.
The utility of NIM as an indicator of malingering has been explored in a num-
ber of research studies. The PAI Professional Manual(Morey, 1991) details the
results of studies where college students were instructed to simulate the responses
of an individual with a severe mental disorder. The distributions of actual clinical
respondents and these malingerers crossed at a score of 84T; this cutoff yielded a
sensitivity of 88.6% in the identification of malingering, with a specificity of 89.8%
among true clinical respondents. The 2 clinical standard deviation cutoff of 92T
resulted in a sensitivity of 86.5% and a specificity of 94.1%.
A sophisticated study of malingering was performed by Rogers, Ornduff, and
Sewell (1993), who examined the effectiveness of the NIM scale in identifying both
naive and sophisticated simulators (advanced graduate students in clinical and
counseling psychology) who were given a financial incentive to avoid detection as
malingerers. Rogers et al. found that the recommended NIM scale cutoff success-
fully identified 90.9% of respondents attempting to feign schizophrenia, 55.9% of
respondents simulating depression, and 38.7% of respondents simulating an anx-
iety disorder. In contrast, only 2.5% of control respondents were identified as sim-
ulators. Interestingly, there was no effect of subject sophistication; the scale was
equally effective in identifying both naive and sophisticated malingerers. Rogers et
al. concluded from these results that the NIM scale is most effective in identifying
the malingering of more severe mental disorders.
Gaies (1993) conducted a similar study of malingering, focusing on the feigning
of clinical depression. Gaies compared four groups of women: an “informed malin-
gering” group who were instructed to malinger depression, and who were given
detailed information about the diagnosis of depression; a “naive malingering”
group who were asked to simulate depression but were given no information
about the disorder; a “depression” group consisting of outpatients being treated for
depression who obtained Beck Depression Inventory scores above 14 and MMPI-2
Scale 2 scores above 64T; and a “control” group of college students responding to

114
 Negative Distortion: Random Responding and Malingering

the PAI in standard fashion. Average scores on NIM were 92T for the informed
malingerers and 81T for the naive malingerers. Sensitivity and specificity results
for particular NIM cutoffs were not reported. However, these results are similar to
those of Rogers et al. (1993) in suggesting that respondents attempting to simulate
milder forms of mental disorder (in this case, depression) will obtain more
“moderate” elevations on NIM, (i.e., scores of around 85T) as opposed to the
scores of 110T that are typical with the simulation of psychosis.
Rogers, Sewell, Morey, and Ustad (in press) investigated the effectiveness of
NIM in detecting individuals feigning three specific disorders: schizophrenia,
major depression, and generalized anxiety disorder. This study compared naive
simulators (undergraduates with minimal preparation) with sophisticated simula-
tors (doctoral psychology students with one week of preparation and coaching)
and actual clinical respondents diagnosed with the three disorders in question.
Although the naive simulators obtained NIM scores that were quite elevated (M =
84T), the scores of the sophisticated simulators (M = 69T) differed only slightly
from those of the bona fide clinical respondents (M = 63T). In this particular mix
of respondents, the optimal NIM cutting score was 77T, which was reasonably
effective in identifying the naive simulators (69% for malingered schizophrenia,
82% for depression, 45% for anxiety) but less effective with the sophisticated simu-
lators (55% for feigned schizophrenia, 19% for depression, 0% for generalized
anxiety). In contrast to the Rogers et al. (1993) study, this study found that sophis-
ticated participants were considerably more effective in avoiding detection by the
NIM scale, with one potential source of the difference involving the use of specific
preparation in the Rogers et al. (in press) study. These results suggest that the util-
ity of NIM as a measure of malingering is affected by preparation and coaching for
the evaluation.
In summary, the NIM scale has a place in the assessment of malingering on the
PAI, but it also has limitations. It appears to work best with efforts to simulate
severe forms of mental disorder; where milder forms of disorder are falsified, it is
less effective. In addition to limitations associated with the type of mental disorder
malingered, the NIM scale has limited utility as a specific indicator of malingering.
This is because NIM was designed as a general measure of a response style, that
would lead the clinician to form a more negative impression than might be objec-
tively warranted; it is not a malingering scale per se.

Malingering Responding: Scale Configuration

The mean profiles for malingered mental disorder from Morey (1991) and
malingered depression, both naive and informed (Gaies, 1993), are presented in
Figure 4-2. The most prominent characteristic of the profile from Morey (1991) is

115
PAI Interpretive Guide

D E Le ee
PROFILE FORM FOR ADULTS-SIDEA 1 2 3 4
eo——e Malingered mental
disorder
@—=—-@ Malingered depression
(informed)
@-------@ Malingered depression
(naive)

90

RCC
TC
1 2 3 4 5 6 if 8 9 10 41 A B c D — ¥ pe
ICN INF NIM PIM SOM ANX ARD DEP MAN PAR scz BOR ANT ALC DRG AGG Sul STR NON RXR DOM WRM

PROFILE FORM FOR ADULTS - SIDE B

rts 24-
e——e Malingered mental
24 18- disorder
24-
24- @—=—- Malingered depression
(informed)
@---+---@ Malingered depression
(naive)

©3

80

70

|
$8109S

Gey
E o- 025 fo - o-: o-
SOMG SOMS SOMH ANXC ANA ANKP ARD-O ARD-P ARDT DEPC DEP-A DEPP MANA MANG MANI PARH PARP PARR SCZP SCZS SCZT BORA BORI BORN BORS ANTA ANTE ANTS AGGA AGGV AGGP
CONV SOMA HEAL COG AFF PHYS OBS PHOB TRAU COG AFF PHYS ACT GRND IRRi HYPE PERS RSNT PSYC SOC THGT AFF ID NEG SELF ANT £&GO STIM AGG VERB PHYS
SYMP ZATN CONC SYMP SYMP SYMP COMP IAS STRS SYMP SYMP SYMP LEVL JOSY BLTY VIG CUTN MENT EXP DET IS INST PROB REL HARM BEH CEN SEEK ATT AGG AGG

Figure 4-2. PAI profiles for malingered mental disorder (Morey, 1991) and malingered depression,
informed and naive (Gaies, 1993).

116
 Negative Distortion: Random Responding and Malingering

obviously the extreme elevation on NIM, which falls far above the thresholds for
profile validity described in previous sections. The Gaies profiles both demonstrate
NIM elevations, although, in the simulation of depression, NIM appears to be a less
prominent part of the profile. On the validity scales, INF also tends to be elevated
in these samples; this combined elevation of INF and NIM is quite rare in actual
protocols, with scores above the recommended cutoffs on both scales occurring in
only 0.2% of respondents in both the community and clinical normative samples.
Although the random-responding profile shown in Figure 4-1 also had both of
these scales elevated, there are differences in the configuration of the two scales.
Malingered protocols tend to lead to profiles where NIM greatly exceeds INF, typ-
ically by 20T or more; when scores on the two scales are comparable (i.e., within
10T of one another), random responding is suggested.
The malingered profiles also tend to be quite elevated, with many clinical
scales above 70T. Although profiles from random responding also can be ele-
vated, the malingered profiles tend to have sharper differentiations than randomly
produced profiles, with some scales (e.g., SCZ) likely to elevate markedly, and
others (e.g., MAN) influenced less consistently when pathology is simulated.
Nonetheless, there are a variety of profile elements on the clinical scales that are
inconsistent with those generated even in severe mental disorder. For example, it
is very unusual to have both DEP and MAN scores above 70T, but this pattern was
seen in the simulation of severe mental disorder (Morey, 1991); even bipolar
patients rarely obtain these elevations simultaneously. Also, the RXR score tends to
be rather high for profiles with this degree of pathology; as will be discussed later,
this probably results from inaccurate lay stereotypes of individuals with a severe
mental disorder.
An actuarial use of PAI profile information for detecting malingering was pro-
vided in a study by Rogers et al. (in press) described in the previous section. This
study constructed a discriminant function that was designed to distinguish the
profiles of bona fide patients from those simulating such patients (including both
their naive and sophisticated simulator groups). This study found that the result-
ing function was considerably more accurate than the NIM scale in isolation in
identifying the feigned disorders. The discriminant function loadings are presented
in Table 4-1; to obtain the function score, each weight is multiplied by the T score
for the corresponding PAI scale and the resulting numbers are summed (including
a value for a constant). Rogers et al. evaluated the effectiveness of a cutting score
of .12368 for this function; scores above this value were predicted to be feigned,
whereas scores below this value were predicted to be bona fide cases. The function
was found to have a sensitivity of 87% in identifying feigned disorder and a 96%
specificity; in cross-validation, the sensitivity for malingering identification was

be
PAI Interpretive Guide

Table 4-1
Discriminant Function Weights Used in Computation of the
Discriminant Function for the Assessment of Malingering
Scale Weight
ICN + .01718613
INF + .01976398
SOM-C — .03403340
SOM-H + .02824221
ANX-A — .04109886
ANX-P + .05324155
ARD-O — 01773748
ARD-P + .02758030
ARD-T — .01741280
DEP-C + .04121700
PAR-H + .01603311
PAR-R + .01554190
SCZ-P + .01775538
SCZ-T — .02750892
BOR-! — .02909405
BOR-N + .03675012
BOR-S — .01793721
ANT-E + .02152554
STR — 01917862
RXR + .02103711
Constant — 6.60458400

Note. Material in this table was adapted from “Detection of Feigned Mental Disorders on the Per-
sonality Assessment Inventory: A Discriminant Analysis,” by R. Rogers, K. W. Sewell, L. C. Morey,
and K. L. Ustad, in press.

found to be 80% and the specificity was 81%. These results suggest that this func-
tion, which uses information from 20 different PAI scale and subscale scores, can
successfully identify over 80% of individuals attempting to simulate a wide array
of emotional disorders, ranging from mild (e.g., generalized anxiety) to severe
(e.g., schizophrenia) pathology.
To investigate the generalizability of this function and to obtain an estimate of
expected distributions when the function scores are calculated, Table 4-2 presents
descriptive statistics for the results of the function when applied to various sam-
ples from the PAI Professional Manual (Morey, 1991). This Table reveals that the
only group to obtain a mean score above the cutoff recommended by Rogers et al.

118
 Negative Distortion: Random Responding and Malingering

Table 4-2
Descriptive Statistics for the Rogers Discriminant Function (RDF)
for Relevant Samples From the PAI Professional Manual
a
Sample type N M SD
Community sample : 1,000 —1.00 1.08
Clinical sample 1,246 —1.15 Wel
Student sample, “fake bad” responding 44 2211, 1.16
Student sample, “fake good” responding 45 —1.15 0.72

(in press) was the “fake bad” or malingering group; all other groups obtained values
that were 2 to 3 standard deviations below this group. It is interesting to note that
the community and clinical samples obtain very similar scores on this function,
even though they tend to obtain dissimilar values on other validity indicators such
as NIM. Such a result suggests that this function may be useful in identifying
malingering in a variety of assessment contexts, including both clinical and com-
munity settings. To assist in interpreting the results of the Rogers et al. Discrimi-
nant Function (RDF), Table 4-3 lists T-score equivalents for different values of the
function as applied to the community norms presented in Table 4-2. Using the
results presented in Table 4-2, it appears that individuals instructed to simulate the
responses of patients with severe mental disorders typically obtain scores of
around 80T; the empirically based cutoff recommended by Rogers et al. corre-
sponds to a value of roughly 60T. Thus, scores at or above 60T suggest the possi-
bility of efforts to feign mental disorder, whereas scores at or above 70T on the
RDF index are indicative of overt attempts at malingering.
The Appendix provides a variety of correlates for RDF scores; some of the most
informative of these correlates are presented in Table 4-4. This table reveals that
the RDF score is positively but weakly related to most other indicators of negative
distortion; correlations are minimal with NIM and with the MMPI F scale, and
somewhat higher with the Malingering Index (to be discussed in the following sec-
tion). Within the realm of clinical constructs, the RDF is most associated with
PAR-R of the PAI scales and subscales, and with Si on the MMPI. On other indica-
tors, State Anxiety from the STAI (Spielberger, 1983) is among the largest corre-
lates. In summary, the RDF appears to be tapping an element of negative distor-
tion that is largely independent of most other indicators of profile validity, providing
the potential for a separate evaluation of validity concerns. In part, this indepen-
dence is achieved through the inclusion of suppressor variables in the function
that serve to remove the overall level of pathology manifest in the profile to a

me
PAI Interpretive Guide

Table 4-3
T-Score Conversions for Rogers Discriminant Function (RDF)
Scores, Standardized Against the PAI Normative Sample

Function T-score Function T-score

result equivalent result equivalent

—5.00 13 0.25 62
—4.75 15 0.50 64
—4.50 18 0.75 66
—4.24 20 1.00 69
—4.00 22 1.25 71
—3.75 25 1.50 iis
—3.50 27 1.75 75
—3.25 29 2.00 78
—3.00 31 2.25 80
—2.75 36 2.50 82
—2.50 36 2.75 85
—2.25 38 3.00 87
—2.00 A 3.25 89
-1.75 43 3.50 92
—1.50 45 3:75 94
-1.25 48 4.00 96
—1.00 50 4.25 99
-0.75 52 4.50 101
—0.50 55 4.75 103
—0.25 57 5.00 106
0.00 59

considerable degree, reducing correlations with validity indicators that are associ-
ated with global psychopathology. Evidence of this can be seen in Table 4-2, which
demonstrates that the clinical and community normative samples have similar scores
on the RDF despite marked differences in the overall elevation of the profiles in these
samples. To the extent that the RDF is related to specific forms of psychopathology,
it is modestly related to a hostile and anxious withdrawal from others. |

The PAI Malingering Index (MAL)

The need for a specific indicator of malingering more specific than NIM led to
the development of the Malingering Index (Morey, 1993). The Malingering Index
(MAL) is comprised of eight configural features of the PAI profile that tend to be
observed much more frequently in the profiles of respondents simulating mental
disorder (particularly severe mental disorders) than in actual clinical patients.

120
 Negative Distortion: Random Responding and Malingering

Table 4-4
Selected Correlates of the Rogers Discriminant Function (RDF) Score
Se
aia a leet Setanta,
Correlation with
Variable description RDF score
PAI INF ; 37
PA! NIM .09
PAI Malingering Index .26
MMPI F 13
PAI PAR-R 43
MMPI Si ‘ot
State-Trait Anxiety Scale, State Anxiety 39
Note. MMPI = Minnesota Multiphasic Personality Inventory.

Table 4-5
Prevalence of Features of the PAI Malingering Index (MAL) in
Community, Clinical, and Simulating Samples
Fake Fake
Item Community’ Clinical? bad° good‘
Index item weight M M M M
1. NIM > 110T 1 .00 01 64 .00
2. NIM minus INF >20T 1 .04 19 13 .00
3. INF minus ICN = 15T 1 .08 .08 61 s¥/
4. PAR-P minus PAR-H = 15T 1 .05 .07 43 .00
5. PAR-P minus PAR-R > 15T 1 .05 .09 .39 .00
6. MAN-I minus MAN-G = 15T 1 .09 23 52 .00
7. DEP > 85T and RXR > 45T 1 .00 01 .36 .00
8. ANT-E minus ANT-A = 10T 1 15 M2 we #5
Total M 46 .80 4.41 HE
Total SD 74 .98 1.80 73
AN = 1,505. PN = 1,246. °n = 44. Sn = 45.

Table 4-5 presents these eight features and the proportions of individuals mani-
festing the features in four samples: the community and the clinical normative
samples, and samples of college students instructed to “simulate a major mental
disorder” (fake bad) or to “present your best possible front” (fake good). This table
reveals that each feature was observed with far greater frequency in the “fake bad”
group than in actual clinical or community samples.

121
PAI Interpretive Guide

The computation and significance of the eight Malingering Index (MAL) items
are as follows:

1. NIM2=11O0T (1 point). As mentioned previously, NIM represents the starting

point for the assessment of malingering, although its limits for this use have been
described. Nonetheless, although elevations on NIM are not uncommon in clinical
samples, extreme elevations (i.e., 2 110T) are very uncommon in samples of
respondents completing the test under standard instructions. In contrast, eleva-
tions in this range are the rule rather than the exception for respondents instructed
to simulate disorders; in fact, the average NIM score for such respondents is typi-
cally above 110T. Although only 1% of respondents in the clinical normative sam-
ple and none of the respondents in the community normative sample obtained
scores of 110T or greater, a full 64% of “fake bad” respondents did so. Thus, any
NIM T score at or above 110T adds 1 point to the Malingering Index.
2. NIM minus INF 2 20T (1 point). Both NIM and INF are validity scales with
extremely small means and standard deviations in the community normative sam-
ple; as such, endorsing only a few items on each of these scales can lead to eleva-
tions. This means that, statistically, the two scales are quite similar and that, on
each scale, item endorsements are infrequent. The only difference lies in the con-
tent of the scales; although the content of INF items is unusual (e.g., having your
favorite sporting event be the high jump), it does not necessarily appear to be
pathological, even to the lay person. In contrast, many NIM items were selected
specifically because they address lay stereotypes of mental disorder in a dramatic
and overt way. Therefore, individuals attempting to simulate mental disorder are
much more likely to endorse the pathological-sounding NIM items than the
unusual, but relatively benign, INF items. A difference of at least 20 T-score points
in favor of NIM is typical in such individuals. The T score of NIM is at least 20
points higher than that of INF in 73% of “fake bad” respondents, but this differ-
ence occurs in only 19% of clinical patients and less than 5% of community
respondents. A difference of 20 points or more in this direction adds 1 point to the
Malingering Index.
3. INF minus ICN 2 I5T (1 point). Although the content of INF items is less
pathological than that of NIM items, the former are sufficiently unusual sounding
to sometimes be misinterpreted as indicators of mental disorder by respondents
attempting to simulate psychopathology. For example, the item “Most people look
forward to a trip to the dentist” is generally recognized as a false statement by most
test-takers. However, respondents attempting to simulate severe disorders may
misinterpret this item as measuring insight and perceptual accuracy and give a
Very Irue response. As such, INF elevations in malingering samples, although

122
 Negative Distortion: Random Responding and Malingering

nowhere near as prominent as NIM elevations, are commonly obtained. In con-

trast, the ICN scale, which, like INF, was designed to identify carelessness or ran-
dom responding, is rarely elevated in simulation studies. This is most likely
because an individual who is motivated to simulate a mental disorder will be care-
ful and consistent in responding to questions; the PAI profile elevations that result
are not from carelessness, but rather a careful attempt to manipulate self-presenta-
tion. As a result, INF scores are typically more than 15 T-score points higher than
ICN scores in such samples. A difference of this magnitude is found in 61% of
“fake bad” respondents, but in only 8% of the clinical and community normative
respondents. A difference of 15 T-score points or more in favor of INF adds 1 point
to the Malingering Index.
4. PAR-P minus PAR-H 2 15T (1 point); 5. PAR-P minus PAR-R > 15T (1 point).
Both of these indicators are obtained by examining the configuration of subscales
on the PAR scale. The item content of the three subscales of PAR is fairly different,
with PAR-P comprised largely of items tapping paranoid psychotic content (e.g.,
ideas of reference, or delusions of persecution), whereas PAR-R and PAR-H assess
more personological or cognitive features of paranoia (e.g., hostility, resentment,
and wariness). The latter two aspects of the syndrome are more commonly
encountered in the general population (as witnessed by their relatively high mean
score in comparison to PAR-P) and, thus, do not seem as overtly pathological to a
lay population. Nonetheless, these elements are invariably part of the paranoid
syndrome, and these scales should be elevated in any full-blown delusional
episode of the sort that might lead to an elevation on PAR-P. Most individuals who
have a smattering of knowledge about psychopathology can recognize persecutory
symptoms as evidence of paranoia, but they understand little else about the disor-
der. Thus, lay respondents instructed to simulate psychosis do not recognize vigi-
lance and resentment as being central to a paranoid psychosis, and they rarely pro-
vide scores on either PAR-H or PAR-R that approach that of PAR-P. Differences of
15 points between these scales are obtained roughly 40% of the time in simulation
samples, but less than 10% of the time in actual clinical respondents. For either
comparison, a difference of 15 T-score points or more in favor of PAR-P adds 1
point each to the Malingering Index (MAL).
6. MAN-I minus MAN-G = I5T (1 point). The items of the MAN-I scale address
a ready irritability that arises out of a low tolerance for frustration and an impa-
tience with interference from others. Implicit in these frustrations is a fundamen-
tal belief in the importance and significance of the respondent's own plans, and,
as such, there is an underlying grandiosity typically associated with this irritabil-
ity, However, to the lay observer, these two components seem quite unrelated;

128.
PAI Interpretive Guide

whereas MAN-I items suggest marked problems with other people and are inter-
preted as a negative attribute, MAN-G items denote positive self-esteem and do not
overtly appear to be indicative of psychopathology. Thus, in malingering simula-
tion samples, these scales often are found to be inversely correlated, whereas, in
clinical samples, they are positively associated. Over one half of simulation respon-
dents will obtain a 15-point differential in favor of MAN-I, whereas less than a
quarter of clinical respondents will display this pattern. Hence, a difference of 15
T-score points, with MAN-I larger than MAN-G, adds 1 point to the Malingering
Index (MAL).
7. DEP = 85T and RXR 2 45T (1 point). One of the most prominent lay stereo-
types of mental disorders is that individuals suffering from such disorders all lack
insight into the nature and severity of their condition. These beliefs can be found
in expressions such as, “The first step to being normal is knowing that you are
crazy.” These stereotypes often lead lay observers to underestimate the marked dis-
tress and motivation for change that characterize most forms of mental disorder.
On the PAI, it is very uncommon to find respondents who report both a signifi-
cant degree of distress (as indicated by elevations on DEP) and little motivation to
change (as indicated by scores on RXR approximating those obtained from non-
patients). In actual clinical samples, these two scales display a strong inverse rela-
tionship, and elevations on DEP are typically associated with very low (i.e., < 357)
scores on RXR. Scores at or above 45T on RXR are almost never associated with
DEP scores of 85T or above in true patients. However, in simulation samples, RXR
is often found to be close to 50T, even in the presence of indicators of marked dis-
tress; more than a third of such patients meet this MAL item. Thus, a score at or
above 45T on RXR coupled with a DEP score of 85T or above adds 1 point to the
Malingering Index (MAL).

8. ANTE minus ANT-A 2 10T (1 point). As noted previously with the scales
assessing mania and paranoia, lay observers often recognize certain dramatic or
prominent elements of syndromes of mental disorders without understanding
other facets of these syndromes that invariably are associated with these elements.
This tendency also appears to occur in the consideration of the different facets of
antisocial personality. Here, the unempathic and amoral elements of antisocial per-
sonality strike lay observers as severely pathological; in fact, the majority of such
observers appear to attribute such characteristics to all patients with severe men-
tal disorder. Thus, when such lay persons attempt to simulate severe psy-
chopathology, they invariably obtain marked elevations on ANT-E. Apparently
however, the antisocial behaviors that are actually the invariable sequelae of this
character style are not part of the stereotype of mental disorder: scores on ANL-A

124
 Negative Distortion: Random Responding and Malingering

in such samples are often within normal limits. In simulation samples

, 73% of
malingerers obtain ANT-E scores that are 10 or more T-score points above
the
ANT-A score, whereas only 12% of true clinical respondents yield this differen
tial.
When observed, this difference adds 1 point to the Malingering Index (MAL).
Factors Underlying Malingering Index (MAL) Items. In order to further facilitate
interpretation of the Malingering Index, the eight items of the index were factor
analyzed (principal components analysis followed by varimax rotation) using data
obtained from both the community and clinical normative samples. In both sam-
ples, two factors could be extracted, and the loadings were quite similar in both
samples. The first factor involves an endorsement of severe and rather unusual
psychotic symptoms, without the marked anxiety and wariness in dealing with the
environment that typically accompany these symptoms. Malingering Index (MAL)
Items 4, 5 and, to a lesser extent, 1 are included on this factor. Prototypic of indi-
viduals endorsing such items would be a person who believes that others are plot-
ting against him or her (as indicated by PAR-P), yet who professes to experience
relatively little anger, anxiety, or resentment that this is occurring. The second fac-
tor involves a tendency to portray oneself (i.e., a very low MAN-G) and one’ envi-
ronment (i.e., a very high NIM) in a very negative light, accentuating the negative
and minimizing the positive elements of each. Malingering Index (MAL) Items 1,
2, and 6 demonstrated substantial loadings on this factor. Prototypic of such indi-
viduals would be those who reflexively deprecate all elements of their experience,
including their experience of themselves.

PAI Malingering Index (MAL) Interpretation

Table 4-5 reveals that the average score for a malingering sample on the Malin-
gering Index was 4.41 items, compared to a mean of 0.80 for the clinical stan-
dardization sample and 0.46 for the community normative sample. Transforma-
tion of Malingering Index items to T scores, based on the means and standard
deviations of both the community standardization sample and the clinical stan-
dardization sample, are listed for convenience in Table 4-6. This table reveals that
the malingering sample obtained a score that was more than 5 standard deviations
above the mean of the community sample; this score was also more than 3 stan-
dard deviations above the mean for the clinical sample. The latter is probably the
better referent, as malingering involves the distinction between actual and feigned
disorders, rather than a comparison to community normal respondents. Thus,
using a 2 standard deviation referent against the clinical sample, MAL scores of 3
or above should raise questions of malingering; scores of 5 or more are highly
unusual in clinical samples, and they tend to occur only when severe mental dis-
order is being feigned.

125
PAI Interpretive Guide

Table 4-6
T-Score Equivalents for the PAI Malingering Index (MAL)
Standardized Against Community and Clinical Samples
Normative ae
er AEEOEE ELA DI ATE
T-score equivalent, T-score equivalent,
MAL score community* norms clinical> norms

0 44 42
1 57 52
2 71 62
3 84 72
4 98 83
5 111 93
6 125 103
7 138 113
8 151 123
aN = 1,000. °N= 1,246.

The Gaies (1993) study of the malingering of depression (described earlier in

this chapter) provides additional support for the utility of the Malingering Index.
Using a cutoff of 3 or greater as an indicator of malingering, Gaies found a
sensitivity of 56.6% for identifying the informed malingerers and 34.2% for iden-
tifying the naive malingerers. Specificity of the index in a sample of patients who
were actually depressed was 89.3%, whereas normal controls demonstrated a
specificity of 100%. Similar to the results obtained using NIM, it appears that the
sensitivity of the Malingering Index will decline somewhat when milder forms of
psychopathology (e.g., depression or anxiety) are being simulated. In settings
where the malingering of such disorders is a concern, adjustments to the Index
cutoff may be needed to optimize the utility of decisions.
The Appendix provides a variety of correlates for the Malingering Index
(MAL), of which selected results are presented in Table 4-7. This table reveals that
the Malingering Index is moderately related to NIM (which should be expected, as
NIM elevations comprise part of the Index) and also somewhat related to the
MMPI F scale. Within the realm of clinical constructs, the MAL is most associated
with PAR-P of the PAI scales and subscales, and with Pa on the MMPI, confirming
that dramatic paranoid psychotic features are most likely to lead to Index eleva-
tions. On other indicators, an Antisocial presentation on diagnostic interview is
among the largest correlates of this Index. In summary, the Malingering Index
appears to be tapping an element of negative distortion related to an atypical pre-
sentation of severe mental disorder, particularly that of paranoid psychosis. In

126
 Negative Distortion: Random Responding and Malingering

Table 4-7
Selected Correlates of the PAI Malingering Index (MAL) Total Score
Correlation with
Variable description MAL score
PAI INF .00
PAI NIM 61
PAI PIM —.39
MMPI F .39
PAI PAR-P 156
MMPI Pa .38
Diagnostic Interview for Personality Disorder, Antisocial diagnosis 42

Note. MMPI = Minnesota Multiphasic Personality Inventory.

combination with other features such as NIM and the RDF score, the Malingering
Index can help identify profiles that suggest overt attempts at the feigning of men-
tal disorder.

127
 k

>

i
4 ' iptv ey i

a \

2] { ~ =
a t whi»)

i tice

° ‘ Pe) y it ;

=a Pry, wi te Vet “Lin or 7) oe eee he >

Fi d Wlbis
ariti eu - { im ? y Ta
Se Ser ee wv = ; ue

Dt Se rivets! wes ah. aries suMataihd : alerts Pose

re #%. SuTial toy 1 1201 obisiiatl ¢ wh a

: a — Pian
 CHAPTER 5
IDENTIFYING DEFENSIVENESS
ON THE PAI
The stigma of mental illness and the limitations of psychological insight in
most people can give rise to an underreporting of clinical problems that can poten-
tially distort the accuracy of self-reported information. The distorting factors can
be quite diverse. They may result from personality traits or from situational influ-
ences on a respondent; they may involve intentional distortion or a genuine lack
of insight; and they may involve selective defensiveness about some problem areas
(e.g., substance abuse) but not about others (e.g., depression). For this reason, the
PAI does not include any “correction” factors of the sort employed by other inven-
tories; these corrections invariably fail to enhance validity, mainly due to the use
of one omnibus correction that cannot discriminate among such varied influences
on defensiveness. Nonetheless, it is clear that the assessment of such potential dis-
tortions is an important part of interpreting any self-report instrument, and iden-
tifying such distortions has arguably been the most difficult assessment task for
researchers and clinicians. Most procedures that have been developed to identify
defensiveness show large overlap with normal functioning, leading some investi-
gators to speculate about widespread “illusory mental health” (Shedler, Mayman,
& Manis, 1993) when assessments are based upon self-reports.
Detecting generally defensive response patterns with the PAI may be accom-
plished through the use of a number of strategies, including the PIM scale and pro-
file configuration information such as that used in the Defensiveness Index; spe-
cific denial of substance abuse problems is discussed in chapter 2. Each strategy
provides information useful in assessing profile validity, but none is infallible; the
identification of defensiveness continues to be one of the most difficult challenges
to self-report psychological assessment. As with any assessment, all sources of
information should be considered. The supplementation of PAI profile information
with concurrent reports from family members, peers, documents or records, and
other psychological and/or laboratory testing is recommended when situational
factors raise the a priori probability of underreporting of clinical problems. Such
situations can include, but are not limited to the following: preemployment
screenings, fitness for duty evaluations, child custody suits, criminal dispositions,
and involuntary hospitalization or treatment decisions.

129
PAI Interpretive Guide

Positive Impression (PIM )

The content of PIM scale items involves the presentation of a very favorable
impression or the denial of relatively minor faults. The items were selected by exam-
ining the distributions of scores for normal respondents, patients, and research
respondents responding to the PAI under positive-impression-enhancement instruc-
tional sets. The items were selected on the basis of low endorsement frequencies
in both normal and clinical respondents; however, PIM items are endorsed with
greater frequency in normal adults than in clinical patients. Marked elevations in
clinical respondents are particularly rare and, hence, are interpretively significant
if obtained. Both patients and normal respondents score considerably lower than
research respondents completing the PAI under a positive impression enhance-
ment instructional set.
For the most part, PIM items offer the opportunity for individuals to say some-
thing negative about themselves. Hence, elevated scores indicate that respondents
do not take many opportunities to say negative things about themselves. There are
a number of reasons why individuals might not report negative characteristics.
One possibility is that they have no negative characteristics, or, at least, they have
fewer of these than most people. A second possibility is that they are not telling
the truth, that they are trying to deceive the recipient of the test results into believ-
ing that they have more positive features than they really do. A third possibility is
that they simply are not aware of certain faults that they may have, that they lack
insight into some of their personal shortcomings. In either of the latter two
instances, the results of a self-report test will lead one to form a more positive
impression of the respondent’s life circumstances and psychological adjustment
than would probably be merited in the opinion of an independent observer. It is
these latter two characteristics that PIM was designed to measure.
It should be recognized that the tendency for favorable self-presentation is
actually fairly common in the normal population. Typically, most clinically derived
cutting scores on indices of social desirability will endup identifying 30% to 40%
of the general population as “faking good.” Such results underscore the difficulty
of distinguishing defensive responding from normality with respect to clinical
instruments. A number of instruments have used scales similar to PIM to “correct”
other scales on the test for defensive responding, as if such scales tapped pure sup-
pressor variables, but the PAI makes no such correction for a number of reasons.
First, using a single scale to correct numerous other scales lessens the discriminant
validity of those other scales by forcing them to intercorrelate artifactually. A sec-
ond reason involves the mistaken assumption that social desirability is a suppres-
sor variable; this is not the case, because it actually does correlate with clinical

130
 Identifying Defensiveness on the PAI

criteria in most instances. In other words, most forms of mental disorder

involve
symptoms that are not socially desirable. For example, hallucinations or drinking
problems simply are not desirable characteristics, and people who have such prob-
lems often say negative things about themselves because they are true, not just
because they are more willing to admit to them. Thus, attempts to remove “social
desirability” from clinical scale scores will remove criterion-related variance and.
hence, lower validity.
A third problem with defensiveness “corrections” is that this construct has
strong situational influences on it. Although people certainly may be defensive in
responding to self-report inventories, there are a variety of reasons why they are
defensive and an equal variety of reasons why some particular scales might be
affected and other scales might not. If an individual is acutely paranoid and does
not want the examiner to know this, the biggest effect is likely to be on PAR. If an
individual is using drugs, but is trying to hide this, the effect will be observed on
DRG. A person masking depression may have lowered effects on DEP and, per-
haps, inflated scores on MAN. To be useful, any correction to such scores should
be made based on the nature of the individual and the type of problems he or she
has and on whatever operations of denial and defensiveness are taking place.
Application of a single correction to multiple scales by treating the characteristic
as a trait is doomed to failure, because these influences are not cross-situationally
consistent. This does not mean that scales such as PIM are unimportant; it simply
means that such corrections must be made integratively by the test interpreter who
has access to all situational and contextual information.
The original decision points for interpreting PIM scores were based on the per-
formance of respondents who were attempting to manage their impression in a
positive direction. The point of rarity between the impression management sam-
ple (i.e., “fake good”) and the community normative sample can be used as a cut-
ting score, above which a person's scores are more similar to those of simulated
respondents than to those of typical respondents. This point of rarity was 18 or
above (57T) on PIM. Application of this rule resulted in a sensitivity in the iden-
tification of positive dissimulation of 82%, and a specificity with respect to normal
respondents of 70%. Interestingly, a study by Cashel, Rogers, Sewell, and Martin-
Cannici (1995) also identified 57T as their optimal cutting score, which yielded
sensitivity and specificity rates of 48% and 81%, respectively. Cashel et al. (1995)
further found that scores below 43T could be used with very high specificity (e.,
virtually no respondents attempting to manage their impression obtained scores
below that value on PIM). Finally, Fals-Stewart (1996), in a study of defensive
responding and denial among drug abusers (described in more detail in chapter 2),
found that the 57T cutting score on PIM had a sensitivity of 88% and a specificity

131
PAI Interpretive Guide

of 80% in identifying substance-abusing individuals motivated to avoid detection

by the PAI.
Thus, low scores on PIM (i.e., < 44T) are strongly indicative of honest respond-
ing. Generally, scores between 44T and 57T suggest that the respondent did not
attempt to present an unrealistically favorable impression in completing the test,
although scores in the upper end of this range tend to be unusual in clinical set-
tings. Moderate elevations (i.e., 57T to 66T) suggest that the examinee responded
in a manner to portray himself or herself as relatively free of the common short-
comings to which most individuals will admit. With PIM in this range, the accu-
racy of interpretations based on the PAI clinical scales profile may be distorted, and
interpretive hypotheses should be reviewed with caution. It is likely that the PAI
profile will underrepresent the extent and degree of significant test findings.
Another method of identifying problematic protocols uses cutting scores
derived from the distributions of clinical respondents. Use of the reference data
from clinical respondents is particularly relevant for the impression management
scales, as normal respondents and clinical respondents tend to differ in this regard.
For PIM, a raw score of 22 or above (66T) corresponds to a score that is more than
2 standard deviations above the mean for clinical respondents. Applying this more
conservative decision rule results in decisions that are more specific to attempts at
impression management than those provided by the empirical cutoffs. In other
words, relatively few actual protocols are identified as invalid using the 2-standard
deviation clinical norms cutoff; the specificity rates for this cutoff are 95% or
greater, although the sensitivity in identifying dissimulated protocols falls to 52%
(Morey, 1991), and sensitivity drops to 17% when respondents are coached regard-
ing believability of results (Cashel et al., 1995). Thus, such high scores on PIM (i.e.,
2 667) suggest that the respondent attempted to portray himself or herself as excep-
tionally free of the common shortcomings to which most individuals will admit.
When scores in this range are obtained, the validity of the PAI clinical scale profile
is seriously questioned, and no clinical interpretation of other PAI scales is recom-
mended. However, such scores are usually rare, and concerns about defensiveness
should be raised at even lower scores, as noted earlier.

Defensive Responding Profile Configuration

The PAI Professional Manual (Morey, 1991) described the results of a study of
45 college students who were instructed to respond as if they were taking the PAI
to qualify for a desired job and they wanted to appear psychologically fit for this
job. The mean profile for this group of respondents is presented in Figure 5-1. The

132
 Identifying Defensiveness on the PAI

scores for this sample were extremely suppressed, with nearly all scores below
50T. Only five scales demonstrated scores above 60T: PIM, MAN, RXR, and the
two interpersonal scales, DOM and WRM. This basic portrayal is of an individual
with above-average self-esteem and energy who is interested and effective in inter-
personal relationships and good at controlling these relationships. Such individu-
als admit to no problems in functioning and would adamantly resist the notion
that some aspect of their lives could be changed for the better. Not only are there
no signs of any personal weaknesses, but the individuals surrounding the. person
are also described as exceptionally efficient and supportive, as evidenced by the
suppressed score on NON. Similar trends can be found in “fake good” profiles
reported in a detailed study of defensive responding on the PAI conducted by
Cashel, Rogers, Sewell, and Martin-Cannici (1995). Cashel et al. instructed two
types of respondents (i.e., college students and prison inmates) to answer the PAI
in a way that would portray them in the best possible manner, but stressed the
believability of the resulting profile. The mean profiles for the offender and student
groups (collapsed across order of taking the test) also are presented in Figure 5-1.
Again, the same five scales reveal elevations, and NON is again one of the lowest
scales. Also, it is important to note that for all three samples ICN scores are quite
low, demonstrating that intentional distortion of PAI responses in a positive direc-
tion is more likely to suppress ICN than to lead to elevations on this scale.
Cashel et al. (1995) also constructed a discriminant function that was designed
to distinguish optimally between defensive and honest responding. The authors
found that this function was more accurate in identifying dissimulated responding
than either the PIM score in isolation or the score on the Defensiveness Index, or
DEF (described later). The Cashel et al. Discriminant Function (CDF) demon-
strated sensitivities ranging from 79% to 87% in identifying falsified profiles,
with specificity of 88%. This discriminant function, calculated by multiplying PAI
T scores by a weighting factor, is as follows:

CDF = 1.67(BOR) + .97(PIM) + .72(MAN) + .60(RXR) — .52(ALC) — .68(STR)

Because these standardized discriminant function coefficients are calibrated to

the means and standard deviations of their particular samples, interpretation of the
number resulting from this formula is not straightforward in other settings. To
assist researchers and clinicians in using and interpreting the results of this for-
mula, the formula was applied to the normative community sample of the PAI,
resulting in a mean value of 138.14 and a standard deviation of 14.91 for this com-
posite. To give some idea how different groups score on this composite, Table 5-1
presents results from a number of different samples.

133
PAI Interpretive Guide

FOR ADULTS-SIDEA 4 2 3 4 5 6 if 8 9 10 14
PROFILE FORM
110 : re eo—e “Fake-good” prison inmates
(Cashel et al., 1995)
e@—-—-e “Fake-good” students
(Cashel et al., 1995)
100 @- eee ooo “Fake-good” students
(Morey, 1991)

90 $ as

: z

80

9sai00g
|
Scores
T
a fo}
60

250

40
40

30 - 5

1 2 a 4 5 6 7 8 9 10 11 A 5 c D E Y z
ICN INF NIM PIM SOM ANX ARD DEP = MAN PAR scz BOR ANT ALC DRG AGG sul STR NON RXR DOM WRM

Figure 5-1. Mean “Fake-good” PAI profiles for prison inmates and students.

Table 5-1 reveals that all of the dissimulating groups obtained scores on the dis-
criminant function that were over | standard deviation above the mean of the nor-
mative sample. It is also interesting to notice that the “fake good” samples obtained
scores that were quite similar, even though the profiles portrayed in Figure 5-1
show considerable differences in the configurations from the different settings.

Table 5-1
Discriminant Function Means for Honest and Dissimulating Groups
Sample type N M SD
Community sample* 1,000 138.14
Clinical sample* 1,246 135.28
Offender sample, honest responding® 45 135.11
Student sample, honest responding” 38 144.66
Student sample, fake good responding* 45 161.41
Offender sample, fake good responding® 45 157.04
Student sample, fake good responding® 38 161.60
Note. The discriminant function formula was developed by Cashel et al., 1995.
*From the PAI Professional Manual (Morey, 1991). "From (Cashel et al., 1995). Standard deviations
were not provided.

Lot
 Identifying Defensiveness on the PAI

Table 5-2
T-Score Conversions for Cashel Discriminant Function (CDF)?
Results Standardized Against the PAI Normative Sample
Function T-score Function T-score
result equivalent result equivalent
75 8 132 46
78 10 135 48
81 12 138 50
84 14 141 52
87 16 144 54
90 18 147 56
93 20 150 58
96 22 153 60
99 24 156 62
102 26 159 64
105 28 162 66
108 30 165 68
ad 32 168 70
114 34 171 72
117 36 174 i4
120 38 177 76
123 40 180 78
126 42 183 80
129 44 186 82
“Cashel et al. (1995).

This suggests that the function proposed by Cashel et al. (1995) may be useful in
detecting defensiveness in a wide variety of different settings. To assist in inter-
preting the result of the function, Table 5-2 lists T-score equivalents for different
values of the function. From the results presented in Table 5-1, it appears that
groups instructed to “fake good” typically yield mean scores in the range from 65T
to 70T on this composite; scores above 70T thus suggest the operation of overt
efforts at positive impression management.
The Appendix provides a variety of correlates for the Cashel Discriminant
Function (CDF; Cashel et al., 1995), of which selected results are presented in
Table 5-3. This table reveals that the CDF score is relatively independent of most
other indicators of defensiveness; the correlation with PIM is close to zero and the
association with the MMPI K scale is negative (—.29). Interestingly, the association
of the CDF score with NIM was greater than that with PIM. More consistent with

135
PAI Interpretive Guide

Table 5-3
Selected Correlates of the Cashel Discriminant Function (CDF) Score
eee ee ect Na etna ertea anata Goan = RNR Se Be© LS SA RAS SO a we a ee
Correlation with
Variable description CDF score
PAI PIM .06
MMPI K —.29
Marlowe-Crowne Social Desirability Scale PU
PAI Defensiveness Index (DEF) OL
PAI NIM .26
PAI MAN 40
PAI ALC —.52
Wiggins MMPI Hypomania 61
Wiggins MMPI Psychoticism .63
PAI SCZ-P 18
Note. MMPI = Minnesota Multiphasic Personality Inventory.

expectations, positive but modest associations were noted with the DEF the PAI
Defensiveness Index (to be discussed in the following section) and the Marlowe-
Crowne (Crowne & Marlowe, 1964) Social Desirability Scale. Within the realm of
clinical constructs, the Cashel Discriminant Function is most associated with indi-
cators of mania and hypomania, such as MAN and the Wiggins (1966) HYP MMPI
content scale. For the PAI MAN scale, the association with the full scale score was
larger than for any of the individual subscales; of the three subscales, MAN-G was
least associated with the CDF score, despite being greatly influenced in most
response set studies. The CDF score was also fairly highly correlated with the
Wiggins Psychoticism content scale, a puzzling finding, as it displayed minimal
relationships with other indicators of psychotic features, such as the PAI SCZ-P
subscale.
In summary, the Cashel (Cashel et al., 1995) Discriminant Function (CDF)
score appears to be tapping an element of positive dissimulation that is largely inde-
pendent of most other indicators of profile validity, providing the potential for a
supplemental evaluation of validity concerns. This independence is achieved
through the inclusion of suppressor variables in the function that serve to remove
the overall level of pathology manifest in the profile to a considerable degree,
reducing correlations with defensiveness indicators that are associated more glob-
ally with social desirability. Evidence of this characteristic can be seen in Table 5-1,
which demonstrates that the clinical and community normative samples have sim-
ilar CDF scores despite marked differences in the overall elevation (and hence,

136
 Identifying Defensiveness on the PAI

social desirability) of the profiles in these samples. To the extent that the
CDE score
is related to specific forms of psychopathology, it appears to be related to
a hypo-
manic and disorganized manner. In combination with other features such as PIM
and the Defensiveness Index (DEF), the CDF score can help identify profiles that
suggest Overt attempts at favorable impression management.

The PAI Defensiveness Index (DEF)

To further supplement the tools for identifying defensive responding, a set of
indicators known as the PAI Defensiveness Index (Morey, 1993) was developed.
The Defensiveness Index (DEF) is comprised of eight configural features of the
PAI profile that tend to be observed much more frequently in the profiles of
respondents instructed to present a positive impression than in actual normal or
clinical respondents. One of the items, involving the PIM scale, is double-weighted
if it exceeds the 50T threshold; all other items are worth 1 point, if present. Table
5-4 presents these eight features and the proportions of individuals manifesting the
features in four samples: the community and the clinical normative samples, and
samples of college students instructed to “simulate a major mental disorder” (“fake
bad”) or to “present your best possible front” (“fake good”). This table reveals that
each feature was observed with far greater frequency in the “fake good” group than
in actual clinical or community samples.
The computation and significance of the eight DEF items is as follows:
1. PIM 2 50T (2 points) or PIM 2 45T to 49T (1 point). The most obvious place
to begin a search for defensive responding on the PAI is with the PIM scale,
because the scale was constructed for this purpose. The first item on the Defen-
siveness Index (DEF) reflects elevations on PIM that are atypical in clinical pop-
ulations; it is quite unusual for individuals in clinical settings to obtain scores
above 50T, and even scores of 45T are uncommon among this group. This item
is weighted 2 points for PIM scores at or above 50T, and 1 point for a score
between 45T and 49T. Obviously, elevations in this range are common in the gen-
eral population, as about one half of such respondents would be expected to score
above the mean of 50T. However, the mean scores presented in Table 5-4 reveal
that nearly all respondents instructed to “fake good” score above 50T, because the
mean on this item is 1.95 out of a possible 2.0 for this group. Thus, this DEF item
appears to be a useful indicator, but, in isolation, it is not sufficient evidence of
defensiveness, because many normal respondents also will be positive for this
item.

13%
PAI Interpretive Guide

Table 5-4
Prevalenc e of Features of the PAI Defensiveness Index (DEF) in
Simulating Samples
Community, Clinical, and ee
ee eee —————

Fake Fake
Item Community® Clinical? bad° Good®
DEF Index item weight M M M M

1. PIM > 50T 2

or 45T to 49T 1 1.24 79 A 1.95
2. RXR>45T 1 70 32 61 90
3. ANT-E minus ANT-A > 10T 1 15 ne 73 35
4. ANT-S minus ANT-A > 10T 1 03 03 ov 19
5. MAN-G minus MAN-/ > 10T 1 19 16 11 0
6. ARD-O minus ANX-A > 10T 1 23 2 07 74
7. DOM minus AGG-V > 15T 1 10 07 02 65
8. MAN-A minus STR > 10T 1 19 05 aa 67
Total M 2.81 1.66 oats 6.23
Total SD 1.52 1.54 1.45 1.82
aN = 1,000. °N = 1,246. °n = 44. In = 45.

2. RXR 45T (1 point). The RXR scale (described in more detail in chapter 10)
was designed to identify openness to psychological treatment. As should be expected,
most adults in the community tend to score higher than is typical for clinical respon-
dents, indicating that individuals in clinical settings (and, hence, probably in treat-
ment) acknowledge a greater need for treatment than individuals in the commu-
nity (and, hence, probably not in treatment). However, it should be recognized
that the typical adult in the community obtains scores that suggest some openness
to the idea of changes in his or her life and a willingness to accept responsibility
for the direction such changes need to take. People who respond to the PAI in a
defensive fashion demonstrate a rigidity and opposition to psychological change
that often result in some elevation in scores on RXR. In many ways, the RXR scale
is a more subtle indicator of defensiveness than PIM. Defensive responding drives
RXR up, and even fairly sophisticated individuals instructed to “fake good” often
do not recognize that their responses are rather rigid and unwilling to consider the
possibility of personal change. RXR elevations suggest that such individuals believe
nothing about themselves needs to change. On PIM, guarded respondents may be
willing to admit to a few faults, yet on RXR they often deny any need to change
those faults, indicating they are fine precisely as they are.
The threshold for this item in the Defensiveness Index is set at 45T, which may
initially seem quite low given that it is below the mean of the normative sample.

138
 Identifying Defensiveness on the PAI

However, such elevations tend to be rare in clinical samples; more

than two thirds
of respondents in clinical settings obtain scores below 45T. In contrast, only 9%
of individuals taking the PAI under “fake good” or dissimulation responding
instructions obtain RXR scores below 45T. Thus, this criterion is particularly
effec-
tive in the identification of defensive responding in clinical settings.
3. ANT-E minus ANT-A 2 10T (1 point). ANT-E scores that are markedly elevated
reflect unempathic and amoral elements of antisocial personality that strike even
lay observers as severely pathological personality features. However, in its milder
forms, this construct seems much less negative to the general public. In fact, it rep-
resents a certain mental toughness and a self-reliance that are valued aspects in this
culture. However, the troublesome behaviors associated with ANT-A do not have
this evaluative complexity; they are unambiguously undesirable, and individuals
instructed to make a positive impression yield raw scores close to zero on this sub-
scale. People instructed to answer the test defensively often are willing to admit
that they look out for themselves, yet they will deny the possibility that this ten-
dency has ever created the types of problems for them that are typically associated
with this attitude. This is a fairly curious configural item, in that it is rarely
obtained unless the person is attempting to impression-manage in either a positive
or a negative direction. When respondents distort in a positive direction, the con-
figuration is ANT-E moderate, ANT-A very low; when distorting in a negative direc-
tion, the configuration becomes ANT-E very high, and ANT-A more moderate.
Thus, this item has the peculiar distinction of being on both the Defensiveness
Index (DEF) and the Malingering Index (MAL).

4. ANT-S minus ANT-A 2 10T (1 point). The rationale behind this item is simi-
lar to that for Item 3: certain personality traits tend to lead to an increased risk for
behavioral problems. ANT-S, which reflects a craving for novelty and excitement,
is one of these traits. A person who is constantly seeking new sensory experiences
and challenges is more likely to have experienced some behavioral problems (as
reflected by ANT-A) than a person who responds anxiously to novel situations,
and, in fact, the ANT-S and ANT-A scales correlate at .53 in the general population.
The 10-point discrepancy described in this item is almost never found in an actual
clinical community sample; being a risk-taker tends to lead to trouble, and the two
scales tend to elevate together. When there is a discrepancy with the stimulus-
seeking scale being higher, it is generally an indication that the person desires to
make a favorable impression on the recipient of the test results.
5. MAN-G minus MAN-I = 10T (1 point). The fifth item of the Defensiveness
Index (DEF) involves a comparison between the grandiosity and irritability com-
ponents of the MAN scale. Individuals attempting to manage a positive impression

wae
PAI Interpretive Guide

show a large difference on these two subscales; 77% of “fake good” respondents
show this 10-point split in favor of MAN-G, whereas only 19% of normal respon-
dents do so. This configuration involves saying a number of positive things about
oneself without admitting to the downside of these positive statements. The pro-
totype of this combination would be an individual who claims to have both the
“genius of Einstein” and the “patience of Job.” However, as self-esteem increases,
one’s patience with others who are perceived as less capable often decreases. When
less capable people get in one’s way, the natural response is one of frustration, but
the low desirability valence of frustration suppresses this score in respondents
attempting to “fake good.” The 10-point split described in this item tends to be
quite infrequent in any other population.
6. ARD-O minus ANX-A = 10T (1 point). The sixth item of the Defensiveness
Index (DEF) represents a difference in portrayed levels of obsessionality and anx-
iety. At the extreme, the Obsessive-Compulsive subscale of ARD can reflect rigid-
ity and perfectionism, but, in the moderate range, it represents a certain number
of desirable characteristics: concern with order, high standards for oneself and oth-
ers, and careful control over emotions. Often the motivating force to maintain that
level of order and that level of affective constraint is to avoid the experience of anx-
iety; when these strategies are working reasonably well (i.e., for most of the gen-
eral population) the T-scores for the two scales will be comparable. However, indi-
viduals instructed to “fake good” will portray themselves as exceptionally orderly
without any accompanying anxiety; they place a premium on control, but the loss
of this control does not concern them in the least. Among such respondents, 74%
report this level of discrepancy between their need to maintain order and anxiety
around order, whereas only 23% of normal respondents demonstrate a compara-
ble split.
7. DOM minus AGG-V 2 I5T (1 point). The seventh item of the Defensiveness
Index (DEF) involves a 15-point discrepancy between reported levels of domi-
nance and the tendency to express anger verbally. The DOM scale often is ele-
vated in dissimulation samples, as it represents the ability to assume leadership
roles effectively. A person demonstrating this difference between DOM and AGG-V
is describing himself or herself as very effective in controlling other people with-
out ever having to raise his or her voice; they are such natural leaders that others
will follow their commands without any need for assertiveness on their part. This
quite desirable, but unlikely, combination of features appears six to seven times
more frequently in individuals “faking good” than it does in the general population.
8. MAN-A minus STR 2 10T (1 point). The last DEF item involves a person
reporting very high activity levels without any stress associated with this level of

140
 Identifying Defensiveness on the PAI

involvement. People with this pattern describe being involved in numerous impor-
tant activities, yet being able to function in an effective and controlled way amid
all this commotion. This configuration is three to four times more likely to occur
in “fake good” samples than in the general population; the appearance of this dis-
crepancy in a clinical population is also quite unusual.
Structure of Defensiveness Index (DEF) items. The Defensiveness Index can be
treated as an eight-item scale with a factor structure. A standard principal compo-
nents extraction with varimax rotation factor analysis, performed using the data
from both clinical and community respondents, yielded three highly converging
factors across the two samples. The first factor, which includes sizable loadings on
DEF Items 1, 2, and 5, involves a refusal to admit that any aspects of life are less
than optimal or to acknowledge negative elements about oneself. The second fac-
tor involves Items 6, 7, and, to a lesser extent, 5. In each of these items, the posi-
tive aspects of a personality trait are emphasized, and the negative aspects are min-
imized; there is a tendency to deny that some positive characteristics often can
have negative consequences in certain contexts. Thus, these items suggest that
there is a denial of the internal consequences of the respondents’ personality style;
high standards can lead to anxiety, and capability can lead to impatience, yet not
for these respondents. The third component, involving Items 3, 4, and 8, reflect
individuals who are more oriented to denying that some of their behaviors have
external consequences, ones that adversely affect other people. The problems being
denied here tend to be ones in the societal realm, whereas the second factor
involves more internal repercussions.

Defensiveness Index (DEF) Interpretation

As can be seen in Table 5-4, the average score for a “fake good” sample on the
Defensiveness Index (DEF) was 6.23, as compared to 2.81 for these features in the
normative community sample. Transformation of Defensiveness Index scores to T
scores based on the mean and standard deviation of this normative sample are
listed for convenience in Table 5-5; this table reveals that the “fake good” sample
obtained a score that was more than 2 standard deviations above the mean of the
community sample. However, a comparable sample obtained by Cashel et al. (1995)
obtained a mean score of only 4.27 on the Defensiveness Index, a value that lies
roughly 1 standard deviation above the norm. The results of the Cashel et al. study
raise questions about the sensitivity of the Index in samples coached for “believ-
ability” in being defensive and raise the possibility that the Defensiveness Index
may be a better inclusion sign for defensiveness than it is as an exclusion sign. Thus,
respondents with score of six or greater on the Index are most likely defensive;

141
PAI Interpretive Guide

Table 5-5
T-Score Equivalents for the PAl Defensiveness Index (DEF)
Standardized Against a Community Normative Sample
DEF score T-score equivalent
32
38
45
51
58
64
71
78
84
CO
(O
9
Ne
C2)
w=
Ol
COMO) 91

Note. N = 1,000.

however, defensive people may not score high on DEF, and scores within normal
limits should not be considered to rule out the possibility of a dissimulated proto-
col. Nonetheless, the Defensiveness Index has promise and merits further study as
a tool for addressing this most difficult of assessment issues.
The Appendix details numerous correlates for the Defensiveness Index, of
which selected results are presented in Table 5-6. This table reveals that the Defen-
siveness Index (DEF) is positively correlated with other indicators of defensiveness
and/or socially desirable responding. DEF correlates moderately with PIM, which
should be expected, as PIM elevations comprise part of the Defensiveness Index.
However, the correlation with the MMPI L scale is nearly identical to that found
with PIM, which provides an independent verification of the positive dissimulation
content of the Defensiveness Index items. Within the realm of clinical constructs,
DEF is negatively associated with depressed mood as rated by an independent
observer on the Brief Psychiatric Rating Scale (BPRS; Overall & Gorham, 1962).
This finding is interesting, because it may suggest that the self-portrayal of positive
mental health reflected in the DEF items can be convincing to others. The Defen-
siveness Index also displays a strong negative relationship with somatic com-
plaints, indicating that respondents with elevated DEF scores are as unlikely to
complain about their physical health as they are to complain about their mental
health. In summary, the Defensiveness Index appears to be tapping an element of
positive dissimulation related to a presentation of various unlikely virtues. In com-
bination with other features such as PIM and the CDF (Cashel et al., 1995) score,

142
 Identifying Defensiveness on the PAI

Table 5-6
Selected Correlates of the PAI
Defensiveness Index (DEF) Total Score
eee

Correlation with
Variable description DEF score
PAI PIM 56
MMPI L As)
MMPI K ‘25
Marlowe-Crowne Social Desirability Scale .28
CDF score (community sample) OZ
PAI NIM —.18
PAI MAN-G 52
PAI RXR D2
Brief Psychiatric Rating Scale, Depressed Mood (BPRS) —.49
Wahler Physical Symptoms Inventory —.79
Note. MMPI = Minnesota Multiphasic Personality Inventory.

the Defensiveness Index can help with the particularly difficult challenge of iden-
tifying profiles resulting from overt attempts at positive impression management.

143
 c-VAM (a9
FR iA
maw) girient} 2 eed!) shiurtves lana
tornyo leavers itsw
_——_ —_ . — ~ = - — ———. —
a _ —

igi iV 7 ak Jiu sioaentiiM = (Si eit

i }
ater indy eer

Sant 'n Sallis Woe Yel psi Hj a irl eabetl acca >
reas Meties t Peers “iru 4 fi Bs
AP eTaAArettg! I" vir ane — ae
. - ‘ es
j au << ei

[ie Ape ene: arta omens stnialea Ler ote, DSleninseess

ie ost Abs are outiire!inTake St Th eile wage sa
shsiieesvedic gre! dix iy. eptay lantBoe ite
- em nie —— bear
_ re a

an te Set
| | 7
 CHAPTER 6
USE OF THE PAI IN DIAGNOSIS

There are a variety of ways to use the PAI in deriving diagnostic hypotheses,
all of which rely on the configuration of the PAI profile. The profile configuration
represents the highest interpretive level of the instrument, and traditionally, the
premise behind multidimensional inventories such as the PAI has been that the
combination of information provided by the multiple scales is greater than any of
its parts. The following sections discuss some of the major diagnostic classes of
mental disorder and aspects of the PAI profile configuration that are useful for
assigning such diagnoses. The sections discuss hypotheses drawn from three
primary sources of information: (a) mean profiles; (b) actuarial functions; and
(c) configural decision rules. For example, with respect to mean profiles, the PAI
manual presents the average profiles derived from 24 different groups isolated on
the basis of a particular diagnosis. The following sections describe these profiles
and, where available, present similar information from other studies. However,
mean diagnostic profiles are limited in the context of interpreting the PAI, because
they do not represent a “prototypic” profile for a diagnosis; rather, they present the
“lowest common denominator” for the diagnosis. Because of the extensive comor-
bidity among emotional disorders and variations in diagnostic practice among
clinicians, the resulting mean profile for a given diagnosis may not fully capture
the elements of the PAI that most reflect that disorder. Thus, the mean profile is
only a beginning point in understanding the relationship between diagnoses and
profile configurations.
Various analyses have also been conducted to identify actuarial decision rules
for diagnostic assignment. In one example of such efforts, LOGIT analyses (Finney,
1971) were performed to construct models of diagnostic decisions provided by
clinicians on patients who completed the PAI. These LOGIT functions were incor-
porated into the PAI Software System (Morey, 1991) in an attempt to realize the
promise of computerized actuarial interpretation, and these functions are dis-
cussed in the sections of this chapter where such analyses have been performed.
In these functions, the probability that a person carries the given diagnosis is esti-
mated by solving the function provided and including that value in the following
formula:
exp[2(x,,—5)]
Probability of diagnosis = L+exp[2(x,—-5)]

45
PAI Interpretive Guide

where x,, is the result of the function described for the diagnosis and exp is the
exponential of the bracketed expression. However, only diagnoses with adequate
numbers of respondents (at least five for each predictor variable) were investigated
using LOGIT analyses, and the calculations are sufficiently complex to hinder use
of such functions in routine clinical contexts other than their incorporation into
the interpretation software. Nonetheless, the composition of such functions is
often informative in illustrating important points about the diagnosis and its PAI
profile. One important aspect of these functions is that they involve contrasts
between a particular diagnostic group and the clinical respondents as a whole
(rather than comparing the group to normal controls). Thus, the parameters of
these functions can be useful guides to discriminating among different clinical
groups, facilitating the discriminant validity of any resulting diagnoses.
Finally, DSM-based configural rules also have been developed for a number of
the DSM-IV (APA, 1994) diagnostic categories; these rules were designed to match
the DSM-IV criteria with corresponding constructs on the PAI and also were incor-
porated into the interpretation software. The primary scales used in these rules are
described in the text and summarized in tables for each of the following major
diagnostic categories. It should be recognized that these decision rules were based
on trends in the standardization samples of the PAI, and there is a clear need for
cross-validational research. However, these rules are in keeping with the nature of
the symptomatology specified in the DSM-IV and, as such, they provide a useful
starting point in identifying particular disorders.

Depression and Related Disorders

In thinking about depression, the obvious place to begin diagnostic consider-
ation is the DEP scale, which is described in detail in chapter 2. A diagnosis of
depression in the absence of some elevation on DEP is unlikely, unless defensive-
ness is distorting the profile. However, there are a number of different disorders in
which depression is a core element, and examination of profile configurations can
help to sort out these different diagnostic possibilities. For example, Figure 6-1
presents the mean profiles (originally presented in the PAI Professional Manual
[Morey, 1991]) for three disorders in which depressed affect is prominent: Adjust-
ment Disorder, Dysthymic Disorder, and Major Depressive Disorder. The figure
demonstrates that the mean profiles for the three disorders are similarly shaped,
but there are interesting configural differences between them. The following para-
graphs offer suggestions for the identification of these three diagnostic groups.

146
 PROFILE FORM FOR ADULTS -SIDEA 1 2 3
- 4 5 6 7 8 9 10 EGE A B c D E Y
- - | -
70=3 70-
:= 70-5
-
70-
-
_60-== 2 ; -
o——e Adjustment
- S | 65:
¥ 65 65-
= 65 = 35-
: = Z : wes : 55
: =
i a =
eo:
seeeee
teat
ee *® Dysthymicj Disorder
1s- ae oe 60 60- - 60 2 = = x E | = ae, ee eee
= ns. Gussz Eee = 65 Ee hee Lc. isorder
55- = = : = 3 = = <t
= = 55- = = -
par Bae 3 a VEs = pce = 5
er: : : z osBy |} 8°= = pa 28 leesz =
= 50- = E 2 ame, = 2
E ae = eS S| fe . ee . |
= SOS rs = 2 50 20.

10- 40- ~ =
Sie ee

Oi » 30== |
rm =
ri] - - ¢ E
Pale
”n =
|
ES =
60 —
= |
=
|

50 = =

= ie = |
= = ace = =
= = 10- = J o-
40 — = eS aE De
= = = = : : 2 |
= - : = = 40- = = o- |
- o- - o- = a = ce S
: c RE 3 = o- o-
30 = = 5- =
= = Z ce

5 ee =
= ya
S
o- = F =
4 2 3 4 5 6 7 8 9 10 u A e c D E Y z
ICN INF NIM PIM SOM ANX ARD DEP MAN PAR scz BOR ANT ALC DRG AGG sul STR NON RXR DOM WRM

PROFILE FORM FOR ADULTS - SIDE B

110 — 24- Z
2 : Fe ; ite 24- ie o—e Adjustment
: 24- ae : Disorder
ae ind E 2 24- 2
; 20- : = @--+++--@ Dysthymic Disorder
100 2 “ ‘
aes = Pe TEL = YHA P - F
F : @—-—-~-e Major Depressive
7 gee z oe ae | Disorder
20-
= - = - 4 15.
3 = 24-
90 20- = = 20-
< 18- 18- a 90

24- 7 pies = - 15- -

24 > ize - 20- |45- ae = - = =
: 4 ‘ere : 3 = = a = 20-
80 = = . eee 4 =e AS eae: Bales = B80
20- 20- E = % c 15-

15= > 5 = = _ 10 4 2 _
- = be ES - =
Z 15- 15- z BES ac =
= = = i— e —}410- =< = = (=e== := === = E 2— 70
Fd 5
~ = 15- A " 4 = : e
Se 15 e 2 = =~ <3
a = fe z = 10- - 3
EO fe - = ~ |)40- = =
60 = 10- = q = ee 2 = = 2 = ee
. E 10, 4, :
= : ¥ E Spee =

50 — =
~t-
= =
cake =, =
=
= as = - ee 2 es - - ae 2 = = = = oes ae S
e- = = a. = a ee ek = ne c 7 ie. AF Se -
= - € ‘ = 2 mn = 7 = ‘ = A y — 40
ie 4 s a g oO . - ‘ - = 0 by 2 = - oe 0 - - =

g os = ‘ oa o- = j _ = 10 Paes “ee sc al ne a
= oO = - - - 0. o- = :

30: Sete aes x o- Pa = 30

SOMC SOMS SOMH ANX-C ANXA ANXP ARD-O ARD-P ARDT DEP-C DEP-A DEPP MANA MANG MANJ PARH PARP PARR SCZP SCZS SCZT BORA BORI BORN BORS ANTA ANTE ANTS AGGA AGGV AGGP
CONV SOMA HEAL COG AFF PHYS OBS PHOB TRAU COG AFF PHYS ACT GRNO IRRI HYPE PERS RSNT PSYC SOG THGT AFF ID NEG SELF ANT EGO STIM AGG VERB PHYS.
SYMP ZATN CONC SYMP SYMP SYMP COMP IAS STRS SYMP SYMP SYMP LEVL |OSY BLTY VIG CUTN MENT EXP DET DIS INST PROB REL HARM BEH CEN SEEK ATT AGG AGG

Figure 6-1. Mean profiles for Adjustment Disorder, Dysthymic Disorder, and Major Depressive
Disorder samples on the PAI (Morey, 1991).

147
PAI Interpretive Guide

Adjustment Disorder
The cardinal feature of an Adjustment Disorder is the development of clinical
symptoms (typically involving neurotic spectrum features of depression and/or
anxiety) in response to some psychosocial stressor or stressors. On the PAI, the
adjustment reaction profile presented in Figure 6-1 demonstrates some elevation
on the “neurotic” scales, including DEP, yet the greatest single elevation on the
profile is the STR scale. This comparison of STR to the clinical scales is a useful
means to explore the possibility of an adjustment disorder. If the primary problem
that the person seems to be identifying involves aspects of his or her environment
(represented by STR elevations), this suggests that the respondent's primary con-
cerns are external and, perhaps, crisis related; the lack of suppression on MAN-G
supports the conclusion that self-esteem has not suffered in the face of these envi-
ronmental events.
One distinguishing feature of Adjustment Disorders is that they are acute in
nature; the DSM-IV specifies that symptomatology begins within 3 months of the
onset of the stressor and lasts no longer than 6 months after the stressor or its con-
sequences have ceased. In contrast to Adjustment Disorders, Posttraumatic Stress
Disorder can have a delayed onset and a far more enduring course. Adjustment
Disorders can have elevations on ARD-T if the recent stressor was extremely severe.
However, Adjustment Disorders can develop in response to a wide array of stres-
sors, and it is more typical that STR is elevated to a greater extent than ARD-T in
this diagnostic group. Also, posttraumatic stress conditions have a particular con-
stellation of features (described in a later section), whereas the psychological
sequelae of Adjustment Disorders are typically limited to the first four clinical
scales: SOM, ANX, ARD, and DEP.

The DSM-IV lists a variety of subtypes of adjustment disorder, subdivided

according to the symptomatic reactions to the stressors experienced. The primary
subtypes include the following:
Adjustment Disorder With Depressed Mood, used when the predominant mani-
festations are unhappiness, tearfulness, or feelings of hopelessness. Primary
PAI indicators would include moderate elevations on DEP-A and DEP-C.
Adjustment Disorder With Anxiety, used when symptoms such as nervous-
ness, worry, or fearfulness are predominant. On the PAI, such features
would be indicated by moderate ANX-A and ANX-C elevations.
Adjustment Disorder With Disturbance of Conduct, where features involving
violations of norms and rules, or of the rights of others, are seen. Elevations
on ANT-A and AGG-P would be expected to accompany such behaviors.

148
 Use ofthe PAI in Diagnosis

The DSM-III-R listed an additional category where the primary symptomatic

reactions to the stressors involved physical complaints; this pattern has been rele-
gated to a residual (“Unspecified”) category in DSM-IV. If observed, the PAI equiv-
alent of such features would be most likely to involve SOM-S. However, if some
repression of psychological involvement with the stressors is occurring, ANX-P
and/or DEP-P also might display’moderate elevations in this constellation.

Dysthymic Disorder
The hallmark feature of Dysthymic Disorder is a chronically depressed mood
occurring most days over a 2-year period. It can be distinguished from Major
Depressive Disorder by a relatively milder severity of symptomatology, the stabil-
ity and chronicity of its course, and by its early and insidious onset (often in ado-
lescence, or even in childhood). Because of the centrality of depressed mood and
its chronic nature (rather than being a reaction to external events), DEP plays a
more prominent role than STR in identifying dysthymia as compared with adjust-
ment disorders. This can be seen from the mean profile for Dysthymic Disorder in
Figure 6-1; the DEP scale (as well as the neurotic level scales such as ANX, ARD,
and SOM) are all at elevations comparable to or beyond that of STR. Also, SUI has
begun to elevate, as the issues are beginning to become more internal than exter-
nal in focus.
Although depressed mood is central in Dysthymic Disorder, this group does
not typically meet the full criteria for Major Depressive Disorder. Indeed, the dys-
thymia must be present for 2 years in the absence of meeting the Major Depressive
Disorder criteria in order for the diagnosis to be assigned. Most typically, the phys-
iological features of depression are the symptoms absent in dysthymia; because the
depressed mood is central to the disorder, DEP-A is invariably elevated. Thus, a
difference between DEP-A and DEP-P, with the former 10T or more above the lat-
ter, is often an indicator of Dysthymic Disorder.

Major Depressive Disorder

The diagnosis of Major Depressive Disorder reflects a constellation of
depressed mood, vegetative signs and symptoms, and low self-esteem and help-
lessness. This disorder is one of the more common psychological conditions, with
lifetime prevalence rates estimated from 10% to 20% of the general population. It
is a serious emotional condition; up to 15% of patients with severe depression
commit suicide, and it is also associated with dramatically increased death rates
among older individuals. The seriousness of the mental state is reflected in the
major depression profile in Figure 6-1, which displays an overall elevation higher

149
PAI Interpretive Guide

than that of either of the previously mentioned disorders. The profile is character-
ized by marked elevations on DEP and SUI; these two scales emerge consistent
with the greater prominence of depressive symptoms over the more general neu-
rosis and personality features typical of the lower grade, more chronic dysthymic
condition. Also noted with the Major Depressive Disorder diagnosis is greater
involvement of social withdrawal and greater indifference (SCZ-S) and cognitive
inefficiency (SCZ-T) than is found in the milder conditions.

The LOGIT function for the diagnosis of Major Depressive Disorder (as con-
trasted with other clinical categories) reveals empirical confirmation of many of
these indicators. This function (presented primarily for illustration purposes,
rather than for routine clinical use) is as follows:

.016(DEP-A) + .011(DEP-P) + .007(SUI) + .002(DEP-C) — .015(MAN-G) + 2.564

This function highlights the central role of the aforementioned scales in

arriving at the diagnosis of Major Depressive Disorder. The weight for DEP-C is
relatively low primarily because of that scale’s high correlation with DEP-A, sug-
gesting that the two scales contribute considerable overlapping information in the
identification of depression. However, the relatively large negative loading of
MAN-G demonstrates that low scores on this scale contribute important diagnos-
tic information independent of scores on any of the DEP subscales. Finally, the
large constant at the end of the function serves as a reminder that this is a high
base-rate diagnosis, and the odds that any given patient has a diagnosis of Major
Depressive Disorder are relatively high.
The profile distinctions between Adjustment Disorders and Dysthymic Disor-
ders that were made earlier also can help to distinguish between endogenous and
exogenous forms of depression. Differences may be found in the nature of the
symptomatology and also search for evidence of marked external stressors. In
endogenous depression, one often finds a preponderance of vegetative signs of
depression, including sleep and appetite problems, lack of energy, and apathy and
indifference. Telltale features here would involve marked elevations on DEP-P
(with T scores at least equal to, if not higher than, the full DEP score) and the sup-
pression of MAN-A, with a T-score here expected to be below 40T. For exogenous
depression, indicators of external stressors should predominate. For example,
elevations on STR or ARD-T, or both, would be indicators. It is worth noting that
global elevations on these two scales are very common among individuals pre-
senting for treatment. As such, the focus should be on the relative elevation of
these scales, relative to the full DEP T-score. In exogenous depression it would be
expected that these scores would be at least as high as DEP.

150
 Use ofthe PAI in Diagnosis

Table 6-1
Summary of Key PAI Diagnostic Indicators for Depressive Disorders
oe a eel etree
Diagnostic Elevation Suppression
consideration indicators indicators
Adjustment Disorder STR and/or ARD-T > 60T Full clinical scales < 70T
Check SOM, ANX, DEP for
nature of symptoms
Dysthymic Disorder DEP-A, DEP-C > 70T DEPP Oin
BOR > 65T with BOR-A SCZ-T < 70T
predominant

Major Depressive Disorder DEP-A, DEP-P, DEP-C, SUI MAN-G < 45T
all typically > 70T MAN-A often low, can
SCZ-T, SCZ-S often elevated be moderately elevated
in severe depression with agitation

Mean profiles for two additional relevant samples are presented in Figure 6-2:
a sample of 47 women patients with “primary affective disorder” (Major Depres-
sive Disorder or Dysthymic Disorder) reported by Cherepon and Prinzhorn (1994)
and a sample of 28 women patients in treatment for depression who had Beck
Depression Inventory scores above 14 and MMPI-2 Scale 2 scores above 64T
(Gaies, 1993). The profiles from these two studies are quite similar and reflect the
mixed composition of the sample, as both studies apparently included patients
with Major Depressive Disorder as well as patients with Dysthymic Disorder. How-
ever, the configural similarities of these profiles to those presented in Figure 6-1
reveals that the respondents were clearly demonstrating complaints within the
affective realm.

Anxiety Disorders
Nearly all clinical disorders share anxiety as a feature, and the group of condi-
tions known as “anxiety disorders” span such diverse conditions as panic attacks,
compulsive hand washing, and posttraumatic reactions to intense life stress. At the
level of individual scales, ANX and ARD (described in detail in chapter 2) both
provide critical information about these conditions. ANX is generally a nonspe-
cific, global measure of anxiety that, as is the case with the construct in question,
could be prominent in a number of different clinical conditions. ARD, on the other
hand, presents behavioral information that is more closely tied to specific anxiety
conditions. The following paragraphs provide suggestions for using these and
other scales in identifying particular anxiety-related conditions.

Lil
PAI Interpretive Guide

FOR ADULTS-SIDEA 14 2 3 4 5 6 7 8 9 10 44 A B c dD E Y Zz
PROFILE FORM

@--++++-@ Depression
(Gaies, 1993)
o——e Primary affective
disorder (Cherepon
& Prinzhorn, 1994)

PP
LCR
Sn
1 2 3 4 5 6 7 8 9 10 a4 A 8 c D E ¥ z
ICN INF NIM PIM SOM —ANX ARO DEP = MAN PAR SCZ BOR ANT ALC ORG AGG sul STR NON RXR DOM WRM

PROFILE FORM FOR ADULTS - SIDE B

@-------@ Depression
(Gaies, 1993)
eo—— Primary affective
disorder (Cherepon
& Prinzhorn, 1994)

reypereupes

Pe
Cn

Preeepeereprerrpreeep
SOMC SOMS SOMH ANKC ANKA ANXP ARD-O ARDP ARDT DEPC DEPA DEPP MANA MANG MANI PARH PARP PARR SCZP SCZS SCZT BORA BORI BORN BORS ANTA ANTE ANTS AGGA AGGV AGGP
CONV SOMA HEAL COG AFF PHYS OBS PHOB TRAU COG AFF PHYS ACT GRND IRRI HYPE PERS RSNT PSYC SOC THGT AFF ID NEG SELF ANT EGO STIM AGG VERB PHYS
SYMP ZATN CONC SYMP SYMP SYMP COMP IAS STRS SYMP SYMP SYMP LEVL IOSY BLTY VIG CUTN MENT EXP DET DIS INST PROB REL HARM EH CEN SEEK ATT AGG AGG

Figure 6-2. Mean PAI profiles for depression and primary affective disorder samples.

152
 Use of the PAI in Diagnosis

Phobias
The essential characteristic of a phobia is an intense and persisting fear of
clearly identifiable objects or situations, or both. The individual generally recog-
nizes that the fear is excessive relative to the actual threat, although this recogni-
tion does little to circumvent his or her anxiety. The fear will generally appear
immediately on exposure to the situation, or it can emerge in anticipation of such
an encounter; such fear can be sufficient to precipitate a panic attack in severe
cases. In order to be a diagnosable condition, the fear and associated avoidance
behaviors must interfere with functioning or cause marked distress.
The DSM-IV lists a variety of types of phobias, subdivided according to the par-
ticular objects or situations that precipitate the anxiety. Two major subtypes
include the following:

Specific Phobia
This disorder (known as “Simple Phobia” in DSM-III) involves marked anxiety
reactions to circumscribed objects or situations. Among the most common of these
phobias are those involving animals (e.g., snakes, insects), natural environment
(e.g., storms, heights), blood/injection/injuries (often accompanied by fainting), or
situations (e.g., enclosed spaces, flying, public transportation). Obviously, the first
place to look for elevations with these conditions is ARD-P, which inquires directly
about some of these situations. However, in order to be a diagnosable condition,
there must be sufficient impairment or distress associated with the situation or its
avoidance, and other scales should be examined for indicators of this distress. The
most typical elevations will involve ANX-A, a sign that the anxiety has become
debilitating and somewhat generalized, and ANX-C, a sign that the fear has
become somewhat of a ruminative preoccupation.

Social Phobia
The critical feature of this disorder is an intense and persistent fear of social or
performance situations that might result in embarrassment or humiliation. Fears
of certain social situations, such as public speaking, are quite common in the gen-
eral population, but the anxiety must be interfering with social-role functioning or
causing marked distress in order to receive the diagnosis. Nonetheless, this is
perhaps the most prevalent phobia, with prevalence estimates ranging from 5% to
10% of the general population. Again, the starting point for identifying social pho-
bias is with ARD-P, which inquires directly about such anxieties. However, other
sources of information about social impairment are also useful. Low scores on WRM
are often observed, demonstrating a lack of effectiveness in social situations; DOM
also is generally low, because serving in a leadership role subjects an individual to

Lek,
PAI Interpretive Guide

a great deal of public scrutiny. As with the specific phobias, signs of impairment
and distress are likely to appear on ANX-A and ANX-C. The DSM-IV includes an
additional specifier of a “Generalized” form of social phobia, where the fears are
related to most forms of social interactions. Such persons are particularly likely to
show deficits in social skills and to have resulting social and occupational impair-
ments. In addition to the features noted above, this more generalized form may be
expected to show elevations on SCZ-S, documenting the lack of interpersonal
skills, accompanied by ANX-A, showing that the interpersonal withdrawal is due
more to anxiety than to a lack of interest in relationships.

Panic Disorder
The defining feature of this disorder is the presence of recurrent panic attacks
not tied to some situational trigger. A panic attack is a discrete period of very
intense fear with a variety of autonomic nervous system features, such as heart pal-
pitations, sweating, chest pain, dizziness, numbness or tingling, shortness of
breath, and chills or hot flashes. Cognitive signs also may be present and typically
involve fears of dying, losing control, or going crazy during the attack. Following
the attacks, such individuals have persistent concerns about the implications of
having another such attack and may show significant behavioral change in order
to prevent or control such a possibility.
The features of the panic attacks are largely physiological in nature. Therefore,
the ANX-P scale is particularly useful in identifying this condition, as it inquires
directly about a number of panic symptoms; marked elevations on ANX-P are par-
ticularly suggestive of disorders with panic attacks. However, elevations on the
other two subscales of ANX are also likely; ANX-C will elevate as the person begins
to ruminate about the recurrence of panic attacks, whereas ANX-A will reflect the
apprehension surrounding the unpredictability of their occurrence. DEP elevations
also are common, as over half of individuals with Panic Disorder will experience a
major depressive episode at some point in their lives. Some patients fear that the
attacks may indicate the presence of an undiagnosed, life-threatening illness, such
as a heart condition or a seizure disorder. These beliefs often will lead to elevations
on SOM-H, but not necessarily on SOM-S or SOM-C, as the panic symptoms are
not directly consistent with the physical complaints reflected on those scales.
Panic Disorder can often lead to generalized avoidance behaviors known as
agoraphobia. The DSM-IV outlines a subtype known as Panic Disorder with Ago-
raphobia. The agoraphobia involves anxiety about being in places or situations
from which escape might be difficult or where help may not be available should
the person experience a panic attack. This results in a pervasive avoidance of any

L354
 Use ofthe PAI in Diagnosis

place in which the person does not feel safe, and, in its most extreme form, the
person will not leave his or her home, and even then may refuse to remain home
alone. The presence of agoraphobic features adds a number of elements to the PAI
profile, in addition to those described earlier for panic disorder. First, ARD-P is
typically highly elevated, because the person is fearful of a wide array of situations.
Also, the person is extremely apprehensive about novelty and unpredictability in
life, and, as such, tends to obtain scores on ANT-S that are considerably below the
mean. Finally, such people often rely heavily on companions to help them deal
with feared situations. The agoraphobic’s dependence on such companions is
reflected in low DOM scores, with WRM scores more likely to be in the average-
to-high-average range.

Obsessive-Compulsive Disorder
The Axis I diagnosis of Obsessive-Compulsive Disorder is characterized by
recurrent or intrusive thoughts, impulses, images, and behaviors that are time con-
suming or a source of distress for the individual, or both. By definition, the person
realizes (or has realized at some point) that the obsessions and/or compulsions are
excessive, inappropriate, or unreasonable, although the level of insight varies.
Impairments can be identified in many areas; the obsessions can interfere with
cognitive tasks, and the disorder can lead to avoidance of situations that provoke
the intrusive thoughts or compulsive behaviors. For example, a person with obses-
sions about germs will avoid public restrooms or shaking hands with strangers.
The ARD-O scale is the beginning point for an investigation of this diagnosis,
as many of the questions on this scale inquire directly about obsessions and com-
pulsions. Because the disorder is generally a source of marked distress, the ANX
scale is typically elevated as well. Among the ANX subscales, ANX-C is the most
characteristic of the disorder, as it captures the rumination and uneasiness of the
obsessional individual. Other elevations may be seen as a function of some of the
problems often associated with this disorder. SOM elevations may be obtained, as
obsessive-compulsive individuals often have hypochondriacal concerns. Because
patients with this diagnosis often have overwhelming guilt and sleep disturbances,
DEP elevations are fairly common as well.

Posttraumatic Stress Disorder

This disorder (abbreviated PTSD) involves characteristic symptoms that some-
times emerge following the experience of an extreme traumatic stressor that leaves
the person intensely afraid, horrified, or helpless. Such events might include mil-
itary combat, violent personal assaults, severe automobile accidents, or witnessing

155
PAI Interpretive Guide

an event that involves death or injury to another person. The symptoms typically
begin within 3 months of the trauma, although there may be a delay of months or
even years before symptoms appear. It appears that the severity, duration, and
proximity of the patient's exposure to the trauma play the largest role in the devel-
opment of the disorder and in the severity of the resulting symptomatology.
The mean profile of a group of patients diagnosed with PTSD (originally pre-
sented in the PAI Professional Manual [Morey, 1991]) is presented in Figure 6-3. As
would be expected, the most striking aspect of the profile involves the marked ele-
vation on ARD-T, which makes direct inquires about the existence of traumatic
stressors. Although many clinical groups demonstrate ARD-T scores above 7OT,
PTSD patients will typically score at least 80T on this scale, and elevations above
9OT are not uncommon in this group. However, the diagnosis of PTSD should not
be based solely on an ARD-T elevation; there are a number of other features of the
profile in Figure 6-3 that are consistent with the characteristic symptomatology of
PTSD. For example, the DEP scale reflects a variety of symptoms associated with
PTSD. Individuals with PTSD often describe painful guilt feelings associated with
the experience, leading to DEP-C elevations. Also, recurrent distressing dreams of
the event are diagnostic, leading to sleep disturbance and subsequent DEP-P ele-
vations. Finally, diminished interest in significant activities are associated with
elevated DEP-A scores.
Figure 6-3 also includes the mean profile for 44 women psychiatric patients
who reported a history of childhood abuse (physical or sexual), adapted from a
study by Cherepon and Prinzhorn (1994). This profile is quite similar to the one
obtained by Morey (1991), with a few interesting differences. In particular, the
Cherepon and Prinzhorn profile displays a greater elevation on the SUI scale, and
WRM is greater than DOM (unlike the profile from Morey). These differences may
suggest differences in the manifestation of PTSD related to gender; all of the
Cherepon and Prinzhorn respondents were women, whereas many of the patients
obtained by Morey (1991) were male veterans with combat-related PTSD. None-
theless, the similarities between the profiles obtained from these two very different
samples are striking.
A variety of other features of the PTSD syndrome are also directly reflected in
PAI scales. DSM-IV symptoms of PTSD include (a) physiological anxiety reactivity,
reflected on ANX-P; (b) feelings of detachment or estrangement from others, man-
ifest on SCZ-S and low WRM; (c) hypervigilance, evidenced in PAR-H elevations:
and (d) irritability, which can be gauged using MAN-I. Difficulty in concentration
and hazy recall surrounding the event often lead to prominent SCZ-T elevations,
as seen in Figure 6-3, and, at the full scale level, SCZ elevations are far more likely

156
 Use of the PAI in Diagnosis

PROFILE FORM FOR ADULTS-SIDEA i 2 3 4 5 6 7 8 9 10 da A B Cc 12) E Y Z

e————* Abused psychiatric

patients (Cherepon
& Prinzhorn, 1994)
@=—=-~-@ PTSD (Morey,
1991)

il &

rerives

peseTeeuererertireeres
°i

drt
or
cel
lap
hioee
iit

Teena

1 2 3 4 5 6 7. 8 9 10 Ee A 8 c o E
ICN INF NIM PIM SOM ANX ARD DEP MAN PAR scz BOR ANT ALC DRG AGG sul STR NON RXR DOM WRM

PROFILE FORM FOR ADULTS - SIDE B

e— Abused psychiatric
patients (Cherepon
& Prinzhorn, 1994)
@=——-@ PTSD (Morey,
1991)

ee

myer

ee

Vepreeepecrepereege

ee
DEPP MANA MANG MANI PARH PARP PARR SCZP SCZS SCZT BORA BOR BORN BORS ANT-A ANTE ANT-S AGGA AGG-V AGGP
'SOMC SOM-S SOMH ANKC ANXA ANXP ARD-O ARDP ARD-T DEP-C DEP-A
GRND IRRI HYPE PERS RSNT PSYC SOC THGT AFF 10 NEG SELF ANT EGO STIM AGG VERB PHYS
CONV SOMA HEAL COG AFF PHYS OBS PHOB TRAU COG AFF PHYS ACT
SYMP LEVL IOSY BLTY ViG CUTN MENT EXP DET DIS INST PROB REL HARM BEH CEN SEEK ATT AGG AGG
SYMP ZATN CONC SYMP SYMP SYMP COMP IAS STRS SYMP SYMP

Figure 6-3. Mean PAI profiles for psychiatric patients reporting childhood abuse and patients diag-
nosed with PTSD.

157
PAI Interpretive Guide

to be seen in PTSD than in any other anxiety disorder. Finally, outbursts of anger
are also diagnostic, and the V-shaped pattern on the AGG subscales (i.e., AGG-A
and AGG-P elevated and AGG-V low) shown in Figure 6-3 reflects both the under-
lying anger and the tendency for it to be expressed as “outbursts” (i.e., anger is
expressed suddenly and explosively).
The LOGIT function for the diagnosis of PTSD (as contrasted with other clin-
ical categories) provides additional support for many of these observations. This
function (presented primarily for illustration purposes, rather than for routine
clinical use) included the following weights for PAI variables:

051(ARD-T) + .001(ANX-P) +.001(DEP-P) + .006(SCZ-T) — .010(BOR-A) + .245

This function highlights the central role of the previously mentioned scales in
arriving at the diagnosis of PTSD, particularly the central role of ARD-T. The inclu-
sion of the physiological features of depression and anxiety and the cognitive inef-
ficiency reflected in SCZ-T demonstrate unique contributions beyond their associ-
ation with ARD-T. The negative weighting of BOR-A shows it to be a suppressor
variable, as the scale itself tends to be elevated in individuals with PTSD, as seen
in Figure 6-3. However, the finding of suppression suggests that many individuals
display PTSD-like symptoms associated with personality pathology, and that
obtaining the PTSD constellation of features in the absence of such pathology is
particularly informative to the diagnosis.

Generalized Anxiety Disorder

The defining feature of this disorder is excessive anxiety and worry that is per-
sistent, is generalized across situations, and lasts for at least 6 months. The diag-
nosis is somewhat of a residual category, as anxiety of a more specific nature (e.g.,
phobias, posttraumatic stress, or obsessions) results in the corresponding diagno-
sis; Generalized Anxiety Disorder is only to be used if a person is anxious, but does
not meet criteria for such specific disorders. Despite this exclusion criterion, the
disorder is apparently fairly common, with a lifetime prevalence of roughly 5% in
the community.
Because the disorder involves rather nonspecific anxiety, the ANX scale is cen-
tral in establishing the Generalized Anxiety Disorder diagnosis, as it is a measure
of this type of anxiety. In particular, the ANX-A subscale captures the free-floating
nervousness and ANX-C taps the apprehensive expectation that characterize this
disorder. As an illustration, Figure 6-4 presents the profile of a group of 28 patients
diagnosed with Generalized Anxiety Disorder; these patients were a subset of a
group diagnosed with various anxiety disorders that was originally presented in

158
 Use of the PAL in Diagnosis

PROFILE
Tre
FORM FOR ADULTS - SIDE A 1- 2 3 4 5 6 7 8
= 9 10 oF A B C
Cc D E Y Zz
s 25 70== -= So - 70 60-==
a o—e Generalized
= =
- E = : 6 |
Anxiety
j

65 E 65 65 a
5 35 e | 30
Disorder
i
=
ae 0
70. 70 i - 60: =
== 15 15 60: = a
60 60- 606 2:
= - i 65 = 65 - 50-
Sy mesos Fe - zo- C— | =
es 55- |

S “ | ~ = 55- re
= ., fe 55- = 25 50
of = | 50
> = 50 | 7

= = aS 50 45 a
2 : {452 = 45-: | 20
20- 2
‘ie | = 50- : 20 a Tal 20
15 10 40= 203 40 =
2 : = | 45 2 40- al
= lees : :
=
2 25- ae ae 35 45.
‘ai 40: 35- - 40: = 15
= — s ss = 30- 15-
= 30- a =a = - ?
= = 2 35 oO
= a . = 35= 35- = 40 - - 30- 15- | 30 =

2 10 = 10- =. oo 82
3
== Se Be | 25 , eS 20:
| 30
<i

= 5 - = || | 2

> 15
5 L
o °
15
Ci
4 3
5 10
|
| 10
5-
Lovveliae | |
i z =
ICN INF NIM PIM ARD
5 6 7 8 ey 10 14 A 8 c o E v z
MAN PAR sez BOR ANT ALC DRG AGG sul STR NON RXR DOM WRM

PROFILE FORM FOR ADULTS - SIDE B

eo——e Generalized
Anxiety
Disorder

veebarer
dana

2 ro)

wo S Tovendacee

SOMC SOM:S SOMH ANX-C ANX-A ANXP —ARD-O ARD-P ARD-T DEP-C DEPP MANA MANG MANI PARH PARP PARR SCZP SCZS SCZT BORA BORI BORN BORS ANTA ANTE ANTS AGGA AGGV AGGP
CONV SOMA HEAL COG AFF PHYS OBS PHOB TRAU coc PHYS ACT GRND IRRI HYPE PERS RSNT PSYC SOC THGT AFF ID NEG SELF ANT EGO STIM AGG VERB PHYS
SYMP ZATN CONC ‘SYMP SYMP SYMP COMP IAS STRS SYMP SYMP LEVL IOSY BLTY VIG CUTN MENT EXP DET DIS INST PROB REL HARM BEH CEN SEEK ATT AGG AGG

Figure 6-4. Mean PAI profile for patients with Generalized Anxiety Disorder (Morey, 1991).

159
PAI Interpretive Guide

Table 6-2
Summary of Key PAI Diagnostic Indicators for Anxiety Disorders
Diagnostic Elevation Suppression
consideration indicators indicators

Specific Phobia ARD-P

ANX-C and/or ANX-A

Social Phobia ARD-P WRM < 45T

ANX DOM «< 45T
SCZ-S if “Generalized”

Obsessive-Compulsive ARD-O BOR-S

Disorder ANX, particularly ANX-C
DEP often elevated
SOM-H often elevated

Posttraumatic Stress ARD.-T, typically > 807 BOR-A at least 107 < ARD-T
Disorder ANX, particularly ANX-P
DEP, all DEP subscales
SCZ-T, SCZ-S
MAN-I!
PAR-H
AGG-P

Generalized Anxiety ANX, particularly ANX-A ARD < ANX

Disorder SCZ at least 10T < ANX-A

Panic Disorder ANX, particularly ANX-P ANT-S < 50T

DEP often elevated SOM-S, SOM-C < ANX-P

Panic Disorder with ARD-P ANT-S < 45T

Agoraphobia ANX, particularly ANX-A DOM < 45T

the PAI Professional Manual (Morey, 1991). This figure highlights the prominence
of ANX relative to other scales when this disorder is present. Although the Gener-
alized Anxiety Disorder profile, in general, is not characterized by marked eleva-
tions, there are other features incorporated into the DSM-IV diagnostic criteria that
might be elevated in particular patients. Among these are (a) sleep disturbances
and low energy levels that would manifest in the form of DEP-P elevations; (b) irri-
tability, evidenced on MAN-I; and (c) difficulty concentrating that might take the
form of a moderate SCZ-T elevation (i.e., approximately 60T).

160
 Use of the PAI in Diagnosis

Somatoform Disorders
The common characteristic of the group of conditions known as Somatoform
Disorders is the presence of physical symptoms that suggest some type of medical
problem but which are not fully explained by any diagnosable medical condition.
The physical symptoms are a significant source of distress and are not intentional
or under voluntary control (i.e., they are not malingered physical symptoms). The
disorders are often encountered in general medical settings as well as mental health
settings.
At the outset, it should be noted that no self-report test can adequately distin-
guish between “functional” and “organic” foundations of somatic complaints; such
distinctions, themselves, may have little meaning. What instruments such as the
PAI can do is present a picture of the role of somatic complaints in the overall psy-
chological makeup of the individual. In any instance where somatic concerns are
a prominent part of the clinical picture, a complete medical evaluation is far prefer-
able to a personality test in ruling out various organic bases for the conditions.
Such a ruling-out process is inherent in the diagnostic criteria for Somatoform Dis-
orders: for example, “After appropriate investigation, the symptoms cannot be
fully explained by a known general medical condition or the direct effects of a sub-
stance” (DSM-IV, p. 451). Any diagnostic guidelines for the PAI presented in the
following sections are predicated on the assumption that such appropriate medical
evaluation has been performed.
The overall discriminant validity of the PAI does make the test useful in assess-
ing emotional conditions within a general medical population. For example,
Osborne (1994) reported data on 105 general medical patients (73 women and 32
men) who were seen by internists within a multispecialty group practice. Patients
completed the computer-administered version of the PAI without difficulty. Mean
profiles for men and women medical patients are presented in Figure 6-5. This fig-
ure reveals that the SOM scale was generally the highest point of the PAI profile in
this population, although the average score on this scale was below 70T (Osborne,
1994). In fact, with the exception of SOM and mild indications of negative affect
among the women patients on DEP and ANX, no PAI scales obtained even a mean
score of 60T in the medical patients. Such results support the conclusion that
medical problems alone will not produce significant elevations on the PAI, and
that, when such elevations are noted, they most likely reflect associated emotional
issues rather than physical symptoms per se.
PAI Interpretive Guide

PROFILE FORM FOR ADULTS -SIDEA 1 oy © 33 Dee ees § 7. bese oer so a NB } Dd E

: = 7 70 60- e@—-—-e Female medical

c 85: | 2 patients
e al = 25 a e— Male medical
| : Bis 30 ; patients

90

ao

$a100S
|

6o

— 50

40

30

— 20
1 2 a 4 5 6 7 8 9 10 44 A B c D E y z
ICN INF NIM PIM SOM ANX ARD DEP = MAN PAR SCZ BOR ANT ALC DRG AGG SUI STR NON RXR DOM WRM

PROFILE FORM FOR ADULTS - SIDE B

@ ——~-e Female medical

patients
eo—— Male medical
patients

SOME SOMS SOMH ANX-G ANKA ANKP ARD-O ARDP ARDT DEP-C DEP-A DEPP MANA MANG MANI PARH PARP PARR SCZP SCZS SCZT BORA BORI BORN BORS ANTA ANTE ANTS AGGA AGGV AGGP
CONV SOMA HEAL COG AFF PHYS OBS PHOB TRAU COG AFF PHYS ACT GRND IRRI HYPE PERS RSNT PSYC SOC THGT AFF ID NEG SELF ANT EGO STIM AGG VERB PHYS
SYMP ZATN CONC SYMP SYMP SYMP COMP IAS STRS SYMP SYMP SYMP LEVL IOSY BLTY VIG CUTN MENT EXP DET DIS INST PROB REL HARM EH CEN SEEK ATT AGG AGG

Figure 6-5. Mean PAI profiles for male and female general medical patients who completed the
computer-administered version of the PAI (Osborn, 1994).

162
 Use of the PAI in Diagnosis

Conversion Disorder
The distinguishing feature of Conversion Disorders is the presence of symp-
toms or deficits that involve voluntary sensory or motor functioning and, hence,
are suggestive of some neurological disorder. Motor deficits might include paraly-
sis or localized weakness, loss of, coordination, or inability to speak; sensory symp-
toms could include blindness, deafness, or loss of feeling or pain sensations. As
with all Somatoform Disorders, these symptoms cannot be explained by any gen-
eral medical conditions. Such features typically emerge during periods of conflict
or stress and are thought to be psychological in origin, although they are not inten-
tionally produced.
The characteristic sensory-motor disturbances in Conversion Disorders are
directly addressed by the SOM-C scale. However, recall that SOM-C also will ele-
vate with actual neurological disorders, and interpretation of elevations as indicat-
ing Conversion Disorder should be advanced only after such general medical dis-
orders have been ruled out. There are a number of other features associated with
Conversion Disorders that can be identified using the PAI scale configuration. For
example, the feature known as la belle indifference involves a relative lack of con-
cern or distress about the symptoms and their implications; such indifference
might be reflected in a DEP-A score considerably below the level of health con-
cerns indicated by SOM-H. A certain degree of dependency and adaptation to the
“sick role” also may accompany conversion symptoms, resulting in low scores on
DOM and moderate-to-high scores on WRM. Often such individuals tend to min-
imize both personal distress and interpersonal conflict; in addition to WRM scores
that are generally above average, they tend to have low AGG scores and often are
above average on PIM.

Somatization Disorder
The central characteristic of Somatization Disorders is the recurrence of multi-
ple somatic complaints, with no known medical origin, over a period of several
years. The complaints typically involve multiple organ systems, such as gastroin-
testinal, sexual—reproductive, sensory-motor, and chronic pain, and the descrip-
tion of the symptoms can be quite dramatic, but vague and lacking in specific fac-
tual information. The somatic complaints have an onset prior to age 30 and
continue for several years, although no overt physical signs or structural abnor-
malities become apparent.
The items of the SOM-S scale directly tap the vague and diverse nature of
somatic symptoms in Somatization Disorder, and, as such, the scale represents
the starting point for establishing the diagnosis. Other associated features of the

163
PAI Interpretive Guide

Table 6-3
Summary of Key PAI Diagnostic Indicators
for Somatoform Disorders

Diagnostic Elevation Suppression

consideration indicators indicators

Conversion Disorder SOM-C DEP-A at least 107 < SOM-H

WRM > 50T DOM < 45T
PIM > 50T AGG < 50T

Somatization Disorder SOM-S

DEP-A and/or ANX-A
BOR-N and/or NON

disorder also can be investigated from the profile configuration. Prominent depres-
sion and anxiety are common, and these affects, in particular, can lead to eleva-
tions on DEP-A and ANX-A. The somatic complaints often seem to play a functional
role for the individual with this disorder, particularly in the interpersonal domain.
For example, a patient may receive some secondary gain by controlling a spouse
through repeated complaints of fatigue, illness, or malaise. However, the chronic-
ity of this behavior invariably leads to conflict with and/or resentment by family or
friends. Such a pattern can manifest in elevations on BOR-N or NON, or both.

Psychotic Disorders
The term psychosis involves a number of different aspects of mental status. In
the DSM manual, the term is used primarily to refer to conditions that involve
delusions, hallucinations, or grossly disorganized behavior. The term also has been
used variously to describe gross distortions in reality testing, or even any severe
impairment that severely interferes with life functioning. In the DSM-IV, the term
refers to Schizophrenia and related conditions, such as Schizoaffective Disorder
and Delusional Disorder. However, the following section also will discuss the diag-
nosis of Mania (although this is officially a mood disorder), because it is generally
more difficult to distinguish Mania from Psychotic Disorders than it is to distin-
guish it from other mood disorders, such as depression.

Mania
A Manic Episode is distinguished by a distinct period of unusually elevated,
expansive, or irritable mood. The mood can involve indiscriminate enthusiasm for
interactions with others, but irritability is perhaps even more common, particularly

164
 Use of the PAI in Diagnosis

when the person’s wishes are thwarted. Inflated self-esteem also is typical, with
grandiosity that can reach delusional proportions. The person will report
heightened energy and activity, with decreased need for sleep and excessive
planning and participation in multiple ventures. The thoughts and speech may
race faster than can be successfully articulated. Unwarranted optimism, self-
aggrandizement, and poor judgment often lead to unwise involvements in precar-
ious situations involving spending money, sex, or physical danger.
These diagnostic features are represented on the MAN scale. The pressured
speech, flight of ideas, and overinvolvement in activities lead to MAN-A elevations:
the grandiosity, expansiveness, and unrealistic self-appraisal are captured by MAN-G;
and the instability of mood and propensity for abrupt irritability when thwarted is
manifest in MAN-I elevations. Other scales can provide confirmation and additional
refinement of manic indications. The unwarranted optimism typical during a Manic
Episode tends to suppress DEP-C such that it often is considerably below the mean
of community respondents. The DOM scale often is elevated, although this does not
necessarily indicate that the person is actually effective in controlling others; rather,
it reflects a belief (probably unrealistic) that he or she is quite effective in a leader-
ship role. The individual’s interest in and uninhibited interactions with others
means that WRM is usually at or above the mean. Lack of insight into the nature
and severity of problems often leads RXR scores to be at or above 50T. Impulsiv-
ity, lack of inhibition, and poor judgment often result in elevations on BOR-S and
ANT-S. Finally, individuals in a Manic Episode often feel that large forces are work-
ing to thwart their efforts toward greatness; hence, elevations on PAR-P are fairly
common, although this sense of persecution does not seem to be accompanied by
the bitterness characteristic of the more purely paranoid individual. As such, PAR-
R tends to be considerably lower than PAR-P.
Individuals who have had one or more Manic Episodes are assigned a diagno-
sis of Bipolar I Disorder, regardless of whether or not they have ever experienced
a depressive episode (although most of these individuals do). When an individual
has recurrent depressive episodes with at least one “hypomanic” episode (similar
to a full Manic Episode, but briefer in duration), the diagnosis of Bipolar II Disor-
der is assigned. In either instance, such individuals may present on the PAI with
the unique combination of elevations on both DEP and MAN; generally, one of
these elevations is caused primarily by historical experiences, although mixed-
mood episodes can occur. It is unusual to find both MAN and DEP at 60T or
above; when this pattern is obtained, particularly with BOR-A also elevated, the
diagnosis of Bipolar Disorder should receive careful consideration.
As a final note, it should be recognized that many individuals in the midst of
a severe, acute Manic Episode are not compliant with, or are too agitated for,

165
PAI Interpretive Guide

psychological testing. In such instances when the examinee is too distracted or

uncooperative in completing the full administration of the test, the examiner may
find the 160-item short form of the PAI useful.

Schizophrenia
The diagnosis of Schizophrenia results from the presence of a number of
diverse features that have persisted for a duration of at least 6 months. These fea-
tures include positive symptoms that reflect an excess or distortion of normal func-
tions, as well as negative symptoms that involve a reduction or loss of different
functions. Other features central in establishing the diagnosis include grossly dis-
organized speech or behavior, or both.
The most prominent positive symptoms of Schizophrenia are delusions and
hallucinations. These aspects are assessed by the SCZ-P subscale, although other
scales are useful in supplementing SCZ-P to give a more precise picture of the spe-
cific symptoms involved. Delusions involve a distortion of inferential thinking that
results in the misinterpretation of experiences and perceptions. The most common
forms of delusion in Schizophrenia are persecutory delusions, where respondents
believe that they are being followed, deceived, spied on, or harassed. Ideas of ref-
erence are often associated with such experiences, where individuals believe that
certain events in the environment (e.g., comments on television, song lyrics, or
newspaper stories) convey specific information for them. Such persecutory and
referential beliefs result in elevations on PAR-P. Somatic delusions also can occur,
(e.g., believing that someone has surreptitiously removed various internal organs).
These delusions, when present, are most likely to appear in the form of SOM-C ele-
vations. Hallucinations are typically auditory and are generally experienced as
voices that are distinct from the person’s own thoughts. Hallucinations in other
sensory modalities can occur, although they are less common. Items inquiring
about hallucinations appear solely on SCZ-P; a few highly unusual hallucinations
are referenced in NIM items, but these experiences are not typical of schizophrenic
symptoms.
The negative symptoms of Schizophrenia tend to be the most stable and
unremitting feature of the disorder. Among these features, flattened affect is par-
ticularly common and refers to a restricted range of emotional expressiveness.
Another prominent negative symptom is decreased productivity of speech and
interaction with others. The combination of these features leads to impoverished
relationships and poor rapport, behaviors that are ascertained with SCZ-S items.
Other PAI scales also can be used to gauge the severity of negative symptoms. The
social isolation leads to low scores on WRM. Avolition, which refers to deficits in

166
 Use of the PAI in Diagnosis

initiating and persisting in goal-directed behaviors, can lead to low scores on DOM
and MAN-A, scales which indicate a degree of initiative in interpersonal (DOM)
and behavioral (MAN-A) realms. The diminution and emptiness of affect associ-
ated with schizophrenia can lead to scores on BOR-A that are low in relation to
other aspects of the profile.
Disorganized thinking is another characteristic of Schizophrenia that does not
fit neatly into the positive—negative symptoms distinction, yet it has been singled
out by some (e.g., Bleuler, 1950, who coined the term schizophrenia) as perhaps the
core defining feature of the disorder. The associations and speech of such individ-
uals tend to drift off topic, with difficulties in focusing answers to questions and
problems in the logical sequencing of ideas. In the extreme, the person may be
incoherent or the thought process may be completely blocked, often experienced
as if some external force was obstructing or removing thoughts from the person’s
head. The SCZ-T subscale includes items relevant to these experiences. Such indi-
viduals also may demonstrate idiosyncrasies in responding to PAI questions, with
distorted inferences about the questions interfering with their ability to respond to
the question as written. In such instances, the INF score may be elevated in a pro-
file that otherwise appears to accurately capture the clinical picture.
The diagnosis of Schizophrenia has a variety of additional specifiers that refer
to different patterns, courses, and phases of the disorder. Some of these specifiers
and the PAI profile information relevant to their identification are described in the
following sections.
Schizophrenia, Paranoid Type. The paranoid subtype is distinguished by the
presence of prominent auditory hallucinations and/or paranoid delusions,
with a relative preservation of cognitive functions and affective responsive-
ness. The delusions may be multiple, but they often are organized around
some coherent theme. Relative to other subtypes, the paranoid schizo-
phrenic displays deeper affect, most notably anxiety, but also anger and
hostility. The psychotic features of this disorder would be expected to ele-
vate SCZ-P relative to SCZ-S and SCZ-T, as well as all three subscales of
PAR. During acute phases of the disorder, ANX elevations would be antic-
ipated. Another possible elevation would involve MAN-I, to capture the
haughtiness and superior manner that such individuals often display.
Schizophrenia, Disorganized Type. The disorganized subtype is characterized
by a lack of goal orientation, flat and/or inappropriate affect, and disrup-
tions in thought process and communication. Such a pattern would be
expected to lead to marked elevations on SCZ-T as well as SCZ-S, with
SCZ-P considerably lower than those subscales.

167
PAI Interpretive Guide

Schizophrenia, Catatonic Type. Catatonic schizophrenia is characterized by

motor abnormalities, such as catatonic stupor, or by purposeless motor
excitement and/or stereotyped movements. In either instance, such
patients are not likely to complete the PAI.
Schizophrenia, Residual Type. The residual form of schizophrenia refers to
the lingering features of the disorder that remain after the active psychotic
symptoms have resolved. These features are predominantly negative symp-
toms, such as flat affect and poverty of interactions. Such features would
be likely to result in SCZ-S elevations without accompanying elevations on
SCZ-P. In addition, the SCZ-S elevation should occur in the absence of
other noteworthy elevations on the PAI, such as on DEP or ANX, since
these more neurotic spectrum features can lead to SCZ-S elevations. The
impoverished affect characteristic of residual schizophrenia should result
in low scores on these scales relative to SCZ-S scores.
Schizophrenia, Undifferentiated Type. This qualifier is added to the diagnosis
if prominent negative symptoms are present during or between active
phases of the disorder. As noted earlier, SCZ-S items are directly pertinent
to these symptoms, and marked elevations (i.e., 2 80T) might merit this
additional specification. Other PAI indicators would include low scores on
WRM, DOM, and MAN-A, indicating a lack of interest and initiative in
interpersonal (WRM, DOM) and behavioral (MAN-A) domains. Also, affec-
tive blunting may suppress scores on BOR-A relative to other aspects of the
profile.
The mean profile of a group of patients diagnosed with Schizophrenia (origi-
nally presented in the PAI Professional Manual [Morey, 1991]) is presented in Fig-
ure 6-6. Unfortunately, this sample was small (32 patients) and the respondents
were not in acute phases of the disorder; most were outpatients in a medication
maintenance clinic at the time of assessment. Patients were excluded from the
sample if their scores on any of the validity scales exceeded 2 standard deviations
above the mean of the total clinical sample. As can be seen in the figure, the Schizo-
phrenia group mean profile had no elevations that exceeded 70T; the SCZ and PAR
scales were above 60T, as were all of the neurotic scales. This pattern should be
considered to reflect the residual phase of the disorder, given the characteristics of
the sample. The most significant aspect of the profile involves the elevation of the
SCZ scale to a degree comparable to DEP and ANX; this is not typical of individu-
als with milder clinical conditions (e.g., see the Adjustment Disorder profile in
Figure 6-1).

168
 Use of the PAI in Diagnosis

PROFILE FORM FOR ADULTS-SIDEA 14 2 3 4 5 6 if 8 9 10 11 A

@—=-® Schizoaffective
Disorder (Morey, 1991)
@-++++*-@ Schizophrenia (Boyle &
Lennon, 1994)
e—e Schizophrenia
(Morey, 1991)

ce
a
1 2 3 4 5 6 ve 8 cy 10 a1 A 8 c o e y z
ICN INF NIM PIM SOM ANX ARD DEP MAN PAR SCZ BOR ANT ALC DRG AGG su! STR NON RXR DOM WRM

PROFILE FORM FOR ADULTS - SIDE B

b B° i
@—=—- Schizoaffective
Disorder
(Morey, 1991)
e— Schizophrenia
(Morey, 1991)

@ So

preeepreerpercepereeg

Ce

vreerbaveetbvcees
Pervert
SOMC SOM-S SOMH ANX-C ANXA ANXP ARD-O ARD-P ARD-T DEP-C DEPA DEPP MANA MANG MANI PAR:H PARP PARR SCZP SCZS SCZT BORA BOR BORN BORS ANT-A ANTE ANTS AGGA AGGV AGGP
CONV SOMA HEAL COG AFF PHYS OBS PHOB TRAU COG AFF PHYS ACT GRND IRRi HYPE PERS RSNT PSYC SOC THGT AFF ID NEG SELF ANT EGO STIM AGG VERB PHYS
SYMP ZATN CONC SYMP SYMP SYMP COMP IAS STRS SYMP SYMP SYMP LEVL JOSY BLTY VG CUTN MENT EXP DET DIS INST PROB REL HARM BEH CEN SEEK ATT AGG AGG

Figure 6-6. Mean PAI profiles for patients diagnosed with Schizoaffective Disorder and Schizo-
phrenia. Boyle & Lennon (1994) did not score the ICN scale or report subscale scores.

169
PAI Interpretive Guide

Figure 6-6 also includes the mean full scale profile of a sample of 30 patients
diagnosed as schizophrenic using DSM-III-R criteria, adapted from a study by
Boyle and Lennon (1994). This profile demonstrates an elevation on SCZ that
exceeds 7OT, but it resembles the Morey (1991) profile in that the clinical scales
show little differentiation. The stage of illness for these patients is not known, but
such a profile could represent a mixture of patients in acute stages (i.e., elevations
on SCZ and PAR) with patients in more residual stages (i.e., elevations on DEP and
ANX).
Because of the limited nature of these schizophrenia samples, additional
groups of patients were identified from the clinical standardization sample on the
basis of three characteristics: presence of auditory hallucinations, presence of per-
secutory delusions, and current treatment with antipsychotic medications. These
groups of patients (originally described in the PAI Professional Manual [Morey,
1991]) yielded the mean profiles displayed in Figure 6-7. All three of these groups
tended to obtain scores on SCZ that were elevated relative to most other popula-
tions. The group experiencing auditory hallucinations displayed the most elevated
profile, with mean scores above 70T on DEP and ANX as well as SCZ, and this
sample was notably higher than other groups on the psychotic experiences
(SCZ-P) and thought disorder (SCZ-T) indicators. The consistent elevations on
each of the ANX subscales most likely supports the conclusion that these halluci-
nating patients were in an acute phase of the disorder; as the disorder shifts into a
residual phase or is successfully treated, the ANX scale and its subscales appear to
drop along with SCZ-P. The group with persecutory delusions demonstrated a
similar profile configuration, but it was slightly more elevated on PAR; as seen in
Figure 6-7, the primary source of this elevation was from the persecution subscale
(PAR-P). Finally, the group receiving antipsychotic medication differed from the
other groups on the more “positive symptoms” of Schizophrenia (e.g., psychotic
experiences, paranoid beliefs), but not on the more “negative symptoms” (e.g.,
social detachment); interestingly, this is consistent with the established therapeu-
tic profile of antipsychotic medication.
Figure 6-6 also presents the mean profile for a group of patients diagnosed with
Schizoaffective Disorder. This disorder involves the co-occurrence of the active
phase of Schizophrenia with a major depressive or manic episode. The profile for
the schizoaffective group was somewhat more elevated than that of the schizo-
phrenic group; this was probably a result of a more acute symptomatic picture, as
the former were primary inpatients whereas the latter were primarily outpatients,
but it may also have resulted from the greater number of symptoms (i.e., from two
different disorders) required to meet the diagnosis. The highest clinical scale ele-
vations for the schizoaffective group were DEP, BOR, and SCZ, suggesting that this

170
 Use of the PAI in Diagnosis

PROFILE FORM FOR ADULTS -= SIDE A at 2 3 4 5 6 ii 8 9 10 a4:

PAO T 5 : A 8 Cc io} E y ?
ae. 70= 70- A 70 60 E. | e @ Antipsychotic
: 3 | medications
lg 65- 65 65 65 = = e 30-
Z - cc ee = : | @=---@ Auditory
100 — a és 70 70 =
: 60
30- Ate
oe hallucinations
: 15 ee = 60- ~—60- 60 o—— Persecutory
= - = 65 = c =
20- eee = = =5 25 delusions
o6 - 55
~ 55 55 60 25

=i z 55> 50 = : 50- BRE - 25 =

= | a5> = 45.

Pe 10 |
= 15 10.
iat =
=
| 15
= A 35
- a . 6 35 15

To = 4 | 15 |
= 35
2
ry : a= |
SN ones Pet |
eS
n |
|

60 —

50 —

= | |

- |
40 —

= Os %
~
==
5
= S- oO -
30
|
—
a
. z 5 = a :
= = 0. 3 |
oe =
= aa =
= - = o- | 40

20 Ke:
: := | 0 | :
1 2 a a 5 6 7 8 9 C z
ICN INF NIM PIM SOM ANX ARD. DEP MAN PAR scz BOR ANT ALC pee AGG Sul STR NON RXR OM WRM

PROFILE FORM FOR ADULTS - SIDE B

@..---.--@ Antipsychotic
medications
@----@ Auditory
hallucinations
o— Persecutory
delusions

$9J09S
1

SOMC SOM-S SOMH ANX-C ANKA ANXP ARD-O ARD-P ARDT DEP-C DEPA DEP-P MANA MANG MANI PARH PARP PARR SCZP SCZS SCZT BORA BOR! BORN BORS ANTA ANTE ANTS AGGA AGGV AGGP
CONV SOMA HEAL COG AFF PHYS OBS PHOB TRAU COG AFF PHYS ACT GRND IRRI HYPE PERS RSNT PSYC SOC THGT AFF ID NEG SELF ANT GO STIM AGG VERB PHYS
SYMP ZATN CONC SYMP SYMP SYMP COMP IAS STRS SYMP SYMP SYMP LEVL IOSY BLTY VIG CUTN MENT EXP DET DIS INST PROB REL HARM BEH CEN SEEK ATT AGG AGG

Figure 6-7. Mean PAI profiles for current antipsychotic medication, auditory hallucinations, and
persecutory delusions samples (Morey, 1991).

171
PAI Interpretive Guide

Table 6-4
Summary of Key PAI Diagnostic Indicators for
Psychotic Spectrum Disorders
Diagnostic Elevation Suppression
consideration indicators indicators

Bipolar Disorder MAN-G, MAN-A DEP-C

MAN-! usually up, can vary PAR-R relative to PAR-P
DEP-P can be moderate
BOR-A above average
ANT-S, DOM, PAR-P
Schizophrenia SCZ (all three subscales) WRM
ANX during acute episodes DOM
PAR-P for Paranoid subtype MAN-A
lf MAN and DEP, consider
Schizoaffective

Delusional Disorder SCZ-P with SCZ-T relative to SCZ-P

PAR-P, MAN-G, or SOM-C

group was displaying a wide array of symptoms consistent with the “mixed” diag-
nosis they received. DEP elevations are not uncommon among schizophrenics, and
simultaneous elevations on DEP and SCZ should not automatically be interpreted
as an indication of Schizoaffective Disorder. However, the joint elevation of SCZ
and MAN (often with DEP also elevated) is rare and points with greater specificity
to the possibility of a Schizoaffective Disorder diagnosis.

Delusional Disorder
The defining feature of the Delusional Disorder (formerly known as Paranoid
Psychosis) is a relatively circumscribed, nonbizarre delusion, with relative intact-
ness of functions unrelated to the delusional belief. The delusion can take many
forms, including persecutory, somatic, jealous, grandiose, or erotomanic (where
the person imagines that another person is in love with him or her). The nature of
the delusion is distinguished from those typical of Schizophrenia in that it involves
situations that could conceivably occur (e.g., being followed, infidelity).
On the PAI, the most likely markers of Delusional Disorder would involve
some elevation on SCZ-P, accompanied by other scales that would give some indi-
cation of the nature of the preoccupation (e.g., PAR-P for persecutory beliefs,
MAN-G for delusional grandiosity, or SOM-C or SOM-H for somatic delusions). To
assign the diagnosis, it would be important that SCZ-T and SCZ-S be within

Liz
 Use of the PAI in Diagnosis

normal limits, because this disorder does not involve impairments in affective
responsivity or cognitive function. Although some distress may occur, depressive
symptoms are typically mild; marked elevations on DEP should be rare and, when
observed, would tend to be less enduring than the indicators of delusional beliefs.

Personality Disorders
The diagnosis of personality disorders, as a group, may be one of the most
complicated tasks from the PAI protocol, because the disorders themselves tend to
be ill-defined. The DSM definition refers to personality traits that are inflexible and
maladaptive and that cause distress or impairment. In 1937, Allport identified
17,953 nonobsolete trait names in the unabridged Webster's dictionary, a number
constituting 4.5% of the total English vocabulary (the number of words in
Webster's has grown by about 12% since Allport’s study, suggesting that a few addi-
tional traits have probably appeared). The PAI is bound to leave a few of these
traits uncovered. In fact, of the 10 primary personality disorders defined in the
DSM-IV, the PAI includes scales directly assessing only 2: Borderline Personality
and Antisocial Personality. This decision was made for a number of reasons.
First, the two disorders account for the majority of research on personality dis-
orders; Borderline Personality, alone, accounts for over half of all studies on Axis
Il (Blashfield & McElroy, 1985). Second, the focus on discriminant validity in the
construction of the PAI was at odds with the well documented lack of discriminant
validity among the personality disorders; for example, individuals meeting criteria
for only one personality disorder (as opposed to two or more) are a small minor-
ity of personality-disordered patients (Morey, 1988). Given such marked diagnos-
tic overlap, the task of constructing conceptually independent scales is greatly
hampered. Under such circumstances, it seemed more useful to focus on only
those personality constructs that were empirically supported and clinically rele-
vant (e.g., the interpersonal dimensions) and to use these constructs to supple-
ment Axis II diagnosis.
The DSM approach to this area is most notably deficient in its efforts to pro-
vide a coherent definition of this class of phenomena. The DSM-IV goes beyond
previous versions by specifying enduring difficulties in cognition, affectivity, inter-
personal functioning, and/or impulse control, although these guidelines are
described at a level of generality that would make them difficult to use clinically.
Many investigators have a different view of what precisely constitutes Personality
Disorder and how it differs from both Axis I disorders and normal personality.
There is a need to spell out certain assumptions about the nature of these disor-
ders: with each specified assumption, the concept will become more manageable

173
PAI Interpretive Guide

and less ephemeral. The following sections provide some candidate concepts for
defining personality disorder as a whole and suggest some PAI indicators that
might be useful in identifying these concepts.
Manifestations of personality disorder are evident early in life. The DSM-IV notes
that the manifestations of personality disorders are often recognized by adoles-
cence or earlier, although it suggests that caution is warranted in using these diag-
noses with children or adolescents. Relatively little is known about the develop-
mental precursors of personality disorder. Although adult antisocial behaviors can
be predicted with some success by behaviors such as aggressiveness and stealing
in children as young as 6 to 9 years of age (Loeber & Dishion, 1983; Robins,
1966), other candidates for childhood markers of future personality disorder (e.g.,
shyness) seem to have little predictive value for later adjustment (Kagan & Moss,
1962; Parker & Asher, 1987). This suggests that scales such as ANT-A, ANT-E, and
AGG-P are likely to reflect enduring traits and, when elevated, to suggest the pos-
sibility of personality disorder. However, PAI scales pointing to dependency, shy-
ness, or withdrawal (e.g., low scores on DOM or WRM, or elevations on ANX-P or
SCZ-S) should not automatically be assumed to reflect enduring characteristics.
Such indicators can be considerably influenced by mood state, and, as such, con-
comitant elevations on DEP suggest that additional inquiry may be needed to
establish the persistence of the withdrawal and dependency.
Manifestations of personality disorder are stable over time. The DSM description of
Personality Disorder refers to enduring traits that are characteristic of long-term
functioning. This assumption has been the focus of some debate, beginning in
1968 with Mischel’ critique of trait psychology. One result of this debate has been
the recognition that some “traits” are more stable over the lifespan than others; for
certain traits, personality stability is evident well into older age (e.g., McCrae &
Costa, 1984). If stability over time is a core feature of personality disorder, then
the extreme manifestations of these traits should be particularly promising candi-
dates for study (although probably not for treatment). Because the PAI scales tap
most facets of these dimensions reasonably well (Costa @ McCrae, 1991) the test
should be useful in providing an assessment of these more enduring personality
aspects. It should be noted that although the limited evidence confirms that cer-
tain personality disorder diagnoses, particularly Borderline Personality Disorder,
are stable over time (McGlashan, 1983; Pope et al., 1983), it is not clear that this
represents stability of traits. For example, McGlashan’s (1986) work suggests that
the prominent traits of the borderline personality tend to vary across the life span;
the anger and impulsivity seem to diminish, whereas identity disturbances and
relational deficits continue. This pattern suggests that BOR-I and BOR-N may be
useful diagnostic features across the lifespan, but BOR-A and BOR-S elevations

bid
 Use of the PAI in Diagnosis

may be more characteristic of borderline individuals in their 20s than those in

their 50s.
Manifestations of personality disorder are stable across different situations. In his
controversial 1968 book, Mischel pointed out that much psychological research
has not supported the contention that individual differences in behavior are very
stable across different situations. As a result, Mischel and others have argued that
the situation in which the person finds himself or herself may be a stronger deter-
minant of behavior than internal personality dispositions. Although the person-
situation controversy generated more heat than light, two important points of con-
sensus have emerged (Epstein & O’Brien, 1985; Kenrick & Funder, 1988). First,
the stability of personality across situations is most evident when one considers
aggregates of behavior rather than single behavioral instances; our ability to predict
behavior from trait information at a particular point in time is quite limited. In the
personality disorders, where the need to make predictions concerning specific
behaviors is often salient (e.g., immediate threat to self or others), this limitation
has grave implications. The second point of consensus emerging from this litera-
ture is that the interaction effect between trait and situation is more likely to
enhance our ability to predict specific behaviors than a single-minded focus on
either traits or situations (Endler @ Edwards, 1988). We need to address the ques-
tion: What types of people perform what behaviors in what types of situations? As
an example, we know that suicidal gestures are a hallmark feature of borderline
personality (Morey & Ochoa, 1989), but this also seems to be the borderline cri-
terion with the lowest prevalence. Recently, research has begun to focus on the sit-
uations most likely to interact with borderline personality features to bring about
serious suicidal behaviors, identifying interpersonal rejection as a particularly
sensitive situation for these people (Kullgren, 1985; Linehan, 1986). This would
predict that people with BOR elevations will be at particular risk for suicidal behav-
ior when scales indicative of interpersonal rejection (e.g., NON) also are elevated.
Personality disorders are largely ego-syntonic conditions. The early psychoanalytic
writers assumed that personality disorders involve character traits that are an
essential part of the personality, rather than symptoms that are experienced by the
individual as alien to the personality. This distinction is of clear utility in distin-
guishing certain Axis I and Axis II disorders, such as the obsessive-compulsive dis-
orders. However, because the key problems in personality disorder are experi-
enced as a fundamental part of the personality rather than as a distressing
symptom, this allows for limited insight into the nature of the person’s diffi-
culty. For example, the presenting complaint for many personality disorders is
likely to involve marked depression, anxiety, or interpersonal conflict; it is rare

ae)
PAI Interpretive Guide

that such people will complain of identity diffusion or lack of empathy. People
with personality disorders are likely to present for treatment only during times of
crisis, even though their core deficits are there much of the time. For such people,
careful attention should be paid to indicators of character (e.g., BOR, ANT, or PAR
elevations). These indicators can sometimes be overlooked in the context of
extreme elevations on DEP, ANX, or SUI that are associated with the immediate
crisis.

Personality disorders are largely interpersonal in nature. Over the past several
decades, a number of writers have identified interpersonal behavior as an impor-
tant focus for the study of personality and psychopathology (Adams, 1964;
Horney, 1945; Kiesler, 1983; Leary, 1957; McLemore & Benjamin, 1979; Sullivan,
1953; Wiggins, 1982). One focus of such attention has concerned the utility of the
interpersonal approach as a foundation for the diagnosis and classification of per-
sonality disorders. It is clear that most personality disorder diagnoses are based on
reports or observations of interpersonal behavior, although this does not necessar-
ily imply that personality disorders are distinct from Axis I disorder, in that they
are dysfunctional primarily through their expression in the social milieu. Writing
from the interpersonal perspective, McLemore and Brokaw (1987) suggest that
personality disorders are “disturbances” in the sense that the behavior of such peo-
ple is disturbing to someone else, implying that a person stranded alone on a
desert island cannot have a personality disorder. This implies that nearly all per-
sonality disorders will have characteristic patterns on the interpersonal scales
DOM and WRM.
Personality disorders differ quantitatively from normal personality variation. This
assumption bears upon the “categorical versus dimensional” debate; the assumption
is that individuals with personality disorders and those with “normal” personalities
differ in degree rather than in kind. This implies that having a personality disor-
der is not an “either—or” type of distinction; rather, personality issues can play a
role for the person to a greater or a lesser extent. Perhaps the closest operational-
ization of this “degree” of personality disturbance on the PAI is the BOR full scale:
The greater the elevation on BOR, the more likely it is that personality problems
are playing a role in the person’s presenting complaints.
Each of these assumptions represents an area of controversy within psychology
and psychiatry, but without such assumptions there is no explicit boundary to the
domain of phenomena denoted by the concept of “personality disorder.” The DSM
does, however, provide explicit definitions for various specific personality disor-
ders. These disorders, and the PAI indicators relevant to establishing the diagnoses,
are described in the following sections.

176
 Use of the PAI in Diagnosis

Borderline Personality Disorder

This disorder is characterized by instability across multiple behavioral domains,
including self-image, mood and affect, interpersonal relationships, and impulse
control. Individuals with this personality disorder are highly reactive to interper-
sonal events; they are both distrustful of those closest to them and also fearful that
these others will abandon them. Their self-image can fluctuate dramatically and is
heavily dependent on the people around them. They are very impulsive in ways
that are self-damaging, including spending money, driving, sexual activity, and
other reckless behaviors; they often display a pattern of undermining themselves
when things appear to be going well. Their mood is very reactive to external events
and often fluctuates rapidly between anger, panic, and despair; mood disorders are
often comorbid with Borderline Personality.
The mean profile for a group of patients (including both inpatients and outpa-
tients) diagnosed with Borderline Personality Disorder is presented in Figure 6-8.
These data were originally presented in the PAI Professional Manual (Morey, 1991).
This profile demonstrates a prominent elevation on BOR as well as a marked eleva-
tion on DEP. This group also had a mean SUI score that approached 80T; this ele-
vation is consistent with the repeated suicidal gestures that are pathognomonic for
this diagnostic group. Unlike other clinical groups, the borderline group shows
elevations on all four borderline subscales; although many individuals with other
forms of personality problems may show elevations on one or two subscales, thus
perhaps elevating the BOR full scale, few respondents who are elevated on all four
BOR subscales will fail to meet DSM-IV criteria for the disorder.
A study of 22 women inpatients diagnosed with Borderline Personality Disor-
der was reported by Bell-Pringle (1994); the mean profile obtained in that study is
also presented in Figure 6-8. The patterns for the two groups are quite similar, the
greater elevation on DEP observed in the Bell-Pringle sample is likely the result of
the sample being restricted to inpatients. If so, the relationship between the two
patterns suggests that the absolute elevation of DEP is more likely than that of BOR
to be associated with hospitalization for borderline patients; the level of BOR is rel-
atively similar in the inpatient sample and the general sample. This may reflect the
sensitivity of DEP to the precipitating crisis, whereas BOR indicates more stable
and enduring pathology that may well continue after discharge.
Other aspects of these profiles are informative as to the characteristics of the
disorder. For example, borderline individuals tend to have prominent elevations
on ARD-T, indicative of a past history of traumatic stress. This is consistent with
reports in the literature that the rates of physical and sexual abuse in the develop-
mental history of such patients is extremely high (Herman, Perry, & Van der Kolk,

177
PAI Interpretive Guide

PROFILE FORM FOR ADULTS-SIDEA 4 2 3K 4 5 6 7 8 9 10 a4 A c D E a z

110 = ue 5 60 @——- Borderline

= e es (Bell-Pringle, 1994)
30- ;
cs 65 65 ‘ a5 o——e Borderline
; 0. 0 = ° 60 30 (Morey, 1991)
100 :
aS Bo 60. 60.

— 90
90

80 80

| 35) AX

cles 35 = ice
i eee
3 30- 2 g
3
a3 aE z
g
= — 60
60 3 .

| 2 =

- 25- s

= — 50
50 20- = =

40
15 —= 40

ES: a

30 a“ 2 =sE0)
=

Sy Ge
2 s
a5
20 | °Z =) .20
1 2 3 a 5 6 7 8 9 10 1 A c D E Y z
ICN INF NIM PIM SOM ANX ARD DEP MAN PAR scz BOR ANT ALC DRG AGG SUI STR NON RXR DOM WRM

PROFILE FORM FOR ADULTS - SIDE B

T
aay e@——-@ Borderline
24- aes (Bell-Pringle, 1994)
Sy 0. 28 e——e Borderline
100 | (Morey, 1991)
24 = =
= | ae 15- =
24- 15-
= = 20-
90 — a — 90
= 15- :
20- - = =

> 20- 18- =

20- 7 Be =
80 = : E gl hy ~ 80
: 15 - ~_}15- - -
10- 45- = |G =
=
- = 15- fs g : 15- 3 -:
10. - 10- " 5
TO} = Ss = SS | Se ee)
: - ; 4
fos || 0 f- 2 m: ers Ke
zak

60 — ao 10 = 60
: - = | 10- -
5- - - - - 5- =

:= 5 & oleae = -
9 ee RE a See et == 50)
5 5- =
& / MA
: ahi pete i

0: a 2 4 *
40 — = u >, "3 2
2 0- i 0-5 a oe oO Pe ‘- i-
=
o- = 0- o-

30 — : og - Sse: = 6 9a
SOM SOM'S SOMH ANXC ANXA ANXP ARDO ARDP ARD-T DEP-C DEPA DEP-P MANA MANG MANI PAR-H PARP PARR. SCZP SCZS SCZT BORA BORI BORN BORS ANT-A ANTE ANTS AGGA AGGV AGG-P
CONV SOMA HEAL coG AFF PHYS OBS PHOB TRAU coG AFF PHYS: ACT GRNOD IRRI HYPE PERS RSNT PSYC SOC THGT AFF 1D NEG SELF ANT —GO STIM AGG VERB PHYS
SymMP ZATN CONC SYMP SYMP SYMP COMP IAS STRS SYMP SYMP SYMP LEVL JOSY BLTY VIG =CUTN MENT EXP DET bIS INST PROB REL HARM BEH CEN SEEK AIT AGG AGG

Figure 6-8. Mean PAI profiles for two groups diagnosed with Borderline Personality Disorder. No
data were provided by Bell-Pringle (1994) for the PAI subscales.

178
 Use of the PAI in Diagnosis

1989). The elevation on DEP reveals borderline individuals to be quite pessimistic

and hopeless, but DEP-P is not particularly elevated, suggesting the absence of
physiological signs of depression. This also is consistent with the research litera-
ture that indicates that borderline individuals are not responsive to traditional anti-
depressant medication; in fact, such medication may actually worsen their symp-
toms (Soloff et al., 1989). Other features of the disorder that are often reflected on
the PAI include marked anxiety, resulting in ANX elevations; impulsive substance
abuse, manifest on ALC and/or DRG; very low self-esteem, resulting in low scores
on MAN-G; and a broadly negative evaluative set that can elevate scores on NIM.
The LOGIT function for the diagnosis of Borderline Personality Disorder (as
contrasted with other clinical diagnoses) supports many of these observations. The
function (presented for illustration purposes rather than for use in routine clinical
situations) included the following weights for PAI variables:

.020(BOR-A) + .001(BOR-I) + .014(BOR-S) + .005(DEP-C) — .007(BOR-N) — .006(PIM) + .245

This function highlights the significance of BOR-A and BOR-S in determining

whether a person receives a Borderline Personality Disorder diagnosis. This is
probably due to the fact that these elements of the disorder are, perhaps the most
dramatic in presentation and the most difficult for the clinician to manage. The
BOR-I component makes a small independent contribution to the diagnosis,
whereas BOR-N operates as a suppressor variable, demonstrating that volatile inter-
personal relationships are often associated with other disorders and are not neces-
sarily specific to Borderline Personality Disorder. The negative loading for PIM is
not a suppressor, however; borderline individuals tend to obtain very low scores
on PIM, reflecting a very negative self-image. Another facet of this self-image is
reflected in the DEP-C loading, indicating that borderline individuals see them-
selves as ineffectual and shameful people.
A similar actuarial study, based on a discriminant analysis contrasting 22 bor-
derline inpatients with 22 student controls, was conducted by Bell-Pringle (1994).
In this study, the greatest correlations with the discriminant function (which was
based only upon the 11 clinical scales) were found with DEP (r = .73), followed
by BOR (r = .59), ARD (r = .57), and SOM (r = .51). With respect to mean com-
parisons, large group effect differences were also seen on SUI, STR, and RXR among
the treatment scales, with diagnostic group differences accounting for at least 40%
of the variation on these scales. For all scales mentioned, borderline individuals
were elevated relative to controls, with the exception of RXR, where the borderline
patients had lower mean scores.

179
PAI Interpretive Guide

Antisocial Personality Disorder

The diagnostic criteria for Antisocial Personality Disorder specify an enduring
pattern of disregard for the rights of others, including illegal activities, aggressive
acts, and deceit and manipulation of others. Such people tend to be extremely
undependable and irresponsible; impulsive decisions and reckless actions are
common. Although the antisocial routine exploits others, these individuals expe-
rience no remorse, being either indifferent to the plight of others or providing
some superficial rationalization for their actions.
The DSM diagnostic criteria bear some relationship to the construct of psy-
chopathy, but, because the criteria tend to focus on antisocial behavior rather than
personality traits, the psychopathy concept is less inclusive than that of Antisocial
Personality. The PAI features that are most directly related to the DSM concept are
ANT-A, which captures the acting-out behaviors, and AGG-P, which captures the
aggressiveness and history of physical confrontation that are represented in the
diagnostic criteria. The concept of psychopathy includes a host of additional
aspects: lack of empathy, egocentricity, an inflated self-appraisal, and glib, superfi-
cial charm are all elements of psychopathy. These aspects are most likely to be
reflected in elevations on ANT-E. Finally, the recklessness and impulsive gratifica-
tion-seeking are captured in the items of ANT-S; although such features are not
necessary to receive the diagnosis of Antisocial Personality, they are invariably pre-
sent in the narrower constellation of the psychopath.
The mean profile for a group of patients diagnosed with Antisocial Personality
Disorder (originally presented in the PAI Professional Manual [Morey, 1991]) is pre-
sented in Figure 6-9. The prominent elevations on this mean profile include DRG
and ANT. The former elevation is probably artifactually inflated due to the large
number of incarcerated respondents in the antisocial group, although drug abuse
is reasonably common in antisocial personality. Among the subscales, antisocial
behaviors (ANT-A) was the most prominently elevated, a pattern consistent with
the fact that these diagnoses were based on DSM definitions that place a great deal
of weight on behavioral features. Other aspects of the profile that bear upon the
antisocial constellation are the relative lack of neurotic manifestations and suicidal
ideation; these scores tend to be lower than comparable scores from almost any
other diagnostic clinical group. Thus, the profile for Antisocial Personality invari-
ably is more elevated on the “right-hand” side of the clinical scales (i.e., BOR, ANT,
ALC, DRG) than on the “left-hand” side (i.e., SOM, ANX, ARD, DEP). The differ-
ence between the antisocial and the borderline individual on MAN-G is notewor-
thy and points to the arrogant self-appraisal of the antisocial individual that is
maintained in the face of life circumstances that suggest otherwise. Finally, the

180
 Use of the PAI in Diagnosis

PROFILE FORM FOR ADULTS-SIDEA 1 2 3 4 5 6 7 8 9 10 11 A B Cc Dd E ¥ z

140 — ~ 5 70. : E =, |
= 25-
70 oa :- ts we
=
bai =
~
| @=—=—-@ Antisocial
=: Ei
; : : ; : ‘ z : | o-—0
; 65- _ ese _ e = - See | sf? Depencent

fa
mation

rolvs

roads

= = =- - o- 10: Ae 2 =
o- = 5 15
- o-
: - = Soe o-
= : -S 3= | 5 | 10 =
E3 x." : 3oa ks 3
=- 2 o- =- | 10

oa 5
: Be
1 2 3 4 s 6 x 8 3 10 a A 8 c ie) E Y z
ICN INF NIM PIM SOM ANX ARD DEP MAN PAR scz BOR ANT ALC ORG AGG sul STR NON RXR DOM WRM

PROFILE FORM FOR ADULTS - SIDE B

=Bo !
@=—=—- Antisocial
e——e Dependent

ee

robeevebarcedeernad

ee
ee

ANKC ANXA ANXP ARDO ARD-P ARDT DEPC DEP-A DEPP MANA MANG MANI PARH PARP PARR SCZP SCZ-S SCZT BORA BOR! BORN BORS ANTA ANTE ANTS AGGA AGGV AGGP
SOMC SOMS SOMH
GRND IRRI HYPE PERS RSNT PSYC SOC THGT AFF iD NEG SELF ANT EGO STIM AGG VERB PHYS
CONV SOMA HEAL COG AFF PHYS OBS PHOB TRAU COG AFF PHYS ACT
SYMP LEVL IOSY BLTY VIG CUTN MENT EXP DET DIS INST PROB REL HARM BEH CEN SEEK ATT AGG AGG
SYMP ZATN CONC SYMP SYMP SYMP COMP IAS STRS SYMP SYMP

Figure 6-9. Mean PAI profiles for patients diagnosed with Antisocial and Dependent Personality
Disorders (Morey, 1991).

181
PAI Interpretive Guide

downward slope of the interpersonal scales (i.e., with DOM considerably higher
than WRM) is also representative of the cold, controlling interpersonal style of
individuals with this personality disorder.
The LOGIT function for the diagnosis of Antisocial Personality (as contrasted
with other clinical diagnoses) provides further understanding of the scales that are
critical in assigning the diagnosis. The function (presented for illustration pur-
poses, rather than for use in routine clinical situations) included the following
weights for PAI scales:

.044(ANT-A) + .017(AGG-P) — .008(ANT-E) — .002(ANT-S) — .028(ANX-A) + 1.85

This function highlights the centrality of ANT-A and AGG-P in determining

whether a person receives a DSM Antisocial Personality diagnosis. This is consis-
tent with the way the disorder is portrayed in the DSM criteria. The ANT-S and
ANTE scales act as suppressor variables, but the negative loading for ANX-A is not
a suppressor; rather, it reflects the fact that this scale is typically lower for antiso-
cial than it is for most other clinical groups.

Narcissistic Personality Disorder

The defining feature of the narcissistic personality is a pattern of inflated self-
esteem, demand for admiration, egocentricity, and lack of empathy. Such people
overvalue their ideas and accomplishments, appearing self-important and preten-
tious to the people around them. They believe that their special talents and abili-
ties can only be fully appreciated by others of equal talents and superior status,
and they devalue others who fail to admire them or who otherwise disappoint
them. Although inflated, their self-esteem is fragile and vulnerable to insult, which
can cause them to become depressed, humiliated, and/or furious with those who
have not granted them their entitlements.
The haughty and arrogant self-appraisal of the narcissistic individual suggests
that MAN-G is a logical starting point for identification, with elevations of at least
60T being typical. However, because the self-esteem can be fragile, MAN-I can be
elevated as well, particularly if there has been a recent narcissistic injury. In
contrast to an individual in a manic episode, elevations on MAN-A are unusual;
whereas the manic can be involved in a great many activities, the narcissist is more
selective, participating only in those impressive tasks that merit his or-her
self-evaluated special talents and unique abilities. ANT-E is often elevated in
Narcissistic Personality, as it captures the lack of empathy and profound egocen-
tricity specified in the diagnostic criteria. On the interpersonal scales, the narcis-
sistic individuals are invariably above average on DOM, because they feel entitled

182
 Use of the PAI in Diagnosis

to be in control of almost any interpersonal situation. However, in contrast to anti-

social individuals, their elevation on WRM is generally within normal limits, as it
is more important to the narcissist to maintain relationships (if only to assure a
steady supply of admirers).

Histrionic Personality Disorder

The core feature of the histrionic personality style is a pattern of excessive and
superficial emotionality and attention-seeking behavior. Such individuals demand
to be the center of attention and attempt to attract such attention through dramatic
behavior and flirtatiousness. They are overly concerned with physical appearance,
often dressing in a provocative manner and attempting to draw compliments from
others. They tend to express their emotions in an exaggerated and rapidly shifting
way, leading others to feel that the emotions are superficial or even faked. Indi-
viduals with this personality style tend to have difficulty establishing and main-
taining intimate relationships, often controlling their partners with emotional
manipulation. At the same time, relationships with friends also are impaired because
of the histrionic’s competitive and provocative nature.
The excessive and superficial emotionality of the histrionic personality sug-
gests that BOR-A represents an important facet in identifying individuals with this
style. The egocentricity and manipulativeness of such people should result in an
elevation on ANT-E. However, unlike other, more malignant, personality disor-
ders, histrionics place a premium on interpersonal relationships, and the quality
of these relationships (at least their superficial quality) is quite important to them.
Thus, WRM is typically above the mean for this group and, in fact, may be quite
elevated. Also prominent in the histrionic character is the repression of anger and
other disturbing affects; thus, AGG-V is typically quite low, as anger tends to be
indirectly expressed. More comfortable with the expression of physical than psy-
chological symptoms, this personality style is associated with various somatoform
disorders, and SOM-S would be expected to be particularly elevated for this group.

Paranoid Personality Disorder

This disorder is characterized by an enduring pattern of distrust and suspi-
ciousness of others. Such individuals assume that others are malevolent, and they
expect harm or deceit from others. In the absence of any objective evidence, they
feel that they have been treated unfairly and feel deeply injured by the people
around them. They are reluctant to confide in others and feel that any personal
information that might be divulged will be used against them. Such people bear
grudges against others, are quick to feel attacked or threatened, and react to minor
slights with dramatic hostility.

183
PAI Interpretive Guide

The PAR scale is the obvious beginning point for the investigation of Paranoid
Personality. Because these individuals are generally not delusional, it is typical to
find PAR-P to be considerably below PAR-R and PAR-H, thus forming a V-shaped
profile for the three PAR subscales. The interpersonal distrust reflected in PAR-H
and the bitterness and hostility captured by PAR-R combine to provide a reason-
ably complete coverage of the features of the disorder. Other features of the profile
can supplement this information. For example, SCZ-S may be elevated, reflecting
the tendency to keep relationships distant and superficial; AGG-A is often high,
indicating the extent of the anger underlying the mistrust; and WRM scores are
typically below average, because the paranoid character places little premium on
close relationships.

Schizoid Personality Disorder

The diagnosis of Schizoid Personality Disorder is characterized by an indiffer-
ence to and detachment from social relationships, as well as a constricted experi-
ence and expression of affect. Such individuals neither enjoy nor desire relation-
ships, preferring to be alone in both occupational and leisure time pursuits. They
may appear emotionally cold to others, but they are unconcerned about the way
that they are regarded and are unaffected by either positive or negative comments
from others. The emotional constriction includes anger, which they rarely express,
even when directly provoked. Lacking social skills and disinterested in sexual rela-
tions, such individuals have few acquaintances and seldom marry successfully.
The social detachment and affective constriction of the schizoid individual are
directly captured by SCZ-S, and it is unlikely that individuals with this diagnosis
would not demonstrate some elevation on this scale. Also, the WRM scale is likely
to be quite low, given the marked lack of interest in attachment relationships. The
affective constriction of the schizoid individual would also lead to suppression of
BOR-A, which can help to distinguish the disorder from many other personality
disorders with prominent social isolation (e.g., Avoidant Personality Disorder);
other neurotic spectrum indicators such as ANX and DEP may also be low, partic-
ularly if the individual is not in an occupational situation that is taxing the limited
social skills. Finally, the disorder should not be diagnosed if these conditions occur
solely during the course of Schizophrenia (including the residual phase), suggest-
ing that SCZ-P elevations are likely to be counterindicative of this diagnosis.

Schizotypal Personality
The Schizotypal Personality is distinguished by interpersonal deficits as well
as cognitive and behavioral eccentricities suggestive of problems within the

184
 Use of the PAI in Diagnosis

Schizophrenia spectrum. Although not of the severity that would merit a diagno-
sis of Schizophrenia, the person may admit to mild ideas of reference, magical
thinking, or paranoid ideation. Such individuals are not comfortable in social sit-
uations, more typically experiencing tension and anxiety (often associated with
suspiciousness concerning the intentions of others). Expression of affect is typi-
cally constricted, and the individual may also experience mild signs of inefficiency
and confusion in thought process.
Because of the similarity of this disorder to the residual phase of Schizophre-
nia (distinguished primarily by a past episode of active schizophrenic psychosis),
the PAI indicators for schizotypal personality are similar to those for the residual
diagnosis. The disorder includes many negative symptoms (e.g., flat affect and
poverty of interactions); such features would be likely to result in SCZ-S elevations
that are more prominent than any accompanying elevation on SCZ-P. However, it
is unlikely that SCZ-P would be below 50T in such individuals, due to their eccen-
tricity and peculiar beliefs. SCZ-T and PAR-P also would be expected to display
moderate elevations, reflecting the cognitive distortions and mild paranoid idea-
tion characteristic of the disorder. The social awkwardness and anxiety would be
reflected in a number of different PAI domains; suppressed scores on WRM with ele-
vations on ARD-P (driven by the social anxiety items) would be expected. In con-
trast to Schizoid personality, the schizotypal individual experiences considerable
anxiety in social situations, and this latter disorder is more likely to demonstrate
ANX elevations than is the schizoid individual.

Avoidant Personality Disorder

The defining features of Avoidant Personality Disorder involve marked social
inhibition, feelings of inadequacy, and intense sensitivity to negative evaluation.
Such individuals typically demonstrate impairment in occupational settings because
of their anxiety concerning interpersonal contact that might lead to criticism or dis-
approval. They are most fearful of new interpersonal situations (i.e., interactions
involving strangers, whom they are afraid will reject them as inadequate). Low self-
esteem is a core feature of the disorder; they believe themselves to be socially inept,
unappealing, and inferior to others. As a result, they will exaggerate the dangers
inherent in novel situations and use this as an excuse to minimize interactions with
others they do not know well.
Avoidant Personality Disorder overlaps a great deal with the Generalized type
of Social Phobia, to the extent that the two diagnoses may be alternative names for
the same condition. Thus, the PAI indicators for the disorders are similar, with the
core indicator for Avoidant Personality Disorder involving elevations on ARD-P.

185
PAI Interpretive Guide

The social anxiety indicated by that scale leads to interpersonal withdrawal and
avoidance, (i.e., scores on WRM are typically suppressed). Also, scores on DOM
tend to be quite low, because the avoidant personality is particularly uncomfort-
able in leadership roles where there is a great deal of public scrutiny and where
any failures are likely to be widely known. Finally, the desire to avoid any novel
social situations tends to suppress scores on ANT-S, as the avoidant individual
seeks to avoid the stimulation associated with unpredictable interactions.

Obsessive-Compulsive Personality Disorder

The obsessive-compulsive personality is preoccupied with orderliness, perfec-
tionism, and a rigid need for internal and interpersonal control. The preoccupa-
tion with trivial detail is generally at the expense of a grasp of the overall situation;
thus, projects often are not finished because of the attention paid to preventing
minor or irrelevant mistakes. However, productivity is far more important than
leisure to these individuals, who tend to regard leisure activities as wasted time.
Such people are inflexible about matters of morality and rigid in their conviction
that others should adhere to their beliefs and values. Miserly in their handling of
finances, money is hoarded in order to deal with any future misfortunes. Emo-
tional expression is typically constricted, with the exceptional outburst of right-
eous indignation. The quality of their relationships with others, even intimates, is
stilted and formal.
The ARD-O scale includes items that capture these personality elements of the
obsessive-compulsive. Whereas the Axis I manifestation of Obsessive-Compulsive
Disorder involves a marked distress (appearing on ANX and DEP) associated with
intrusive obsessions and compulsions, the affectively constricted personality dis-
order may display no such elevations, aside from a moderate elevation on ANX-C.
Also, as a further indicator of this affective constriction, BOR-A may be quite low
in the personality disorder variant. These individuals do not like to submit to
others’ ways of doing things, and, as such, low scores on DOM are rare. However,
often the obsessive simply will refuse to work with others, rather than experience
the failure of others to cooperate, and, thus, DOM will not necessarily be elevated.
The disinterest in the affective quality of relationships as well as the formal and
somewhat stilted style of relating to others are typically expressed by low scores
on WRM.

Dependent Personality Disorder

The defining trait for this disorder is dependency; an enduring pattern of
submissiveness, neediness, and clinging behavior. Such people have great diffi-
culty making independent decisions, preferring to have others make important

186
 Use of the PAI in Diagnosis

decisions for them. They feel incapable and helpless on their own, fearing that they
will be left alone should their current dependency relationship end. If the rela-
tionship does end, they urgently seek another that will meet their needs for care
and support. In such relationships, they may suppress their own objective best
interests in order to ensure that they will not be abandoned.
The mean profile’for a group of 56 patients in the clinical normative sample
who were diagnosed with Dependent Personality Disorder is presented in Figure
6-9. One important feature of this profile is the low score on DOM, demonstrating
the interpersonal submissiveness characteristic of this group. Other suppressed
scales are also noteworthy. WRM scores will invariably be higher than DOM
scores, because interpersonal relationships are of such importance to the depen-
dent personality. MAN-G scores are typically low, pointing to the poor self-
esteem and feelings of inadequacy that drive these individuals to depend on oth-
ers. Low scores on AGG-V are typical, as expressions of anger would be inhibited
to avoid endangering all-important relationships. Among the BOR subscales, BOR-I
is particularly salient, as these individuals tend to submerge their sense of identity
within the context of a dependency relationship, leading to moderate elevations on
this subscale.
The LOGIT function for the diagnosis of Dependent Personality Disorder (as
contrasted with other clinical diagnoses) provides further understanding of scales
critical in assigning the diagnosis. The function (presented for illustration pur-
poses, rather than for use in routine clinical situations) included the following
weights for PAI scales:

.001(WRM) — .014(AGG-V) — .017(DOM) — .012(MAN-G) + 5.693

This function highlights the centrality of low scores on DOM, AGG-V, and
MAN-G in determining whether a person receives a Dependent Personality Disor-
der diagnosis; these negative loadings are reflective of the inverse relationships of
these scales to the diagnosis, rather than indicating variance suppression. The
small positive loading on WRM supports the conclusion that relationships are
important to the dependent, although the remaining scales in the function point
out the lengths to which this group, as opposed to other clinical groups, will go to
maintain these relationships.

Other Personality Disorders

Two personality disorders are described in an appendix to the DSM-IV manual:
Passive-Aggressive (Negativistic) Personality Disorder, and Depressive Personality
Disorder. Although officially classified as “Personality Disorder Not Otherwise

187
PAI Interpretive Guide

Specified,” these two disorders have either been in the nomenclature for many
years (i.e., Passive-Aggressive personality) or have been the focus of considerable
research (i.e., Depressive personality). Thus, some mention of the PAI indicators
for these disorders is warranted.
In the DSM-IV, Passive-Aggressive Personality Disorder has been relegated to
an appendix of “criteria sets provided for further study.” This disorder is charac-
terized by passive resistance and negativistic attitudes toward others who place
demands on the person. These demands are resented and opposed indirectly,
through procrastination, stubbornness, intentional inefficiency and memory
lapses. Such individuals tend to be sullen, irritable, and cynical, and they chroni-
cally complain of being underappreciated and cheated. On the PAI, the hallmark
combination is one of elevated PAR-R, signifying the hostility and resentment,
combined with low scores on DOM, indicative of the passivity element of the dis-
order. Scores on WRM also are typically low, as such people tend to be unsuccess-
ful in interpersonal relationships because of their capacity to evoke hostility and
negative responses from others. Also, any elevation on AGG-P would contraindi-
cate the disorder, as it suggests that anger and resentment are likely to be expressed
in a direct and overt manner; such elevations would not be typical of the passive-
aggressive individual.
Depressive Personality Disorder is also included in the appendix of criteria sets
for further study. This disorder is characterized by enduring depressive cognitions
and behaviors; some have proposed that this is essentially the same concept as
Dysthymic Disorder. Certainly, the PAI indicators for the two disorders would be
the same, as the primary distinction seems to be one of duration rather than qual-
ity of symptoms. Because of the centrality of depressed mood, DEP would be crit-
ical in identifying this disorder. The proposed criteria emphasize mood quality and
related cognitions, rather than the more somatic features of depression; thus,
DEP-C and DEP-A should be more prominent than DEP-P, with the former two
scales 1OT or more above the latter. Because the depression is chronic in nature,
rather than a reaction to external events, the DEP elevation should be somewhat
higher than STR, which might indicate more situational mood disruptions. Also,
some elevation on SUI would be expected, related to the chronic pessimism and
brooding nature of these individuals.

188
 Use of the PAI in Diagnosis

Table 6-5
Summary of Key Diagnostic Indicators for Personality Disorders
Diagnostic Elevation Suppression
consideration indicators indicators
Borderline Personality Disorder ‘BOR (all four subscales) PIM, RXR
DEP, particularly DEP-C
ARD, particularly ARD-T
SUI, STR, SOM
Antisocial Personality Disorder ANT-A (esp. for DSM), ANX-A, ARD-P
AGG-P, DRG
Narcissistic Personality Disorder MAN-G, DOM, ANT-E
Histrionic Personality Disorder WRM, SOM-S, ANT-E, AGG-V
BOR-A, PIM
Paranoid Personality Disorder PAR-R, PAR-H, SCZ-S, WRM
AGG-A
Dependent Personality Disorder WRM DOM, AGG-V, MAN-G
Passive-Aggressive Personality PAR-R DOM, WRM, AGG-P
Disorder
Schizoid Personality Disorder SCZ-S WRM, BOR-A

Avoidant Personality Disorder ARD-P DOM, WRM, ANT-S

Schizotypal Personality Disorder SCZ-S, PAR-P, ARD-P, WRM
SCZ-T
Obsessive-Compulsive ARD-O WARM, BOR-A
Personality Disorder

Substance Abuse Disorders

The key features of substance abuse and dependence involve the continuing
use of the substance despite significant problems associated with this use. The
diagnostic symptoms presented in the DSM-IV are similar across alcohol and the
different classes of drugs of abuse, although, for some substances, certain criteria
are less relevant (e.g., there is no known withdrawal symptom from LSD). These
symptoms include tolerance, withdrawal, failure to cut down despite repeated
efforts, social-role impairment, and continued use in the face of significant prob-
lems associated with the substance. Because these diagnoses are based heavily on
historical life event information rather than on present mental status, it is neces-
sary to establish that the difficulties have occurred within the preceding 12 months
in order to assign a diagnosis.

189
PAI Interpretive Guide

On the PAI, the obvious beginning points in considering a substance-abuse

diagnosis are the ALC and DRG scales; these scales are described in detail in chap-
ter 2. It should be remembered that a good deal of the information gathered on
these scales is historical (i.e., inquiries are made about events that may have hap-
pened in the past). As such, ALC or DRG, or both, can be elevated in people who
have had a substance-abuse problem in the past, but who are not currently drink-
ing or using drugs. It is certainly possible for a person to score in the vicinity of 70T
on either scale largely through historical information; however, a person who has a
current substance-abuse problem will tend to have scores that are quite elevated.
Because both ALC and DRG items directly address substance use, the scales are
susceptible to denial, a problem of concern to many in the substance abuse field.
Although the research literature supports the use of such a direct questioning
method, chapter 2 addresses in detail the issue of assessing substance abuse denial.
However, such indirect methods have limited ability to circumvent denial issues,
and asking directly about substance use is the most straightforward manner of
obtaining such information. If there is reason to suspect that marked denial of sub-
stance abuse may be occurring, supplementing the PAI with information from col-
lateral informants (e.g., spouse or family member) is recommended.
The mean profiles for alcohol and drug abuse samples (Morey, 1991) are pre-
sented in Figure 6-10. The alcoholic mean profile was markedly elevated on ALC
(i.e., 84T) and also was rather elevated on DRG, reflective of the prevalence of
polysubstance abuse among alcohol abusers. The highest clinical scale elevation
for the drug-abuse sample was on DRG (i.e., 80T). ALC approached 70T in this
group as well, consistent with the relatively high prevalence of alcoholism among
drug abusers. Other than the differences reflecting the primary substance of abuse,
the alcoholic and drug abuser profiles are quite similar, with the drug abuser sam-
ple displaying a slightly more elevated score on ANT that seems to reflect a greater
likelihood of antisocial behaviors among drug abusers. However, the relative speci-
ficity of the ALC-DRG elevations demonstrates that other scales will not necessar-
ily be elevated in these respondents; thus, a DEP elevation in a person with alco-
hol problems is not merely an artifact of the alcohol-related issues; instead, it
indicates a likely comorbid condition.
Figure 6-11 presents the mean profile from a sample of 229 methadone main-
tenance patients as reported by Alterman et al. (1995), and the mean profile of a
group of 30 alcoholics described by Boyle and Lennon (1994). The methadone
patients demonstrated a profile pattern similar to that of the general drug abuser
sample, with some differences. Alterman and colleagues found that their metha-
done patients were significantly higher on DRG and RXR than the drug patients,
whereas their scores were significantly lower on NIM, BOR, ANT, ALC Ul DOM:

190
 Use of the PAI in Diagnosis

PROFILE FORM FOR ADULTS - SIDE A 141 A B Cc D E y 2

ae - 70 E z
Es 70= 70- 70 6
: a : : a @—-—-® Alcohol abuse
$ 65 : ane 30 o—e Drug abuse

-
as
- T0-
60 30
45 : —
60. = >
aS
so- SP 5!
20- - | S i 3o- LL 25

a8 = a 25- 50- ; |
5 as- : : 20 |

w ene Re . |
a 20- |
| 40
;

ie ||

|
relouealos

5 10 :
Phy
la
Welar
adap
- = 40-
M =
ie Z =
a 2 3 4 5 6 7 8 9 10 1 A 8 D E y z
ICN INF NIM PIM SOM ANX ARD DEP = MAN PAR SCZ BOR ANT ALC DRG AGG sui STR NON RXR DOM WRM

PROFILE FORM FOR ADULTS - SIDE B

b B to} |
@-—=-@ Alcohol abuse
o—— Drug abuse

eC
(aCe
Te
Gc

a [-)

prveepererpereegeev
Pere

S3

wors) ee
ARD-O ARDP ARDT DEP-C DEP-A DEPP MANA MAN-G MAN‘ PARH PARP PARR SCZ-P sczS SCZT BORA BOR! BORN BORS ANT-A ANTE ANTS AGGA AGGV AGGP
SOMC SOM-S SOM-H ANX-C ANXA ANX-P
CONV SOMA HEAL COG AFF PHYS OBS PHOB TRAU COG AFF PHYS ACT GRND IRRI HYPE PERS RSNT PSYC soc THGT AFF ID NEG SELF ANT EGO STIM AGG VERB PHYS
STRS SYMP SYMP SYMP LEVL IOSY BLTY WIG CUTN MENT EXP DET DIS INST PROB REL HARM BEH CEN SEEK ATT AGG AGG
SYMP ZATN CONC SYMP SYMP SYMP ‘COMP

Figure 6-10. Mean PAI profiles for alcohol and drug abuse samples (Morey, 1991).

191
PAI Interpretive Guide

and WRM. Although the greater elevation on DRG (ie., signifying a greater degree
of drug related dysfunction) is likely to be a true reflection of a unique aspect of
the methadone sample, methodological variation may account for other observed
differences. In particular, Alterman et al. used a NIM cutoff of 92T to establish pro-
file validity, which resulted in 30% of the methadone patients being excluded from
further analyses; hence, they are not included in the mean profile presented in
Figure 6-11. The drug abuser profile presented in Figure 6-10 did not exclude any
respondents based on PAI-derived indicators, including the NIM scale. Thus, the
greater elevations on scales such as BOR and SUI in the drug abuser samples may
reflect profile distortion of the type measured by NIM. The Alterman study is
important in that it indicates that marked elevations on NIM are relatively com-
mon among severely drug-dependent individuals. The results also suggest that
among their sample of methadone patients, alcohol problems were relatively infre-
quent. This may be a result of screening practices (i.e., opiate addicts are poor risks
for methadone programs if they are also alcoholics), or it may reflect a narrowing
of the drug-taking repertoire among this type of addict.
The mean profile of alcoholics obtained by Boyle and Lennon (1994) (pre-
sented in Figure 6-11) displayed far more psychopathology than the alcoholic
mean profile reported in the PAI Professional Manual (Morey, 1991; shown in Fig-
ure 6-10). Although the ALC scale was markedly elevated in the Boyle and Lennon
sample (.e., M = 88T), so was nearly every other clinical scale. The mean score of
NIM was LOOT, which is above the recommended cutoff for profile validity. This
finding is apparently consistent with the results of Alterman et al. (1995) in doc-
umenting frequent NIM elevations among this population. These results may
reflect the point in treatment at which the PAI was administered; the elevated NIM
scores may result from having the test completed during detoxification, which
could hamper the straightforward interpretation of the profile information. These
findings indicate that extreme NIM elevations when the PAI is given during detox-
ification should be evaluated carefully; a readministration of the test following
completion of detox is recommended.
LOGIT functions derived for the alcohol and drug dependence diagnoses (as
contrasted with other clinical diagnoses) illustrate other elements of the substance
abuser profile. The function for alcohol dependence (presented for illustration
purposes, rather than for use in routine clinical situations) was as follows:

.052(ALC) + .004(ANT-A) — .011(BOR-S) + 1.321

The substantial contribution of ALC is not surprising, as this scale directly taps
signs and symptoms of alcohol dependence. The ANT-A loading reflects the behav-
ioral impairment often associated with alcohol problems. BOR-S appears to act as

192
 Use of the PAI in Diagnosis

PROFILE FORM FOR ADULTS-SIDEA 1 2 3 4 5 6 7 8 9 100 41 A

é.
B c D E Y z
= E E E
70= 70 70- 60- 2
wT

2 Z @ ——-@ Alcoholic (Boyle &

65 : Lennon, 1994)
= e——e Methadone
maintenance
(Alterman et al.,
1995)

9sasoog

— 60

1 2 3 4 5 6 7 8 9 10 1 8 c o € y Zz
ICN INF NIM PIM SOM ANX ARD DEP = MAN PAR SCZ BOR ANT ALC ORG AGG sul STR NON RXR DOM WRM

Figure 6-11. Mean PAI profiles for alcoholic and methadone maintenance samples. Boyle and
Lennon (1994) did not score the ICN scale.

a suppressor variable, indicating that ALC elevations in the absence of prominent

scores on BOR-S suggest that alcohol problems (as opposed to characterological
difficulties) are prominent. This function differs somewhat from the LOGIT func-
tion for drug dependence diagnoses, which was the following (presented for illus-
tration purposes, rather than for use in routine clinical situations):

.032(DRG) + .022(ANT-A) + .OO1(ANT-E) — .014(BOR-S) — .013(SOM-H) + 1.999

There are some similarities between this function and the alcoholism function
just described. The large contribution of the relevant substance abuse scale, the
significance of ANT-A, and the suppressor variable of BOR-S all replicate the pat-
tern seen in the alcoholism equation. However, the ANT-A loading is even larger
for the drug dependence group, indicating that it plays a larger role in discrimi-
nating drug abusers from other clinical respondents. Also, the more psychopathic
elements of ANT-E play some role in the drug dependence function, but they are
not useful in identifying alcoholics. Finally, the negative loading on SOM-H
appears to signify a disregard for health among drug-dependent individuals.

193
PAI Interpretive Guide

Other Diagnoses
There are a number of other diagnoses for which particular PAI configurations
are suggestive. For most of the following, additional information would be neces-
sary to supplement the PAI data for diagnostic purposes.

Intermittent Explosive Disorder

This disorder is officially an “impulse-control disorder not elsewhere classi-
fied.” It is characterized by discrete episodes of aggression that result in serious
assaults or destruction of property. The outbursts may be experienced as a “spell”
which is immediately followed by relief, but these individuals may later feel upset,
remorseful, or embarrassed by their behavior. On the PAI, the AGG scale includes
information directly relevant to such outbursts. The most typical configuration for
this diagnosis would involve a deep V-shaped pattern on AGG-A, AGG-V, and
AGG-P. Such individuals are temperamentally quite angry (AGG-A), and, when
they lose their temper, they have the potential for violence (AGG-P). However,
anger is typically not expressed verbally (AGG-V); rather, it tends to be suppressed
and then released in episodic explosions. Individuals who alternate between vio-
lent outbursts and extreme remorse over their behavior often will have DEP-C
scores at a comparable level to the AGG-P score.

Dissociative Identity Disorder (Multiple Personality)

The Dissociative Disorders are characterized by disruptions in normally inte-
grated areas of consciousness, including memory, identity, or perception of the
environment. The most dramatic manifestation of a dissociative disorder is multi-
ple personality (named Dissociative Identity Disorder in DSM-IV), which involves
the presence of two or more distinct identities residing within one individual.
Switches between these identities result in the individual forgetting important per-
sonal information associated with the other personalities.
The dramatic nature of these disorders makes them particular targets for
malingering, especially in situations where there may be financial or forensic gain.
Indeed, the few cases of multiple personality observed in the standardization of
the PAI tended to have elevated NIM scores, and a few of the NIM items are rele-
vant to dissociative phenomena. Alpher (1995) reported a study of 21 dissociative
disorder patients, who obtained a mean score on NIM of 75T. However, these
patients did not display any noteworthy elevation on the Malingering Index (M
score = 1.6 items), which suggests that a NIM elevation with the Malingering Index
within normal limits may signify a dissociative disorder. Alpher also noted marked

194
 Use of the PAI in Diagnosis

elevations on ARD-T (i.e., 80T), which is consistent with theoretical etiology of

these conditions. Finally, an elevation on BOR-I in the absence of elevations on the
other BOR subscales can also indicate unstable identity of the type seen in disso-
ciative disorders.

Organic Mental Disorders/Cognitive Disorders

The DSM-IV uses the term cognitive disorder to refer to conditions that lead to
significant changes in cognitive ability or memory, including dementia or delirium.
The presence of this type of dysfunction is best established by tests that tap cog-
nitive abilities, and the PAI does not do this. However, the discriminant validity of
the PAI makes it particularly useful in assessing the emotional and personological
aspects of an individual with suspected or confirmed cognitive impairment, and
such considerations are vitally important in diagnosis and treatment planning. On
the PAI, a person with central nervous system compromise will not obtain scale
elevations merely as a function of the organicity; in fact, over 40% of such patients
obtain no clinical scale elevations above 70T (Morey, 1991). Similarly, Schinka and
Vanderploeg (1995) found that scores on the PAI were largely unrelated to
measures of cognitive function from the Neuropsychological Screening Test in a
sample of 309 alcohol- and drug-dependent patients. Such findings indicate that
the PAI can be of particular help in differential diagnosis of emotional problems
(e.g., depression, impulsivity, or paranoia) that are commonly seen in this population.
The pattern of impairment seen in individuals with organic problems will vary
as a function of many factors, including the locus and extent of any damage as well
as the individual's premorbid functioning. This pattern is as true for the person's
emotional status as it is for cognitive status. However, some scales seem to elevate
with some regularity in these populations and, when observed, suggest the con-
sideration of organic impairment. The SOM scale is a relatively common elevation
among individuals with such disorders, with the SOM-C subscale often elevated
above the other two subscales. SOM-C elevations that are accompanied by high
scores on ALC are common among chronic alcoholics who are experiencing symp-
toms of Korsakoff’s syndrome or an alcoholic dementia; in fact, the SOM scale has
been found to be inversely related to verbal fluency in a sample of alcoholics
(Schinka & Vanderploeg, 1995). SCZ-T is another common elevation among the
cognitively impaired, as it can reflect the concentration and memory problems that
accompany clouded mental status. Finally, DEP scale elevations are seen with some
frequency in this group, which may reflect either a reaction to their impairment or
fatigue, sleep, or appetite problems associated with their organic disorder.

195
PAI Interpretive Guide

Family/Marital Difficulties
Although not considered to be formal diagnoses, relationships between part-
ners or family members can often be a major focus of clinical attention. On the
PAI, marital and family issues are most evident on NON and, to a lesser extent, on
STR. Elevations on NON that are 10T points above other scales are particularly
indicative that the respondent views the primary concerns as existing within the
marriage and/or the family. The clinician should pay particular attention to eleva-
tions on PAR or BOR, or both before interpreting the NON elevation in this man-
ner. These scales can indicate a generalized pattern of interpersonal bitterness, of
which the reported family difficulties are simply an example.

No Diagnosis
Finally, consideration should be given to the absence of diagnosable emotional
conditions as well as to their presence. What are the prerequisites of a “clean bill
of health” on the PAI? Several factors must be considered. First, there should be no
indication that defensiveness is playing a factor in suppressing the profile. Thus,
PIM scores should be low and there should be few, if any, items from the Defen-
siveness Index that are positive; establishing these indicators is described in detail
in chapter 5. Second, there should be no indications of problems on the clinical
scales. In general, this means that all clinical scales and their subscales should be
below 60T. Similarly, the treatment consideration scales (with the exception of
RXR) should also be below 60T. On the interpersonal scales, DOM scores should
be within | standard deviation of the mean (i.e., 40T to 60T), indicating that the
person is unlikely to be either overcontrolling or overly submissive in relation-
ships. WRM, on the other hand, should be above 40T although no maximum is
given, as there is no indication that marked elevations on WRM in isolation can
lead to difficulties.

196
 CHAPTER 7
EVALUATING SUICIDE POTENTIAL
‘The assessment of suicide potential is one of the most critical of all clinical
evaluation tasks. Unfortunately, it is also one of the most difficult tasks. Although
suicide is the eighth leading cause of death in the United States (National Center
for Health Statistics, 1992), it paradoxically is still a relatively rare event, involv-
ing 12.2 of every 100,000 people. Thus, it is a low base-rate condition, and such
conditions are extremely difficult for instruments with anything short of perfect
validity (Meehl & Rosen, 1955). The guidelines offered in this chapter reflect a
beginning point for identifying suicidal potential, but, given the difficulty of the
task and the critical nature of the issue, it is particularly critical to supplement the
PAI with additional information for clinical decisions in this area.
One source of information on the PAI that can be useful in evaluating suicide
potential, but which is likely to be overlooked, involves the demographic infor-
mation gathered on the answer sheet. Information such as gender, marital status,
age, and ethnic background provide actuarial information that, in turn, provides a
context for the available clinical information on the rest of the instrument. For
example, suicide attempt rates are roughly three times higher for women than for
men, but the rates of completed suicides are three to four times higher for men
than for women (Clark & Fawcett, 1992). Widowed individuals, particularly
younger ones, demonstrate the highest suicide rates with respect to marital status,
whereas married individuals, particularly those with children under the age of 18,
show the lowest rates. In general, suicide is more common among the elderly, and
this continues to be the trend, although rates have been increasing among adoles-
cents and young adults over the past few decades. Finally, suicide rates tend be
about twice as high for Whites as for non-Whites (National Center for Health Sta-
tistics, 1992). Each of these factors provides important information to consider
when interpreting the PAI test results.

Suicidal Ideation (SUI)

The obvious starting point on the PAI for evaluating suicide potential is the SUI
scale. As with the other scales of the PAI, SUI includes items that range in severity

197
PAI Interpretive Guide

from “thinking about death,” to “ever having contemplated suicide,” to “a current

serious consideration of suicide.” The latter item is placed near the end of the test
as a sort of sign out response, a final opportunity to alert caregivers to the desper-
ateness of the person’s need for help. Like other scales on the PAI, the content of
the SUI items is directly related to thoughts of suicide and related behaviors, and
individuals who wish to disguise suicidal intents can do so easily. However, the
large majority of individuals who completed suicide communicate their intent
(Shneidman, 1989), and the SUI scale offers an in-depth probe of any such intention.
It must be kept in mind that SUI is a suicidal ideation scale, rather than a sui-
cide prediction scale. As such, high scores indicate that a person has thought about
and is thinking about suicide; such scores do not necessarily mean that the person
will actually commit suicide or even attempt it. Ideation (in contrast to completion)
is fairly common in clinical settings, and, in fact, the raw score in the general pop-
ulation is not zero, implying that the average individual in the community is likely,
at some time, to have thought about suicide. Nonetheless, as suicide rates are fairly
low, it is clear that thinking about suicide and actually committing suicide are
quite different matters. There are obviously a host of other factors in addition to
ideation that determine whether or not a person will attempt suicide. However,
ideation still has a central role, as it is a necessary, but usually not sufficient, con-
dition for a completed suicide.
Scores on SUI in the average range are those below 60T, and these scores indi-
cate that the respondent is not reporting being disturbed by thoughts of self-harm.
It is fairly unusual for individuals in clinical settings to score below 45T. If such
scores are accompanied by other risk factors described later in this chapter, the
possibility of denial and masked ideation should be considered. Scores from 60T
to 69T are typical of clinical respondents. Scores in this range suggest that the per-
son is experiencing periodic, and perhaps transient, thoughts of self-harm. Such
people are pessimistic and unhappy about their prospects for the future. Although
such scores are common in clinical settings, specific follow-up regarding the details
of any suicidal thoughts and the potential for suicidal behavior is warranted.
SUI scores from 70T to 84T suggest recurrent thoughts related to suicide.
Although only a small percentage of individuals who entertain suicidal thoughts
actually act on them, a score in this range should be considered a significant warn-
ing sign of the potential for suicide. The presence of additional risk factors is of
particular concern for scores in this range, as such scores still reflect signifi-
cant ambivalence about suicide. For such individuals, an evaluation of their

198
 Evaluating Suicide Potential

life circumstances and available support systems is critical. As scores get

higher, (i.e., 85T to 99T) this ambivalence lessens, and the thoughts of suicide are
intense and recurrent. Such scores are typical of individuals placed on suicide pre-
cautions. As SUI scores become extreme (i.e., > LOOT), the person is likely to be
morbidly preoccupied with death and suicide, and many of the steps toward sui-
cide (e.g., giving away belongings, writing a note, formulating a specific plan) are
likely to have been completed. In such cases, the potential for suicide should be
evaluated immediately, and appropriate interventions should be implemented
without delay. Scores at these levels must be considered a significant warning sign
of the potential for suicide, regardless of the levels of elevation on other scales.

PAI Profile Configurations and Suicidal Behavior

The mean PAI profiles for three clinical groups with various types of suicidal
behavior are presented in Figure 7-1. The three groups, taken from Morey (1991),
include (a) patients on suicide precautions at the time of testing, (b) patients with
a suicidal gesture-attempt in the preceding 6 months, and (c) patients with a his-
tory of self-mutilating behavior. These three profiles are shaped quite similarly and
share a pronounced elevation on SUI; for each group, this scale is the highest on
the instrument. However, there are subtle differences between the profiles that
exemplify differences between the groups. The “current precautions” group dis-
plays the greatest elevations on SUI, but there are also elevations on many of the
neurotic spectrum scales (e.g., DEP and SOM). The scales in this group that would
suggest impulsivity and acting-out (e.g., BOR, ANT, ALC, DRG) are lower than
those for the groups with a history of parasuicidal or self-damaging behavior. The
“suicide history” group has a slightly lower score on SUI, consistent with the
notion that suicide risk, although heightened, is less imminent in this group. The
mean profile for a sample of patients with self-mutilating behavior, although gen-
erally resembling the other two profiles, is quite similar to the mean profile of
patients diagnosed with borderline personality (Morey, 1991). This result is not
surprising, as self-mutilation is considered by some to be a pathognomonic sign of
borderline personality. Although there is a certain artifactual contribution to this
resemblance (i.e., roughly half of the self-mutilating group received a diagnosis of
borderline personality), this pattern was also consistent in those self-mutilating
respondents who did not receive a borderline diagnosis. It is worth noting that of
the three groups portrayed in Figure 7-1, the self-mutilating group has the lowest
ratio of SUI to BOR (i.e., BOR is nearly as high as SUI).

199
PAI Interpretive Guide

14 2 3 4 5 6 8 9 10 d1 A B c D E Y Z
PROFILE FORM FOR ADULTS-SIDEA

se as ST gar roe fea es @——-© Current suicide

= A 65- - precautions
65 65 65: 65- : 7 35 ra @--++---@ Self-mutilation
: 70 70
a tags: 30 history
60
e— Suicide history

8 9 10 14 A 8
ICN INF NIM PIM DEP MAN PAR BOR ANT ALC — DRG AGG sul

PROFILE FORM FOR ADULTS - SIDE B

24-
24- @—=-® Current suicide
24-
24.
24- precautions
24-
20- @-------@ Self-mutilation
100 — = history
24-
24- 24 /20- 924. r e— Suicide history
24- =

90

80

| | 70

TScores {
91095

ah o- sal =
o- o- oOo = ? o-
— ar ‘oe prerrprerepecirpererpr
Peree
seer
eprerepereep
fore
er
eper
peer
pees
SOM-C SOM-S SOM-H ANX-C ANKA ANXP ARD-O ARD-P ARDT DEP-C DEP-A DEPP MANA MAN-G MAN PAR-H PARP PARR SCZP SCZ-S SCZT BORA BORI BORN BORS ANTA ANT-E ANTS AGGA AGGV AGGP
CONV SOMA HEAL COG AFF PHYS OBS PHOB TRAU COG =AFF PHYS: ACT GRNO IRRI HYPE PERS RSNT PSYC SOC THGT AFF 1D NEG SELF ANT —GO STIM AGG VERB PHYS
SYMP ZATN CONC SYMP SYMP SYMP COMP IAS ‘STRS SYMP SYMP SYMP LEVL IOSY BLTY VG =CUTN MENT EXP DET bIS INST PROB REL HARM BEH CEN SEEK ATT AGG =AGG

Figure 7-1. Mean PAI profiles for patients with a suicide history, patients on current suicide pre-
cautions, and patients with a history of self-mutilation (Morey, 1991).

200
 Evaluating Suicide Potential

The PAI Suicide Potential Index (SPI)

Suicidal ideation, although a central component in the evaluation of suicidal-
ity, is only one factor needed to assess acute risk. The fact is that suicidal ideation
is quite common among individuals presenting for treatment, yet the number of
completed suicides is far lower. Thus, ideation may be thought of as a “necessary
but not sufficient condition” for suicide; obviously, other variables are needed to
account for the differences between those who think about suicide but do not act
on it and those who make a serious suicide attempt.
The research literature suggests several risk factors for completed suicide that
can be assessed with the PAI. These features, incorporated from research investi-
gating completed suicide by Bongar (1991), Maris et al. (1992), Motto (1989) and
others, represent an attempt to form a risk constellation from PAI factors. A listing
of these features and their markers is provided in Table 7-1. This constellation, the
PAI Suicide Potential Index (SPI), consists of 20 features of the PAI profile that are
congruent with the research literature on risk factors for completed suicide. Some
evidence for the validity of this index can be obtained from Table 7-2, which com-
pares the means of the community and clinical samples with the three groups pre-
sented in Figure 7-1 (.e., individuals with a history of suicidal behavior, patients
currently on suicide precautions, and individuals with self-mutilating behavior).
The features on the Suicide Potential Index (SPI) tap a wide array of different
psychological problems, and, in general, respondents with globally elevated pro-
files will obtain high scores. The literature supports this approach, as completed
suicide in the absence of some form of emotional, physical, or behavioral problem
is quite rare (Beskow, 1979, Rich, Young, & Fowler, 1986; Robins et al, 1959). The
diagnoses noted most often are depression, alcoholism, and schizophrenia,
although most of these studies have not done an adequate job of assessing per-
sonality disorders. Although the index includes information pertinent to these
diagnoses, it also focuses on key behaviors or environmental circumstances that
contributed to elevated suicide risk. For example, impulsivity (operationalized by
BOR-S) heightens risk, because suicidal intent tends to be a transitional phenom-
enon. This is not to say that individuals completing suicide do so without think-
ing about it; on the contrary, most have thought of little else for some time. How-
ever, the intent itself will vary; a person may be very suicidal one day, and then,
the next day, reconsider this intent. Unfortunately, the tendency to act while the
feelings or impulses are strong will preclude the personal reflection that could lead
to a reduction in intent.

201
PAI Interpretive Guide

Table 7-1
The PAI Suicide Potential Index (SPI!)
Frequency in
Suicide risk factor PAI markers clinical sample*

Severe psychic anxiety ANX-C > 60T 49%

Severe anhedonia, degree of

depression DEP-A > 65T 44%

Global insomnia DEP-P > 60T 44%

Diminished concentration SCZ-T > 60T 39%
Indecision, OCD features, rigidity,
perfectionism ARD-O > 55T 42%
Acute overuse of alcohol ALC > 60T 36%
Panic attacks ANX-P > 60T 36%
Cycling affective disorder MAN-A > 55T 33%
No children in home, little chance of
rescue or interruption NON > 60T 44%
Concomitant drug abuse DRG > 60T 35%
Acute interpersonal disruption BOR-N > 65T 34%
Intensity of current stress STR > 65T 45%
Poor impulse control BOR-S > 60 34%
Anger, held in AGG-P minus AGG-V > 10T 24%
Hopelessness DEP-C > 65T 38%
Mistrust PAR-H > 60T 34%
Withdrawn, isolated WRM < 45T 44%
Worthlessness MAN-G < 45T 40%
Mood fluctuations BOR-A > 65T 38%
Somatic problems SOM-H > 55T 42%

ON = 1,246.

A host of other factors can indicate heightened risk for suicide: confusion and
indecision, hopelessness, feelings of worthlessness, significant health concerns,
and drug abuse. The SPI also includes various environmental features that
heighten suicide risk. Social stresses, interpersonal loss, and a lack of social sup-
port will both heighten the acuteness of the distress and also diminish the possi-
bility of intervening factors. Each of these factors is incorporated within the index.
The Suicide Potential Index (SPI) is scored by counting the number of positive
endorsements on the factors presented in Table 7-1. As the table demonstrates,
each feature in isolation is seen with some frequency in a general clinical popula-
tion. Nonetheless, Table 7-2 reveals that the mean number of positive SPI items in

202
 Evaluating Suicide Potential

Table 7-2
PAI Suicide Potential Index (SPI) Means and
Standard Deviations in Relevant Samples
Sample n M SD
Suicide precautions 46 10.35 Salo
Suicide attempt 95 9.90 5.34
Self-mutilating behavior Te 9.84 5.46
Clinical sample 1,246 7.74 5.30
Community sample 1,000 3.14 Be

patients presenting with self-destructive concerns are all significantly elevated in

comparison to community respondents as well as to clinical respondents as a
whole. Using 1- and 2-standard deviation points above the mean of clinical
respondents as thresholds, Suicide Potential Index (SPI) scores at or above 13
items and 18 items would suggest moderate and marked numbers of suicide risk
factors, respectively. Such scores, particularly in combination with elevated scores
on SUI, should alert the clinician to acute situational as well as ideational factors
related to suicide, and a prompt further evaluation of suicide risk is merited. For
convenience, Table 7-3 provides transformations of SPI scores to T scores, based
on the means and standard deviations of both the community standardization
sample and the clinical standardization sample (Morey, 1991).
Not surprisingly, scores on the Suicide Potential Index (SPI) are related to SUI
(r = .65 in the clinical sample, r = .63 in community respondents) even though the
Index includes no items directly tapping suicidal behavior or intent. However, it is
interesting to note that NIM serves as a mediator of this relationship. The index
demonstrates a substantially larger association with SUI (r = .60 in clinical respon-
dents) for individuals with NIM scores below 75T than it does for those with NIM
scores above 75T (r= .35). Although not conclusive, this interaction suggests that
the SPI will be affected by the global profile elevations typical with inflated NIM
scores; thus, the SPI should be interpreted with caution when NIM is elevated.
However, future research is needed to further explore the potential interactive
relationships between these direct and indirect measures of suicide risk.
To further refine an understanding of this index, a factor analysis (principal
axis extraction, followed by varimax rotation) was performed on the items com-
prising the index, using the data from the clinical standardization sample (Morey,

203
PAI Interpretive Guide

Table 7-3
T-Score Equivalents for the PAI Suicide Potential Index (SPI)
Standardized Against Community and Clinical Normative Samples ste my
Sele eslhrcntchase a a oo Nec aera serns ase neee oy Ti ee
T-score equivalent T-score equivalent
SPI score community norms? clinical norms?

0 40 35
1 43 37
2 46 39
3 50 A1
4 53 43
5 56 45
6 59 47
7 62 49
8 65 50
9 68 52
10 71 54
1 74 56
12 78 58
13 81 60
14 84 62
15 87 64
16 90 66
ili 93 G7

18 96 69
19 99 71
20 102 73
SNESROOU SSALEeT odGarantie nhwena BMGs? TELS 17,ST aCe i eetied ieee kia

1991). Four factors achieved an eigenvalue greater than one; Table 7-4 presents
noteworthy (i.e., above .30) loadings on these factors. Factor 1 appears to involve
a general distress factor associated with marked anxiety and depression (i.e., high
negative affect); Factor 2 involves moodiness, hostility, and interpersonal disrup-
tion (i.e., volatility); Factor 3 is marked by poor impulse control and substance
misuse (i.e., acting-out); and Factor 4 involves listlessness, apathy, and withdrawal
(i.e., low positive affect).
Simplified factor scores may be obtained for each index factor by summing the
number of items exceeding the cutoffs listed in Table 7-1 for those items that load
on each factor (as listed in Table 7-4). Thus, a total of 10 items are relevant to Fac-
tor 1, 11 for Factor 2, and so forth. On Factor 4, 1 point is deducted if MAN-A

204
 Evaluating Suicide Potential

Table 7-4
Factors of the PAI Suicide Potential Index (SPI)
Suicide risk factor Factor 1 Factor2 Factor3 Factor 4
Severe psychic anxiety rU 31
Severe anhedonia, degree of depression ey 42 32
Global insomnia 61
Diminished concentration .64 .30
Indecision, OCD features, rigidity,
perfectionism 43)
Acute overuse of alcohol 61
Panic attacks 72
Cycling affective disorder (—.39)
No children in home, little chance of rescue
or interruption .60
Concomitant drug abuse 78
Acute interpersonal disruption .59
Intensity of current stress 48
Poor impulse control 42 35
Anger, held in 32
Hopelessness 49 46 36
Mistrust .62
Withdrawn, isolated 35 41
Worthlessness 54
Mood fluctuations 47 Sy)
Somatic problems .50
% variance 32.4% 9.4% 6.9% 5.2%
Note. Factor 1 = High Negative Affect; Factor 2 = Volatility; Factor 3 = Acting-Out; Factor 4 = Low
Positive Affect.

exceeds 55T, as this item relates inversely to this factor. Table 7-5 presents a num-
ber of characteristics of these simplified factor scores. First, individuals identified
as at imminent risk for suicide tend to be most prominently elevated on features
of high negative affect, low positive affect, and volatility; acting-out behaviors
appear more related to parasuicidal gestures and self-mutilating behaviors. Second,
although the distress represented by the negative affect and volatility factors is
highly related to NIM, the feature of low positive affect is relatively independent of
NIM. Although not conclusive, this suggests that low positive affect may be of par-
ticular use in distinguishing between severity of suicidal potential when both NIM
and SUI are elevated. For example, an elevation on NIM might lead the clinician
to discount an elevated SUI score, believing that it may reflect a tendency to over-
dramatize personal misery. However, should the low positive affect features of

205
PAI Interpretive Guide

Table 7-5
Characteristics of the PAI Suicide Potential Index (SPI) Factor Scores
Suicide risk factor Factor1 Factor2 Factor3 Factor 4

Maximum score 10.0 ae) 3.0 4.0

Clinical sample, M (SD) 3.91(3..3) 4.4 (3.6) TO.) esis)
Community sample, M (SD) 1.55429) 1.4 (2.0) 0.4 (0.7) 0:5 (1e0)
Current suicide precautions, M (SD) 5.93.3) 6.4 (3.3) 0.9 (1.1) 2501(425)
Recent suicide attempt, M (SD) 52) (G3) 6.0 (3.5) {.22( 41) lle)
Self-mutilating behavior, M (SD) 4.9 (3.5) 6.0 (3.8) ASCs) TEGIE7)
Correlation with SUI (r) .63 .65 .26 49
Correlation with NIM (r) 0 0) oS ‘5
Note. Factor 1 = High Negative Affect; Factor 2 = Volatility; Factor 3 = Acting-Out; Factor 4 = Low
Positive Affect.

Factor 4 also be elevated, there may be an increased threat that the ideation may
unfold into action.
Using a 1-standard deviation cutoff above the mean scores for clinical respon-
dents may serve as a convenient shorthand to determine whether these factors are
elevated. Individuals obtaining these scores thus fall at or above the 84th per-
centile with respect to individuals presenting for treatment in a wide variety of
clinical settings. Application of this strategy would result in cutoffs of eight or
more items from Factor 1, eight or more items from Factor 2, three items from Fac-
tor 3, and three or more items from Factor 4. Individuals exceeding these cutoffs
on several of the factors, particularly with an elevation on SUI, and no elevation
on NIM, raise serious concerns about the risk for self-harm.
The Appendix details numerous correlates for the Suicide Potential Index
(SPI), of which selected results are presented in Table 7-6. This table reveals that
the Index is positively correlated with indicators of distress, depression, and poor
morale. On the PAI, the SPI displays association with BOR, DEP, and ANX, which
should be expected as these scales comprise part of the Index. However, the cor-
relations with comparable indicators from other instruments are nearly as high.
The SPI correlates highly with the total score and most subscales of the Suicide
Probability Scale (Cull & Gill, 1982), with Wiggins (1966) Depression and Poor
Morale content scales from the MMPI, and with the Beck Depression Inventory
(Beck & Steer, 1987).
The Suicide Potential Index (SPI) is also associated with various measures of
profile distortion. Because the SPI is highly associated with (and indeed, com-
prised of) various measures of distress, it is affected by the overall degree of

206
 Evaluating Suicide Potential

Table 7-6
Selected Correlates of the
PAI Suicide Potential Index (SPI) Total Score
Variable description Correlation with SPI score
PAI NIM ; 69
MMPI F 63
Rogers Discriminant Function score (clinical sample) ae
PAI PIM —.67
MMPI kK —.59
Marlowe-Crowne Social Desirability Scale —.36
Cashel Discriminant Function score (clinical sample) 10
PAI BOR .83
PAI DEP 82
PA! ANX 82
MMPI Sc 61
MMPI D 53
Wiggins MMPI Depression .83
Wiggins MMPI Poor Morale ETE
Beck Depression Inventory 63
Beck Hopelessness Scale 49
Suicide Probability Scale, Hopelessness .64
Suicide Probability Scale, Suicidal Ideation .65
Suicide Probability Scale, Negative Self-Evaluation 48
Suicide Probability Scale, Hostility 39
Suicide Probability Scale, Total Score .63
Note. MMPI = Minnesota Multiphasic Personality Inventory.

pathology represented by the profile. Thus, factors that produce distortion in a

pathological direction will inflate SPI scores, whereas factors that suppress pre-
sentation of pathology, such as defensiveness, will also suppress SPI scores. Thus,
the SPI score displays high positive correlations with NIM and MMPI F, and large
negative associations with PIM and MMPI K. However, indices of profile distortion
such as the Cashel Discriminant Function (described in chapter 5) and the Rogers
Discriminant Function (described in chapter 4) that are more independent of
global pathology are less correlated with the Suicide Potential Index (SPI). The
implication of this pattern of results is that NIM elevations will be common when
the SPI reaches critical levels. In such instances, the SPI score should not reflex-
ively be discounted, because the catastrophic cognitions typical of the suicidal
individual can also give rise to NIM elevations. An inspection of the Rogers Dis-
criminant Function (RDF) score can provide further information with which to
investigate this hypothesis. If the RDF score is also elevated, then the profile may

207
PAI Interpretive Guide

contain noteworthy exaggeration. However, if the RDF score is within normal lim-
its, then the SPI elevation merits serious consideration irrespective of the score on
NIM. |

a AS
WIM
« OSM a2
7 Gees ige’) le BD
wh 008 1OMM aniggfiy
‘ ys, naa aS Mares
Ly . ere = Bere eres :
Pa my ; sBulQgOH ene i ?
A vital lsbicwwS “place tividiedos
‘gn. Me olin hee sullsiin” ace Yeah:

hor LP
oll
bp beak = MI |, SMgrel aH. te “nS, TT Tt, laa lad fT Esra e . :

etaierdt’ fou Wy his uel rir ag) te pion, cui gir) were
at ai Pi

“ ghie¢@ dito
eee
ws i
bens»
+ apres : mar es 4
, a ha 7 ai 7 =x. /

Dinteih® Babette aks he

=
wo
 CHAPTER 8
EVALUATING POTENTIAL
FOR AGGRESSION
Ohne of the most important needs, yet one of the most daunting tasks, in clin-
ical assessment involves the prediction of aggression and violent behavior. Deci-
sions about aggressive potential are commonplace in many settings where the PAI
is administered; it is a critical consideration in treatment modality and hospital-
ization decisions, parole and inmate classification decisions, fitness for duty eval-
uations, custody examinations, and a variety of other core assessment tasks. Unfor-
tunately, this is a very difficult task; in general, psychological measures have
limited success in making predictions about highly specific behaviors at highly
specific points in time. This is particularly true of the prediction of aggressive
behavior (e.g., Megargee, 1970; Werner, Rose, Yesavage, & Seeman, 1984), where
judgments about imminent dangerousness often need to be made.
The assessment of aggressive behavior is complicated by the number of differ-
ent ways in which this construct is represented. Although a number of existing
conceptualizations of the construct are multidimensional, many of the dimensions
within these conceptualization are different. Predictably, empirical studies often
find that different measures of aggression correlate minimally with one another
(e.g., Govia & Velicer, 1985). In an effort to clarify distinct elements of the con-
struct, Spielberger et al. (1985) have delineated differences between anger, which
is an emotion; hostility, which involves a set of attitudes; and aggression, which
refers to destructive behavior directed toward other persons or objects. Although
anger and hostility both contribute to aggressive behavior, ultimately it is this
behavior that clinicians are asked to identify and predict.
One study of several multidimensional measures of aggression is particularly
informative about the nature of convergence on the major elements of this con-
struct. Riley and Treiber (1989) examined data from the Buss-Durkee Hostility
Inventory (Buss & Durkee, 1957), the Multidimensional Anger Inventory (Siegel,
1985), the Framingham Anger In/Out scales (Haynes et al., 1978), the Anger Self-
Report scale (Zelin, Adler, & Myerson, 1972), and the State-Trait Anger Scale
(Spielberger et al., 1983). The results of a combined factor analysis of scores from

209
PAI Interpretive Guide

these instruments yielded three factors: one general factor tapping the experience
of anger and hostility; and two factors identifying different modes of behavioral
expression: one factor involving verbal expression of anger, and another involving
physical, maladaptive forms of anger expression (e.g., fighting or smashing things).
The PAI can be useful as part of a comprehensive evaluation of aggressive
potential, anger, and hostility. The obvious starting point for the assessment of
aggression with the PAI is the AGG scale. Thus, this chapter describes the AGG
scale and its subscale configurations. However, there are additional indicators on
the PAI that can supplement the AGG scale in assessing aggressive potential, and
preliminary work in combining these indicators into an aggregated index is
described at the end of the chapter.

Aggression (AGG)
The AGG scale is a “treatment consideration” scale; it has no direct correspon-
dence to any DSM diagnostic category, but, instead, taps fundamental affects and
behaviors involved in many categories. Indeed, the DSM has been criticized for
failing to include any reasonable classification of problems related to anger, aggres-
sion, and their management (Deffenbacher, 1992). There are a variety of diagnos-
tic groups for whom anger control is central. Many of these groups are personality
disorders: Antisocial, Borderline, and Passive-Aggressive diagnoses all have signif-
icant issues surrounding anger management. Intermittent Explosive Disorder is
classified as an impulse control disorder, but failure to control anger is central.
Physical abuse of adults or children is an “other condition that may be a focus of
clinical attention” where anger management problems are involved. Thus, the AGG
scale provides useful information for a wide array of diagnoses, as is demonstrated
in the different configural guidelines described in chapter 6.
The PAI AGG scale was assembled to assess the three elements of aggression
identified by Riley and Treiber (1989) described earlier. One subscale is devoted
to a general assessment of temperamental anger and hostility, whereas the remain-
ing two assess the typical behavioral mode through which anger and hostility are
expressed. This combination of subscales permits assessment of a number of dif-
ferent aspects of aggression and its control (or lack of control). For example, strong
inhibition and suppression of anger (e.g., an individual who turns anger “inward”)
might be reflected in positive indications of the experience of anger, but sup-
pression of scales suggesting that this anger might somehow be expressed. The
composition and interpretation of these subscales is described in the following
sections.

210
 Fvaluating Potential
forAggression

Aggressive Attitude (AGG-A)

The Aggressive Attitude subscale was conceptualized to include general affects
and attitudes conducive to aggressive behavior (e.g., having a quick temper, or a
belief in the instrumental utility of aggression). The concept is distinct from the
expression of anger, in that individuals can be quite angry and yet not express it,
but, instead, suppress it or perhaps turn it inward. This concept resembles one of
anger-proneness: the tendency to become easily frustrated or irritated, to react
angrily when criticized or treated poorly by others.
Low scorers on AGG-A would be described as calm and placid individuals,
very slow to anger and quite tolerant and forgiving of others. Moderate elevations
(i.e., 60T to 70T) suggest individuals who are easily angered and frustrated. Oth-
ers may perceive them as hostile and readily provoked. Scores exceeding 7OT sug-
gest persons who are very prone to anger, often losing their temper with little
provocation. Such people may use anger to intimidate or control others and
become furious when others criticize or obstruct them in some way. However,
such anger may not be readily expressed; if AGG-A is elevated in this range and
AGG-V and AGG-P both lie at or below the mean, the individual is apparently sup-
pressing anger (look for low scores on DOM) or may be turning it inward (sug-
gested with DEP-C elevated and MAN-G low).

Verbal Aggression (AGG-V)

The Verbal Aggression subscale included items indicating a readiness to dis-
play anger verbally in a milder form, perhaps through sarcasm or criticism, and in
a more extreme form through yelling or abusive language. The critical aspect of
this mode of anger expression is its visibility; high scorers will display their
anger readily when it is experienced, rather than attempting to suppress or hide
it. Of the three subscales, AGG-V is probably most (but inversely) related to
efforts to control anger; low scorers make an effort to hide their anger from oth-
ers, whereas high scorers make little or no effort to control their outward expres-
sion of anger. Thus, the ease with which one can tell that a person is angry is
related to this subscale.
Low scores on AGG-V suggest individuals who prefer not to express their anger
when it is experienced; rather, such people tend to overcontrol their anger, keep-
ing it in to the best of their ability. Although control of anger is desirable, exces-
sive control can lead to passivity and withdrawal (look for low DOM scores),
intropunitive attitudes (suggested by DEP-C elevations or MAN-G suppression), or
episodic, poorly controlled outbursts of anger when it is released (suggested if
AGG-P is elevated). Scores on AGG-V that are moderately elevated (i.e., 60T to

211
PAI Interpretive Guide

70T) reflect individuals who are assertive and not intimidated by confrontation;
toward the upper end of this range they may be verbally aggressive (e.g., critical,
insulting, or verbally threatening) with little provocation. Elevations above 70T
suggest that these verbal outbursts are likely to be abusive; such people are gener-
ally not popular with others and are viewed as extremely hostile. It is likely that
others perceive such people as being angrier than they themselves acknowledge,
or of which they are even aware.

Physical Aggression (AGG-P)

The Physical Aggression subscale addresses past history and present attitudes
toward physically aggressive behavior. The questions inquire about a history of
fighting and physical violence during adulthood; it is unlikely that significant ele-
vations would result from conduct problems during adolescence in the absence of
problems during adulthood. This scale has a relatively hard floor (.e., community
adults typically obtain low raw scores). However, elevations in clinical samples are
relatively common.
Average scores on AGG-P indicate a person who reports being generally in con-
trol of angry feelings and impulses and who rarely expresses an angry outburst.
Moderate elevations suggest that losses of temper are more common and that the
person is prone to more physical displays of anger, perhaps breaking objects or
engaging in physical confrontations; such people probably attempt to maintain
close control over their anger, preferring to brood rather than risk expressing anger
in potentially destructive ways. As scores elevate above 7OT, this control often
lapses, resulting in more extreme displays including damage to property and
threats to assault others. Some of these displays may be sudden and unexpected,
as such individuals may not display their anger readily when it is experienced, par-
ticularly if AGG-V is below the mean. It is likely that others are intimidated by their
temper and the potential for physical violence and go to great lengths to avoid pro-
voking them.

AGG Full Scale Interpretation

As a full scale, AGG provides a global assessment of attitudinal and behavioral
features relevant to aggression, anger, and hostility. The item content ranges from
indicators of verbal assertiveness and poor anger control to violent and assaultive
behaviors. Average scores on AGG (i.e., < 60T) reflect a reasonable control over
the expression of anger and hostility; scores below 40T may indicate very meek
and unassertive individuals who have difficulty standing up for themselves, even

212
 Evaluating Potential
forAggression

when assertiveness is called for. Scores between 60T and 7OT are indicati
ve of
individuals who may be seen as impatient, irritable, and quick-temper
ed when
frustrated or crossed. Toward the upper end of this range such people
may be
increasingly angry and easily provoked by the actions of others around them.
Respondents with scores above 70T are likely to be chronically angry and will
freely express their anger and hostility. In this range, at least one subscale is likely
to be elevated and these scores should be examined to determine the typical
modality (e.g., verbal or physical) through which the anger is expressed. AGG
scores that are markedly elevated (i.e., > 82T) are typically associated with con-
siderable anger and potential for aggression. Such individuals are easily provoked,
and they may explode when frustrated; if AGG-V is low and AGG-P is elevated.
this explosion may come with little warning. Others are likely to be afraid of the
respondent’ temper, and close relationships will suffer as a result. There is proba-
bly a history of fights and other episodes where anger has clouded the respondent's
judgment, often leading to legal or occupational difficulties. Aggressive behaviors
are likely to play a prominent role in the clinical picture; such behaviors represent
a potential treatment complication that should receive careful attention in treat-
ment planning.

AGG Subscale Configurations

The particular configuration of subscales that drive AGG elevations is very
informative in determining the nature and severity of any aggressive behaviors that
may occur. The following sections describe some of the implications of different
combinations of elevations on the three subscales. A “high” score generally refers
to scores above 70T; however, it should be recognized that AGG-P scores are more
likely to be elevated than AGG-V scores in clinical samples.

AGG-A high, AGG-V high, AGG-P high

This pattern of responses suggests a person who is easily angered and who is
probably perceived by others as having a hostile, angry temperament. Such people
have difficulty controlling the expression of their anger, often making little effort
to control it and displaying it readily when it is experienced. They are likely to be
belligerent at relatively low levels of provocation and are not intimidated by con-
frontation. They tend to escalate to more extreme displays of anger, which might
include damage to property and threats to assault others. It is likely that those
around such people are intimidated by their temper and the potential for verbal
abuse or displays of physical violence. _

2S
PAI Interpretive Guide

AGG-A high, AGG-V high, AGG-P average

This pattern of responses suggests an individual who is seen as hostile and
angry, with marked difficulty controlling the expression of anger. When angry,
such people tend to display the anger immediately, rather than brood about the
perceived affront. They are unafraid of verbal confrontation, and they will tend to
be verbally aggressive at relatively low levels of provocation. More extreme dis-
plays of anger, including damage to property and threats to assault others, are pos-
sible, but they do not appear to be a significant part of the clinical picture; this pat-
tern suggests a person whose “bark is worse than their bite.” It is possible that the
more frequent venting of anger suggested by this pattern serves to prevent a more
dramatic and overwhelming loss of control over temper.

AGG-A high, AGG-V average, AGG-P high

This pattern reflects a very angry individual who struggles to maintain control
over his or her temper, but who tends to lose this control easily. When this hap-
pens, the person is likely to respond with more extreme displays of anger, includ-
ing damage to property and threats to assault others. When the difference between
AGG-P and AGG-V is substantial (i.e., = 20T), these displays may be sudden and
unexpected, as anger may not be displayed readily when it is experienced. Such
people attempt to hold their anger in, but lose control suddenly and explosively.
It is likely that those around such people are afraid of their unpredictability, their
potentially explosive temper, and the potential for physical violence.

AGG-A average, AGG-V high, AGG-P high

This is an unusual pattern of subscales. It suggests a person who believes that
he or she is assertive but generally in control of angry feelings and impulses,
expressing an angry outburst relatively infrequently. Nevertheless, the behavior of
such individuals suggests that the control over temper is nowhere near as complete
as they seem to believe. It is likely that the people who have experienced one of
the respondent’s angry verbal outbursts regard him or her with considerable wari-
ness; they are likely to view the respondent as being more hostile and angry than
the respondent believes is true of himself or herself. Anger is expressed readily
when it is experienced, and this may involve more extreme displays of anger,
including damage to property and threats to assault others.

PAI Profile Configuration and Aggressive Potential

Although AGG is the natural starting point for a determination of aggressive
potential, there are many other elements of the PAI profile that are informative in

214
 Evaluating Potential
forAggression

this regard. To illustrate some of these elements, mean PAI profiles for a number of
different groups who share problems with anger management are presented in
Figures 8-1 and 8-2. These groups, taken from the PAI Professional Manual (Morey,
1991), include (a) psychiatric patients with a history of assaultive behavior,
(b) patients on precautions for assaultiveness at the time of testing, (c) inmates
incarcerated for rape, and (d) men court-ordered for treatment because of spouse
abuse. These profiles share a number of features that are unique by comparison to
many other clinical groups. For example, all groups tend to have scores on SUI
that are lower than scores on AGG; this configuration is unusual in most clinical
settings and suggests that anger is more likely to be directed outward than inward.
A similar downward slope is seen in the relationship between DOM and WRM:
these individuals seek to control relationships through hostile means. Scores on
MAN-G are all above the mean, an unusual finding in clinical groups. Patterns of
failure and discomfort in social relationships (e.g., BOR-N, SCZ-S) and a history of
victimization (e.g., ARD-T) are also highlights of these configurations.

The PAI Violence Potential Index (VPI)

It is very difficult to estimate the short-term risk of violence in a given indi-
vidual, and the research literature has, thus far, provided a meager empirical base
for making such estimations. Often, the better predictors are demographic factors
such as age and gender, or a past history of violence, as opposed to clinical signs
or symptoms of the type measured by instruments such as the PAI. Nonetheless,
the PAI does address a variety of factors that have been shown to be of promise in
these type of assessments, including hostility and suspiciousness, agitation, or
social withdrawal (McNiel & Binder, 1994; Shaffer, Waters, @ Adams, 1994). Such
studies, in combination with the profile configurations presented in Figures 8-1
and 8-2, suggest a variety of risk factors for violence that can be combined into a
risk constellation using PAI configuration information. This constellation, the PAI
Violence Potential Index (VPI), consists of 20 features of the PAI profile that are
congruent with the available evidence on the prediction of dangerousness. These
features and their operationalization are described in Table 8-1. Some preliminary
evidence for the validity of this index can be obtained from Table 8-2, which com-
pares the means of the community and clinical samples with a variety of relevant
groups for whom dangerousness is a consideration.
The features on the Violence Potential Index (VPI) tap a wide array of psycho-
logical problems, but none of the features involve a reference to the absolute ele-
vation of AGG, as the index is designed to supplement AGG scores in assessing
dangerousness. The VPI is scored by counting the number of positive risk factor
endorsements in Table 8-1. As the table demonstrates, each risk factor in isolation

215,
PAI Interpretive Guide

4 5 6 7 8 9 10 14 A B c io) E Y z
PROFILE FORM FOR ADULTS-SIDEA 1 2 S

310 2er |ee7 70ES= 76 70.80 =: 2 :: @-+++++@ Assault history ;

ere 2 : : ce 3 : e——e Current aggression
- =i)
= 100
100 =
5
15
25- 3

20 :
= 90:
90 =

: = 80
80 a 3 ::
15- 10
35- =.

= 25- Z 5 -
}— - —{- 35- —— ee70
70 —= =
2 = “Er ae wis
S8 : i aba. oi
OE
wn 10 « =
30- =
=
3 = = 60
60 a
25- = =

= 50)
50 S

15 - ~ 40
40
ag ‘ e e
= 15- =

- | = A: .- == 30
on} - 5- 7 . ‘
30 = a
| a = 3
s- fs = = 10- -
oS 7 = P =

= o-
Loe Ee)
20
a a 3 4 5 6 rp 8 9 10 41 A 8 c Oo 3 mM, z
ICN INF NIM PIM SOM ANX ARD DEP MAN PAR sez BOR ANT ALC DRG AGG sul STR NON RXR DOM WRM

PROFILE FORM FOR ADULTS - SIDE B eee

@----++-@ Assault history
o— Current aggression
= =

verges

1
$909$

Peerprereprecepererpr

eepeceny
pre
SOMC SOM-S SOMH ANX-C ANKA ANXP ARD-O ARDP ARDT DEP-C DEPA DEPP MANA MANG MANI PAR-H PARP PARR SCZP SCZS SCZT BORA BOR BORN BORS ANT-A ANTE ANTS AGGA AGGV AGGP
CONV SOMA HEAL COG AFF PHYS OBS PHOB TRAU COG AFF PHYS ACT GRND IRRI HYPE PERS RSNT PSYC SOC THGT AFF 10 NEG SELF ANT EGO STIM AGG VERB PHYS
SYMP ZATN CONC SYMP SYMP SYMP COMP IAS STRS SYMP SYMP SYMP LEVL JIOSY BLTY VIG CUTN MENT EXP DET DIS INST PROB REL HARM EH CEN SEEK ATT AGG AGG

Figure 8-1. Mean PAI profiles for patients on current assault precautions and patients with a history
of assaultive behavior (Morey, 1991).

216
 Fvaluating Potential
forAggression

PROFILE FORM FOR ADULTS-SIDEA 1 2 3 4 5 6 7 8 9 10

bp bp °o |

@=——-@ |ncarcerated
rapists
@-+++*-*® Spouse abusers

Hewat
el
ia
lla

ee

rin
iniotico
teliaiebe
omar
ia
intone
et
neds
aera
0
1 2 3 4 5 6 7 8 9 10 11 A 8 c o E Y Zz
ICN INF NIM PIM SOM —ANX ARD DEP = MAN PAR scz BOR ANT ALC DRG AGG sul STR NON RXR DOM WRM

PROFILE FORM FOR ADULTS - SIDE B

@ —=- |ncarcerated
rapists
@--++**-®@ Spouse abusers
» S rs)
rid

veurlour

cyr=)

~6

>3

8
PPeei
perveprerrprerepeerep
pers
perry
SOM-C SOM-S SOM-H ANXC ANXA ANXP ARD-O ARDP ARDT DEP-C DEP-A DEP-P MANA MAN-G MANI PAR-H PARP PARR SCZP SCZS SCZT BORA BOR BORN BOR-S ANT-A ANT-E ANTS AGGA AGGV AGGP
CONV SOMA HEAL COG AFF PHYS OBS PHOB TRAU COG AFF PHYS ACT GRND IRR HYPE PERS RSNT PSYC SOC THGT AFF 10 NEG SELF ANT EGO STIM AGG VERB PHYS
SYMP ZATN CONC SYMP SYMP SYMP COMP IAS STRS SYMP SYMP SYMP LEVL JIOSY BLTY WG CUTN MENT EXP DET DIS INST PROB REL HARM BEH CEN SEEK ATT AGG AGG

Figure 8-2. Mean PAI profiles for incarcerated rapists and spouse abusers in treatment (Morey,
1991).

217
PAI Interpretive Guide

Table 8-1
The PAI Violence Potential Index (VPI)
Frequency in
Violence risk factor PAI markers clinical sample*

Explosive expression of anger AGG-P 15T higher than AGG-V 26%

Anger directed outward AGG 10T higher than SU/ 18%

Hostile control in relationships DOM 10T higher than WRM 26%

History of trauma without fearfulness ARD-T 15T higher than ARD-P 33%

History of antisocial behavior ANT-A > 70T 22%

Limited capacity for empathy ANT-E > 60T 24%

Sensation seeking ANT-S > 60T 25%
Rapid mood changes BOR-A > 70T 26%
Troubled close relationships BOR-N > 70T 27%
Impulsivity BOR-S > 70T 21%
Agitation MAN-A > 60T 24%
Self-centered MAN-G > 60T 19%
Negative world view NIM > 70T 19%
Hostile suspiciousness PAR-H > 70T 17%
Sense of persecution PAR-P > 70T 14%
Psychotic symptoms SCZ-P > 70T 8%
Social alienation SCZ-S > 70T 20%
Alcohol as disinhibitor ALC > 70T 27%
Drug abuse as disinhibitor DRG > 70T 24%
Estrangement from support system NON > 70T 22%

@N = 1,246.

is seen with some frequency in a general clinical population. Nonetheless, Table 8-

2 reveals that the mean numbers of positive index items in patients presenting with
issues that raise concerns of dangerousness are all elevated in comparison to com-
munity respondents as well as to clinical respondents as a whole.
T-score conversions for the Violence Potential Index (VPI) are presented in
Table 8-3, using the community and clinical samples as standardization refer-
ents. Using 1- and 2-standard deviation points above the mean of clinical respon-
dents as thresholds, VPI scores at or above 9 items and 17 items would suggest
moderate and marked risk of violent behavior, respectively. Such scores, particu-
larly in combination with elevated scores on AGG, should alert the clinician to
both historical and personality factors related to dangerousness, and further eval-
uation of the potential for assault is warranted. However, these scores are dramat-
ically above the mean of community respondents (84T and 121T, respectively),

218
 Evaluating Potential
forAggression

Table 8-2
PAI Violence Potential Index (VPI) Means and
Standard Deviations in Relevant Samples
Sample n M SD
Current assault precautions 73 6.63 4.52
Current antisocial behaviors 102 6.87 4.83
History of assault or violence 231 6.95 4.54
Convicted for assault 124 7.29 4.61
Convicted for rape 14 6.50 5.40
Clinical sample 1,246 4.40 3.98
Community sample 1,000 1.58 PAN!

and, as shown in Table 8-2, samples with a history of violence obtain mean scores
above 6 items (70T). Thus, even scores in this range raise the possibility that
potential for aggression might be a complicating factor in treatment planning.
The Appendix details numerous correlates for the Violence Potential Index
(VPI), of which selected results are presented in Table 8-4. This table reveals that
the VPI is positively correlated with indicators of anger, hostility, and poor judg-
ment. On the PAI, the Violence Potential Index displays its greatest associations at
the full scale level with BOR and ANT, which are both included as part of the VPI.
However, at the subscale level, the greatest association is with AGG-P, and this lat-
ter score is not directly included in the VPI score, as the Violence Potential Index
is intended to independently supplement the information provided by the AGG
scale and subscales. The VPI also correlates highly with indicators of hostility and
poor judgment on the MMPI, with Hare’s (1985) self-report measure of psycho-
pathic features, and with a diagnosis of Antisocial Personality Disorder arrived at
through structured interview.
The Violence Potential Index (VPI) is also associated with various measures of
profile distortion. Because the VPI is highly associated with (and indeed, com-
prised of) various measures of symptoms and character issues, it is affected by the
overall degree of pathology represented by the profile. Thus, factors (e.g., malin-
gering) that produce distortion in a pathological direction will inflate VPI scores,
whereas factors that suppress presentation of pathology (e.g., defensiveness) will
also suppress VPI scores. Thus, the VPI score displays high positive correlations
with NIM and MMPI F, and substantial negative associations with PIM and MMPI
K. However, the discriminant function-based indices of profile distortion that are
more independent of global pathology, such as the Cashel Discriminant Function
(described in chapter 5) and the Rogers Discriminant Function (described in

ZA2
PAI Interpretive Guide

Table 8-3
T-Score Equivalents for the PAI Violence Potential Index (VPI)
Standardized AgainstseCommunity and Clinical Normative Samples
Bhat ichabrneeet ether del vine acs aa Ri a s aSP ee I A a
T- score equivalent T-score equivalent
VPI score community norms? clinical norms?

0 43 39
1 47 42
52 44
:
3 57 46
4 61 49
5 66 51
6 70 54
7 75 57
8 79 59
9 84 62
10 89 64
11 93 67

12 98 69
13 102 72
14 107 74
15 ie 77
16 116 79
17 121 82
18 125 84
19 130 87
20 134 89
aN = 1,000. °N= 1,246.

chapter 4), are less correlated with the VPI. In the assessment of potential for vio-
lence, the guarded or defensive individual is typically of greater concern than the
exaggerating or malingering individual. The implication of this pattern of results
is that guarded responding of the type leading to elevations on PIM or the MMPI
K scale will also lead to the suppression of the VPI score. However, more subtle
indicators of defensiveness such as the Cashel Discriminant Function (CDF) score
or the Defensiveness Index (DEF) seem largely unrelated to VPI scores, suggesting
that the types of intentional positive dissimulation reflected by those indices will
not necessarily suppress VPI scores. Thus, for example, an individual with a
markedly elevated CDF score and a very low VPI total should not necessarily be
assumed to be concealing features of potential violence, particularly in the absence
of an elevation on PIM.

220
 Evaluating Potential for Aggression

Table 8-4
Selected Correlates of the
PAI Violence Potential Index (VPI) Total Score
Correlation with
Variable description , VPI score
PAI NIM .66
MMPI F 2
Rogers Discriminant Function score (clinical sample) PA.
PAI PIM —54
Defensiveness Index (clinical sample) .04
MMPI K =AG
Marlowe-Crowne Social Desirability Scale —.40
Cashel Discriminant Function score (clinical sample) 19
PAI BOR 74
PAI ANT 74
PAI AGG 65
PAI AGG-P JZ
Wiggins MMPI Psychoticism .65
Wiggins MMPI Hostility .63
Hare Self-Report Psychopathy (clinical sample) .67
Diagnostic Interview for Personality Disorders, Antisocial Personality .68

Note. MMPI = Minnesota Multiphasic Personality Inventory.

221
 4 .

; | ( sorevianated
4M
; : Tat} “J

) tt e ’

7 - Poe An.
' IAS
Ich (AW
3-04 104
ne MSCs Ute =r Cg
: Pee egw =f
T” - Tite tegtimiay Tigers teiQuh teeeutM
ue PITTS MOE .2voc TR] yiitenodst sol waned |

lamer gec” elements’ eileen

sestotual eincarr 7 Le
i
\L © oMNGRIIeledL vias
(Olrevel
, tet na — ie.
 CHAPTER 9
EVALUATING SPECIFIC
PSYCHOLOGICAL ISSUES

Tn addition to the many clinical issues that have been discussed in previous
chapters, there are a number of issues of importance in personality assessment that
fall within the more normative range of personality and behavior. This chapter
explores three specific areas where the PAI can provide valuable information: the
domains of self-concept, interpersonal style, and perception of the environment.

Assessment of Self-Concept
The view that people have of themselves can play a critical role in determining
their behavior. On the PAI, three clinical subscales are central in assessing three
important facets of the self-concept. One facet, self-esteem, reflects the evaluative
component of self-perception: Do people like themselves, or do they dislike them-
selves? Are they the way they want to be, or would they prefer to be very differ-
ent? The most direct measure of this self-facet on the PAI is the MAN-G subscale,
with high scorers manifesting high, perhaps even inflated, self-esteem. A second
facet, self-efficacy, reflects a sense of personal competence and perceived control
(Bandura, 1977). The DEP-C scale provides information relevant to the person's
perceived effectiveness: High scorers see themselves as ineffective in controlling
the environment to meet their needs. The third facet involves the stability of the
self-concept: Is it fixed and enduring, or is it unstable and highly vulnerable to
environmental events? For example, two people who each have quite high self
esteem may differ substantially in the secureness of this esteem; one person may
be capable of maintaining high self-esteem in the face of considerable evidence to
the contrary, whereas the other's self-esteem may be quite vulnerable to even the
slightest “blow to the ego.” BOR-I provides a measure of the stability of self-esteem,
with high scorers having the more variable and more vulnerable self-concepts.
The following sections discuss some of the implications for the self-concept
related to different configurations of these three scales. As the implications of the
individual scales in isolation have been discussed previously in chapter 2, these

225
PAI Interpretive Guide

implications will not be reviewed here; rather, the following descriptions address
the implications of 27 different configurations (e.g., high, average, or low scores
for each scale) for the self-concept of the respondent. In dividing the scales into
high, average, or low scorers, the distinctions are drawn in reference to a clinical
population rather than a community population; because certain self-related issues
(e.g., low self-esteem and doubts about self-efficacy) are so common in clinical
groups, it was thought that basing the differentiation on expected scores within
clinical samples would yield finer discriminations when these rules are applied to
such samples. Thus, the ranges differ somewhat across the three scales. For DEP-C
and BOR-I, the high range is considered to be 75T or above, whereas for MAN-G
the high range refers to 70T or above. The average range is between 51T and 74T
(inclusive) for DEP-C and BOR-I, whereas for MAN-G this range is between 41T
and 69T. Thus, the low range for DEP-C and BOR-I involves scores of 50T or
below, whereas the low range for MAN-G includes scores 40T or lower.

MAN-G high, BOR-I high, DEP-C high

The self-concept of individuals with this pattern may be poorly established
(i.e., their attitudes about themselves are likely to fluctuate wildly in response to
situational triggers). The self-perception will vary from states of very poor self-
esteem and severe self-doubt to periods of exaggerated confidence and overvalued
accomplishments. The episodes of positive self-esteem, when they occur, may be
defensive in response to feelings of emptiness and a lack of a sense of purpose. As
a result, this self-esteem will be quite fragile and is likely to plummet in response
to slights or oversights by other people. Associated with this instability in self-
esteem are corresponding shifts in identity and attitudes about major life issues.
Because of the normally inverse relationship between MAN-G and the other two
scales, such a pattern should be quite rare; normatively, it was observed in 0.1%
of clinical respondents and was never obtained in the community sample.

MAN-G high, BOR-I high, DEP-C average

This pattern suggests that the self-concept of the respondent appears to be
poorly established and is likely to fluctuate in response to the situation. For such
individuals, the self-perception will vary from states of poor self-esteem and uncer-
tainty to periods of exaggerated self-assurance and overvalued accomplishments.
The episodes of positive self-esteem may be a defense against marked feelings of
emptiness; overvaluing their everyday accomplishments may help give such indi-
viduals a sense of purpose that they would otherwise lack. However, such feel-
ings of self-importance are likely to be tenuous, and self-esteem during these
times may plummet dramatically in response to slights or oversights by other

224
 Evaluating Specific Psychological Issues

people. Corresponding shifts in identity and attitudes about goals and values are
likely to be associated with this instability in self-esteem. Whereas this pattern was
never observed in the community normative sample, it was obtained in 0.9% of
clinical respondents.

MAN-G high, BOR-L high, DEP-C low

This pattern suggests that the self-concept is poorly established and highly
responsive to the nature of external events. Most likely, the self-perception will
vary from states of inflated self-esteem and overvalued accomplishments to times
of intense uncertainty, resentment, and anger. The generally positive self-esteem
may be a defense against marked feelings of superficiality and emptiness and a lack
of a sense of purpose. One would anticipate that the favorable self-evaluation is
likely to be quite fragile and may internally plummet in response to slights or over-
sights by other people; overtly this might take the form of considerable anger
directed at these people. Corresponding shifts in goals, values, and attitudes about
major life issues may be associated with these evaluative swings. Generally, this is
a very rare configuration, occurring in 0.1% of clinical respondents and never
observed in the PAI normative community sample.

MAN-G high, BOR-I average, DEP-C high

This unusual pattern suggests a self-concept involving a generally positive self-
evaluation, but there are doubts about personal effectiveness in many situations
and a pessimistic view of future prospects. It is possible that the positive self-
esteem may be defensive in response to these feelings of pessimism and a sense of
inadequacy, and these feelings may alternate as a function of the current situation.
As a result, self-esteem will tend to be fragile and very reactive to the quality of the
individual’ interactions with other people. The pessimism may result from a sense
that the external environment consistently provides obstacles to the accomplish-
ment of aims and goals. Responsibility for any setbacks is, thus, likely to be attrib-
uted externally. The pattern is uncommon, obtained in 0.2% of clinical respon-
dents and never obtained in the community normative sample.

MAN-G high, BOR-I average, DEP-C average

The self-concept of such individuals appears to involve a generally positive, but
probably fluctuating, self-evaluation. The positive self-esteem may be a defense
against feelings of uncertainty and self-doubt. Thus, the self-perception is likely to
be vulnerable and may drop dramatically in response to scrutiny or criticism by
other people. Self-esteem may be maintained in such situations through attributing
responsibility for setbacks to some external cause, rather than to personal failings.

222
PAI Interpretive Guide

This pattern is more common in clinical populations (1.0% of respondents) than

in normal populations (0.2%).

MAN-G high, BOR-I average, DEP-C low

This self-concept is likely to involve a generally positive and, at times, perhaps
uncritical self-evaluation. There may be some variability and uncertainty associ-
ated with this self-concept, particularly in the face of scrutiny or criticism from
others. Nonetheless, self-esteem is likely to be maintained in such situations
through attributing responsibility for setbacks to some external cause, rather than
to personal failings. At such times, anger directed at the source is a more likely
result of such criticism than inner self-doubt. This pattern is somewhat more com-
mon in clinical respondents (1.3%) than in normal respondents (0.6%).

MAN-G high, BOR-I low, DEP-C high

This pattern is a contradiction in terms: a highly positive self-evaluation in
combination with a pessimistic view of the personal prospects for the future and
doubts about efficacy in dealing with the challenges of the future. Because BOR-I
denotes a stable sense of self-worth, this pessimism must result from a sense that
the external environment consistently provides exceptional obstacles to the accom-
plishment of personal aims and goals. Responsibility for any setbacks is, thus, likely
to be attributed externally. Because of the inherent contradictions in this pattern, it
is not surprising that it was never obtained in either the normative clinical or the
community samples.

MAN-G high, BOR-I low, DEP-C average

This self-concept pattern appears to involve a generally positive self-evaluation,
which may be occasionally punctuated by periods of pessimism. Such individuals
describe approaching life with a clear sense of purpose and distinct convictions.
The generally stable positive self-evaluation may be vulnerable during times when
the external environment is perceived as providing obstacles to the accomplish-
ment of personal aims and goals. However, given the reasonably stable sense of
self-worth implied by the low BOR-I, responsibility for any setbacks is more likely
to be attributed externally than to personal failings. Because of the contradictions
between the stability of the self-concept and its positive and negative elements, the
pattern is unusual; it was obtained in only 0.2% of clinical respondents and 0.1%
of community respondents.

MAN-G high, BOR-I low, DEP-C low

This self-concept pattern appears to involve a highly positive and, at times,
perhaps uncritical self-evaluation. Such people describe themselves as effective

226
 Evaluating Specific Psychological Issues

and competent in most domains, including having a well established sense of pur-
pose in life and distinct convictions. The high self-esteem is probably quite robust
in the face of insults, given the reasonably stable sense of self-worth implied by the
low BOR-I. To maintain this self-esteem, responsibility for any setbacks may be
more likely to be attributed externally than to personal failings. This pattern is
obtained in about the’same proportions in community samples (1.9%) and clini-
cal samples (1.5%).

MAN-G average, BOR-I high, DEP-C high

This pattern implies a self-concept that is poorly established and likely to fluc-
tuate. The self-perception will vary from states of harsh self-criticism and severe
self-doubt to periods of relative self-confidence and intact self-esteem. It also will
probably vary most as a function of the current status of close relationships; apart
from a sense of identity established from these relationships, such people are likely
to feel ineffectual, unfulfilled and inadequate. As a result, self-esteem is quite frag-
ile and is likely to plummet in response to slights or oversights by other people.
Corresponding shifts in identity and attitudes about major life issues are likely to
be associated with these drops in self-esteem. Although uncommon in the com-
munity sample (0.2%), this pattern was fairly common in clinical respondents
(2%),

MAN-G average, BOR-I high, DEP-C average

This pattern suggests that the self-concept is imperfectly established, with con-
siderable uncertainty about major life issues and goals. Although outwardly such
individuals may appear to have adequate self-esteem, this self-esteem is likely to
be fragile, and the individuals may be inwardly self-critical and self-doubting. Self-
esteem may be particularly vulnerable to slights or oversights by other people, aris-
ing from a self-image that depends unduly on the current status of close relation-
ships. This pattern is nearly six times more common in clinical samples (4.7%)
than in the normative community sample (0.8%).

MAN-G average, BOR-I high, DEP-C low

This configuration, like the preceding pattern, reflects a self-concept that is
imperfectly established, with considerable uncertainty about major life issues and
goals. On the surface, such individuals are likely to appear to be optimistic and to
have adequate self-esteem; however, this self-esteem may be fragile and particu-
larly vulnerable to interpersonal disruption. In part, this vulnerability may arise
from a self-image that depends on the current status of close relationships. The
blame for any interpersonal problems may be repeatedly attributed outwardly in

227
PAI Interpretive Guide

an effort to preserve self-esteem. This pattern is uncommon in both clinical (0.7%)

and community (0.2%) samples.

MAN-G average, BOR-I average, DEP-C high

This pattern suggests a self-concept that involves a generally negative self-
evaluation, which may vary from states of harsh self-criticism and self-doubt to
periods of relative self-confidence and intact self esteem. This fluctuation is likely
to vary as a function of current circumstances, with pessimism and self-doubt pre-
dominating at the present time. Under stress, such persons are prone to be self-
critical, dwelling on past failures and lost opportunities and having considerable
uncertainty and indecision about their plans and goals for the future. Given this
self-doubt, they tend to blame themselves for setbacks and to see prospects for
future success as dependent on the actions of others. This is a relatively common
pattern, although roughly six times more prevalent in clinical (5.9%) than in
community (1.0%) samples.

MAN-G average, BOR-I average, DEP-C average

This configuration, representing as it does the average ranges for all three
scales in clinical populations, is not surprisingly the most common configuration
among clinical respondents. The self-concept of such respondents involves a self-
evaluation that has both positive and negative aspects. Their attitudes about them-
selves may vary from states of pessimism and self-doubt to periods of relative self-
confidence and self-satisfaction. Some fluctuation in self-esteem may be observed
as a function of current circumstances; during stressful times, in particular, such
individuals may be prone to be self-critical, uncertain, and indecisive. However,
these fluctuations would not be as extreme as those noted in many clients in clin-
ical settings; in fact, such instability is experienced by most adults. As noted, this
is the most common configuration in clinical respondents (26.2%), but also the
second most common in community adults (15.4%).

MAN-G average, BOR-I average, DEP-C low

This pattern suggests a self-concept involving a generally positive self-evaluation.
Such people are typically confident, resilient, and optimistic, although the self-
esteem may be fairly sensitive to changes in current circumstances. During times
of stress, these people inwardly may be troubled by more self-doubt and misgiv-
ings about adequacy and competence than is readily apparent to others. Reactive
changes in self-esteem may be accompanied by uncertainty about goals, values,
and important life decisions. This is a common pattern, although obtained some-
what more frequently in community samples (11.0%) than in clinical samples
(9.1%).

228
 Evaluating Specific Psychological Issues

MAN-G average, BOR-I low, DEP-C high

The self-concept of such respondents appears to involve a fixed, rather nega-
tive self-evaluation. People with this pattern are likely to be pessimistic and self-
critical and to dwell on past failures and lost opportunities. They may tend to
attribute blame for any setbacks internally, rather than externally, but their self-
esteem is still likely to be responsive to positive feedback from others. This is a
very uncommon pattern in both clinical (0.2%) and community (0.1%) samples,
perhaps due to the inherent contradictions given the picture of stable self-concept,
combined with very low self-efficacy, yet average self-esteem.

MAN-G average, BOR-I low, DEP-C average

This pattern suggests a self-concept that involves a reasonably stable and pos-
itive self-evaluation which, as is the case with most individuals, may be occasion-
ally punctuated by brief periods of self-doubt or pessimism. Such people describe
approaching life with a clear sense of purpose and distinct convictions, with a well
articulated sense of who they are and what their goals are. This identity should be
robust even in the face of significant stresses in their environment. This positive
and sturdy self-view is more than twice as common in community (9.6%) as in
clinical (3.9%) populations.

MAN-G average, BOR-I low, DEP-C low

This pattern reflects the “picture of health” with respect to self-concept, indi-
cating a stable positive self-evaluation. It reflects a confident and optimistic person
who approaches life with a clear sense of purpose and distinct convictions. These
characteristics are valuable in that they allow the person to be resilient and adap-
tive in the face of most stressors. Such people are reasonably self-satisfied, with a
well articulated sense of who they are and what their goals are. This is the most
common pattern among community adults (39.2%), where it is roughly three
times more frequent than in clinical populations (13.1%).

MAN-G low, BOR-I high, DEP-C high

This pattern suggests a self-concept that is poorly established, although harsh
self-criticism and severe self-doubt seem to predominate. The self-perception will
tend to vary as a function of the current status of close relationships; apart from a
sense of identity established from such relationships, the respondent is likely to
feel incomplete, unfulfilled and inadequate. As a result, self-esteem is quite fragile
and is likely to plummet in response to slights or oversights by other people. Cor-
responding shifts in identity and attitudes about major life issues, which will
largely be derived from those held by important others, are associated with this
instability in self-esteem. This pattern is relatively common in clinical respondents
(4.7%), although is is uncommon in the general population (0.6%).

229
PAI Interpretive Guide

MAN-G low, BOR-I high, DEP-C average

The self-concept represented by this pattern appears to be poorly established,
and self-criticism and adequacy concerns seem characteristic. Although low self-
esteem may not be obvious to others, the self-perception will tend to depend
greatly on the current status of close relationships. Apart from a sense of identity
derived from such relationships, such people are likely to feel uncertain and unful-
filled. Self-esteem is probably generally low and particularly sensitive to slights or
oversights by other people. Corresponding shifts in identity and attitudes about
major life issues are associated with any such shifts in self-esteem. This relatively
uncommon pattern is seen more in clinical respondents (1.3%) than in commu-
nity adults (0.3%).

MAN-G low, BOR-I high, DEP-C low

This very unusual pattern suggests a self-concept that is imperfectly estab-
lished, with considerable uncertainty about major life issues and goals. Although
outwardly such people may appear to have adequate self-esteem, inwardly they are
likely to be troubled by self-doubt and misgivings about adequacy. Some of the
self-image may be derived through close relationships; although this strategy may
be working for the respondent at present, the self-esteem may be quite fragile and
vulnerable to slights or oversights by other people. This pattern is very infrequent
in both clinical (0.1%) and community (0.1%) populations, as it is unusual for
people to maintain optimism in the face of marked identity confusion.

MAN-G low, BOR-I average, DEP-C high

This self-concept pattern involves a generally harsh, negative self-evaluation.
Such people are prone to be self-critical and pessimistic, dwelling on past failures
and lost opportunities with considerable uncertainty and indecision about plans
and goals for the future. Given this self-doubt, they tend to blame themselves for
setbacks and to see any prospects for future success as dependent on the actions
of others. This pattern is fairly common in clinical populations (6.3%), although
rare in the general population (0.7%).

MAN-G low, BOR-I average, DEP-C average

This self-concept pattern involves a rather negative self-evaluation. Such peo-
ple are likely to be self-critical, not handling setbacks very well and blaming them-
selves for past failures and lost opportunities. They may inwardly be more trou-
bled by self-doubt and misgivings about their adequacy than is apparent on the
surface. They may tend to play down their successes as a result, and probably see
such accomplishments as heavily dependent on the efforts or good will of others.

230
 Evaluating Specific Psychological Issues

This is a relatively common pattern in clinical groups (7.0%), although not

uncommon in the general population (4.7%).

MAN-G low, BOR-I average, DEP-C low

This pattern suggests an individual who is reasonably comfortable with him-
self or herself, although self-esteem may be rather reactive to changes in current
circumstances. Such a person may inwardly be troubled by more self-doubt and
misgivings about adequacy than is readily apparent to others, but be relatively
effective in masking these doubts. This person may tend to discount personal suc-
cesses, seeing such accomplishments as heavily dependent on the efforts or good
will of others. This is a configuration more commonly observed in clinical groups
(1.2%) than in community samples (0.3%).

MAN-G low, BOR-I low, DEP-C high

The self-concept suggested by this pattern involves a fixed, negative self-
evaluation. Such people are likely to be harshly self-critical and to dwell on past
failures and lost opportunities. Plagued by self-doubt, they are likely to attribute
blame for any setbacks internally and any successes are dismissed as either good
fortune or as the result of actions by others more competent than themselves. Self-
concepts with such low self-efficacy and such stability are rare, constituting 0.1%
of clinical respondents; this pattern was never observed in the normative sample.

MAN-G low, BOR-I low, DEP-C average

The self-concept of these respondents appears to involve a fixed, rather nega-
tive self-evaluation. Such people are likely to be self-critical and to focus on their
shortcomings and failures. Similar to those in the preceding group, these individ-
uals may be inwardly troubled by self-doubt and misgivings about adequacy to a
greater extent than is apparent to others. They may dismiss their successes as
either good fortune or as the result of the efforts of others, being unwilling to give
themselves credit for their accomplishments. This pattern is not uncommon in
either normative (3.2%) or clinical (2.7%) samples.

MAN-G low, BOR-I low, DEP-C low

The self-concept described by this pattern involves a generally stable, unas-
suming self-evaluation; such people report approaching life with a clear convic-
tions and distinct (but likely modest) ambitions. Although outwardly such people
will appear to have reasonable self-esteem, inwardly they may be troubled by self-
doubt and misgivings about personal competence. Although optimistic about the
future, they may tend to minimize their role in any anticipated successes, tending

Zod
PAI Interpretive Guide

to view achievements as either good fortune or as the result of the efforts of oth-
ers. Fairly common in the general population (9.8%), this configuration is less
common in clinical respondents (2.2%).

Assessment of Interpersonal Style

An individual’ interpersonal style constitutes a significant portion of his or her
personality, The way that a person relates to others is certainly associated with
overall adjustment; however, there are a variety of ways in which people interact
with one another, and there is no one “healthy” style that is necessary for personal
effectiveness. Nonetheless, a person's style can interact with aspects of the situation
to produce outcomes both desirable and undesirable; for example, a person who
is aloof and retiring may do quite well in a job that requires computer program-
ming skills, but the same person may be uncomfortable and ineffective at cocktail
parties. In a clinical context, this person might respond quite differently to some
therapies or therapists than another individual who is more outgoing. The inter-
personal style thus represents a significant aspect of the personality that can medi-
ate a number of clinical concerns.
On the PAI, the two interpersonal scales, DOM and WRM, represent the core
of the interpersonal style assessment. The selection of these two dimensions was
based on the interpersonal circumplex model originally formulated by Leary
(1957) and elaborated upon by many others (e.g., Benjamin, 1973; Kiesler, 1983;
Wiggins, 1982). The basic circumplex model of interpersonal behavior involves
two orthogonal dimensions that, when considered in combination, characterize
one’s preferred manner of interacting with others. Combining these two scales
forms four quadrants: a warm, dominating quadrant; a cold, dominating quadrant;
a cold, submitting quadrant; and a warm, submitting quadrant. For each of the
quadrants, stereotypic behaviors can be described. A prototype for the “warm con-
trol” quadrant would be parenting behavior, which is an example of controlling
others while being interested in maintaining the attachment relationship. An
impersonal, superordinate-subordinate relationship would characterize the “cold
control” quadrant, whereas “cold submission” would be characterized by a person
who is submitting to others unwillingly, perhaps a passive-aggressive reaction.
Finally, the prototype for the “warm submission” style would be dependency, a
person who is very interested in maintaining the attachment relationship and will-
ing to submit in the context of that relationship in order to maintain it.
A particularly interesting aspect of the theory surrounding the interpersonal
circumplex is the principle of complementarity. This principle governs the expected

252
 Evaluating Specific Psychological Issues

nature of interpersonal transactions within the circumplex; every interpersonal

behavior has a complement, which is the natural interpersonal reaction to a given
event or transaction. Complementary behaviors are the same on the warmth
dimension and on the opposing end of the dominance dimension. For example, if
a person controls people (dominance) in a friendly (warmth) Way, as in a parent-
ing relationship, the complementary reaction would be for people to submit in a
friendly way. On the other hand, if a person controls others in a hostile and uncar-
ing way, the complementary reaction would be to submit, but in a hostile manner.
This property of the interpersonal dimensions is useful in allowing one to predict
the types of interpersonal behavior a person is likely to evoke in others.
The DOM and WR\M scales have nearly identical distributions in clinical and
normal respondents. This supports the conclusion that the two scales capture
variation across a normal personality trait, and that variability on these dimensions
exists as widely within normal populations as it does within individuals present-
ing for treatment. On these scales, high scores may be problematic, and low scores
may also reflect problems; the interpersonal scales are probably the most bipolar
of all the PAI scales, in that the low and high extremes are equally interpretable
and have equal potential for problems. Generally, it appears that obtaining higher
scores on WRM is preferable, as this scale is typically positively correlated with
indicators of favorable adjustment. However, high scores on WRM could well
reflect a person who is sacrificing too much to maintain attachment relationships
and is, thus, ineffective in interpersonal relationships in many ways.
The following sections discuss some of the implications for the assessment of
interpersonal style related to different configurations of DOM and WRM. Five score
ranges are provided for these scales: Very low (< 35T), Low (35T to 44T), Average
(45T to 55T), High (S6T to 65T), and Very high (> 65T). Because of the similar
distributions for the two groups, these ranges are applicable to both clinical pop-
ulations and community populations. Interpretations for these scales are some-
times supplemented with references to BOR-N, which can shed further light on
the nature of any difficulties that may be suggested by the interpersonal scales. In
addition, there are several other sources of interpersonal information on the PAI,
most notably SCZ-S (i.e., social disinterest), ARD-P (i.e., social anxiety), ANT-E
(i.e., capacity for empathy), PAR-R (i.e., interpersonal bitterness), and PAR-P (i.e.,
suspiciousness and touchiness).

DOM very high, WRM very high

This interpersonal style seems best characterized as involving very strong
needs for affiliation and attention. This may result in rather extreme behavior, such
as controlling and interfering with the social interactions of others in order to meet

255
PAI Interpretive Guide

the respondent's own needs. Such people may be seen by others as being attention-
seeking and dramatic. The needs for attention and affiliation can be so strong that
the quality of social interactions may be relatively unimportant as compared to
their quantity. As a result, such people may be uninhibited in seeking any oppor-
tunity to interact with others, as long as the interaction permits them to maintain
some control over the relationship. This control, perhaps intended as a protective
measure by the respondent, may be viewed as smothering by others.

DOM very high, WRM high

This interpersonal style seems best characterized as involving very strong
needs for attention and affiliation. These needs may result in the respondent
being perceived by others as controlling; this may take the form of interfering
with the social interactions of others in order to meet personal needs. The need
for attention may be sufficiently strong that any opportunity to interact with oth-
ers will be acted on, as long as the interaction permits the respondent some con-
trol over the relationship. This control, perhaps intended as helpful by the respon-
dent, may be viewed less positively by others. If PAR is below average, this type of
individual may be willing to relinquish some control in the relationship for the
sake of maintaining the attachment; however, elevations on PAR suggest that such
flexibility is unlikely.

DOM very high, WRM average

This configuration suggests an interpersonal style best characterized as being
domineering and overcontrolling. Such people have strong needs to control oth-
ers, and they expect respect and admiration in return. They may be driven to
appear competent and authoritative and are likely to have little tolerance for those
who disagree with their plans and desires. Others probably view them as being
rather overbearing and dictatorial. Although able to express some degree of
warmth, the need to be in control in relationships probably taxes the endurance of
those close to them. Although they are interested in relationships with others, they
are probably quite uncomfortable about the prospect of appearing weak, submis-
sive, or passive in these relationships.

DOM very high, WRM low

People displaying this pattern report an interpersonal style best characterized
as being controlling and rather egocentric. They are likely to view relationships
more as an opportunity for self-enhancement, rather than as a source of enjoy-
ment. As a result, their relationships are likely to be impersonal and pragmatic.
They tend to be quite ambitious and, hence, competitive in relationships; they tend
to be skeptical of close attachments, perhaps viewing them as a sign of dependency

Zo4
 Evaluating Specific Psychological Issues

or weakness. Such people place a high premium on loyalty in others (particu

larly
if PAR is above average) and they are not likely to forgive transgressions or slights.

DOM very high, WRM very low

This interpersonal style may be characterized as being egocentric and suspi-
cious. Such people are likely to demand more from relationships than they are
willing to give, using relationships for self-enhancement. As a result, their interac-
tions with others are likely to be coldly pragmatic and, perhaps, exploitative. Such
people may be quite competitive in relationships, being skeptical of close rela-
tionships and avoiding commitment and any signs of dependency or weakness.
They will tend to remember any social slight and may have a reputation for nur-
turing a grudge.

DOM high, WRM very high

This interpersonal style is best characterized as involving strong needs for affil-
iation and positive regard from others. This may result in rather uninhibited social
behavior that may be seen by others as being attention-seeking and dramatic.
These needs for attention and affiliation can be so strong that the quality of social
interactions may be relatively unimportant as compared to their quantity. These
behaviors, perhaps intended as friendly and sociable by the respondent, might be
viewed as somewhat controlling and overbearing by others. Eager to be seen by
others as popular and socially effective, such people may have little patience with
those who do not view them this way, particularly if MAN-I is above average.

DOM high, WRM high

This pattern suggests an interpersonal style best characterized as friendly and
extraverted. Such people will typically present a cheerful and positive picture in
the presence of others. They are generally able to communicate their interests and
wants to others in an open and straightforward manner, although, if AGG-V is
below average, they may have difficulty expressing displeasure with others. They
usually prefer activities that bring them into contact with others rather than soli-
tary pursuits, and they are quick to offer help to those in need of it. Such an indi-
vidual sees himself or herself as a person who has many friends and as a person
who is comfortable and effective in most social situations.

DOM high, WRM average

This interpersonal style suggests a person who is self-assured, confident, and
dominant. Although they are not unfriendly, such people are likely to be described
by others as ambitious and having a leader-like demeanor. Although they are com-
fortable in social settings, they are not likely to mix indiscriminately, preferring to

235
PAI Interpretive Guide

interact with others in situations over which they can exercise some measure of
control.

DOM high, WRM low

This pattern represents a pragmatic and independent interpersonal style. Inter-
actions with others are likely to be practical rather than sentimental, and relation-
ships may be viewed as a means to an end, rather than as a source of satisfaction.
Such individuals are not likely to be perceived by others as being warm and
friendly, although they are not necessarily lacking in social skills and they can be
reasonably effective in social interactions. Others will probably view such individ-
uals as being shrewd, competitive, and self-confident.
DOM high, WRM very low
This interpersonal style seems best characterized as being remote and, perhaps,
somewhat egocentric. Such people are not likely to be very interested or invested
in social relationships, and they may take more from relationships than they are
willing to give emotionally. As a result, their relationships are likely to be pragmatic
and to be viewed in terms of their potential benefit, rather than as a source of
enjoyment. Others are likely to view such people as harsh and punitive. Such peo-
ple are probably skeptical of close relationships and will avoid commitment if pos-
sible. Above average scores on MAN-I accompanying this pattern heighten the
probability that such people will be impatient and demanding of others.
DOM average, WRM very high
This interpersonal style is characterized by an exceptionally strong need to be
accepted by others. This need for acceptance is likely to dominate the interactions
of such people. They may be seen by others as being too caring, trusting, and sup-
portive for their own good. They are at risk for being so committed to acceptance
that they lose all individuality or creativity. Such people attempt to avoid any con-
flict in relationships, and they are reluctant to accept any hint of hostility in them-
selves, particularly if AGG-V and MAN-I are below average. Others are likely to
take advantage of their strong need to be liked, which may be seen as an invitation
to exploit their trust.

DOM average, WRM high

Such an interpersonal style can be characterized as being warm, friendly, and
sympathetic. These individuals particularly value harmonious relationships and
derive much of their satisfaction from these relationships. Because of the premium
placed on harmony, they may be uncomfortable with interpersonal confrontation
or conflict and will tend to shun controversy. Such people are probably quick to
forgive others and will readily give others a second chance.

236
 Evaluating Specific Psychological Issues

DOM average, WRM average

This interpersonal style is best characterized as one of autonomy and balance.
Such a person probably has the capacity to adapt to a wide range of interpersonal
situations, able to both lead and follow, and able to balance practicality and senti-
ment. With both interpersonal scales scoring in the average range, their assertive-
ness, friendliness, and concern for others is typical of that for normal adults.

DOM average, WRM low

This pattern suggests a person who may be somewhat distant in personal rela-
tionships. Such people may not appear to place a high premium on close, lasting
relationships, and they may well view most social interactions without much
enthusiasm. Others may view them as reserved and possibly aloof and unsympa-
thetic. However, they may view themselves as independent, practical, and less pre-
occupied with the opinions of others than most people.
DOM average, WRM very low
The interpersonal style suggested by this pattern is one characterized as being
cold and unfeeling. Others are likely to see such individuals as being stern, imper-
sonal, and unable to either display affection or make a commitment to personal
relationships. At times, they may appear almost devoid of warmth and friendli-
ness, and they may make others uncomfortable and uneasy. There are probably
few people who consider such people to be anything more than an acquaintance.
DOM low, WRM very high
This interpersonal style is other-oriented to a degree that might be considered
naive, conforming, or gullible. Such people have a strong need to be liked by oth-
ers, and fear of rejection by others (particularly if ARD-P is above average) likely
makes it difficult to be assertive or to display any anger in relationships. A lack of
confidence in relationships (look for a low MAN-G) may make these people some-
what dependent, and they may tend to feel helpless and overwhelmed when under
pressure. Concerns about being well liked and not offending others may provide
situations where others could take advantage.
DOM low, WRM high (with BOR-N < 60T)
This interpersonal style is best characterized as being open, genuine, and con-
forming. Such people are likely to be somewhat unassuming individuals who pre-
fer to avoid the leadership role in social interactions and relationships. Although
not necessarily shy or socially avoidant, they are typically most comfortable in the
background of a social setting. Despite this rather unobtrusive stance in social
interactions, such people tend to be reasonably effective in their interactions. They
often are seen by others as warm, quiet individuals who are fairly eager to please.

237
PAI Interpretive Guide

DOM low, WRM high (with BOR-N 2 60T)

This type of individual is generally interested and rather conforming in rela-
tionships. Such a person is likely to be rather unassuming and will prefer to avoid
being the center of attention in social interactions, particularly if MAN-G is below
average. Such people are quite interested in maintaining relationships, and these
efforts have probably been a significant source of stress at various times. Given
their rather unobtrusive stance in social interactions, such people often value their
relationships more than is readily apparent to those around them. They are likely
to be seen by others as someone who is fairly eager to please but, at times, overly
sensitive in relating to others.

DOM low, WRM average (with BOR-N < 60T)

This interpersonal style is best characterized as being modest, unpretentious,
and retiring. Such people are likely to be self-conscious in social interactions, and
they are probably not skilled or comfortable in asserting themselves. Others
probably view such a person as passive, humble, and unassuming. If ARD-P is
above average, it is likely that the person is shy and somewhat anxious about social
interactions.

DOM low, WRM average (with BOR-N 2 60T)

This pattern suggests an interpersonal style that is generally unassuming and
self-effacing. Such people are likely to be self-conscious in social interactions and
they tend not to be skilled or comfortable in asserting themselves; previous efforts
may have led to conflicts that they did not handle well and would prefer to avoid
repeating. Others probably view such a person as rather passive, modest, yet fairly
sensitive to the appraisals of others.

DOM low, WRM low

This pattern suggests a person who is withdrawn and introverted. Such indi-
viduals are likely to appear to others as if they have little interest in socializing. In
fact, such people tend not to invite social interaction with others and make little
special effort to appear friendly. They may derive little enjoyment from such inter-
actions (particularly if SCZ-S is above average) and, as such, it would not be
expected that they would have an extensive social network. They are likely to be
rather passive and distant in those relationships that are maintained.

DOM low, WRM very low

This interpersonal style characterizes a person who is very uncomfortable in social
situations. Such individuals appear to have little interest or need for interacting

238
 Evaluating Specific Psychological Issues

with others (if SCZ-S is above average) and/or are very anxious when interacting
with others (if ARD-P is above average). In either case, such a person is likely to
take a passive, submissive stance when dealing with others. This lack of interest
and initiative may result in their being socially isolated, avoiding most social inter-
actions rather than run the risk of being forced to make an active engagement and
commitment to a relationship.

DOM very low, WRM very high

This interpersonal style suggests a person likely to be conforming, needy, and
gullible. Such a person is likely to have a strong fear of rejection by others and, as
a result, he or she finds it difficult to be assertive or to display any anger (particu-
larly if AGG-V is below average). Such people will tend to feel helpless and over-
whelmed under relatively mild pressure and dependently seek the assistance of
others. Marked concerns about offending others can provide many situations
where others take advantage of the respondent. An investigation of the self-esteem
indicators may reveal that the dependency needs arise from a poor self-image.

DOM very low, WRM high (with BOR-N < 60T)

This pattern represents an interpersonal style best characterized as being sub-
missive, conforming, and perhaps naive. Such people find it difficult to assert
themselves or to display any anger in relationships; this may be driven by anxiety
about potential rejection by others, particularly if ARD-P is above average. They
will tend to feel helpless and overwhelmed under relatively mild pressure and
dependently seek the assistance of others. Their concerns about offending others
may potentially provide situations where others could take advantage of their
eagerness to avoid interpersonal conflict.

DOM very low, WRM high (with BOR-N 2 60T)

This interpersonal style involves behavior that is generally submissive and con-
forming. Such people have difficulty with assertiveness or with effectively display-
ing anger in relationships; past experiences in this regard have likely led to con-
flicts that these individuals are probably very motivated to avoid. This submissive
style may be driven by anxiety about potential rejection or abandonment by oth-
ers. Such people will tend to feel helpless and overwhelmed under relatively mild
pressure and dependently seek the assistance of others, but they may feel that oth-
ers are not doing enough to meet their needs, particularly if PAR-R is above aver-
age. The motivation to maintain relationships may potentially provide situations
where these individuals feel that others are taking advantage of or exploiting them.

239
PAI Interpretive Guide

DOM very low, WRM average (with BOR-N < 60T)

This interpersonal style can be described as being self-effacing and lacking
confidence in social interactions. These characteristics make it difficult for such
people to have their needs met in personal relationships; instead, they tend to sub-
ordinate their own interests to those of others in a manner that may seem self-
punitive. This failure to assert oneself may result in mistreatment or exploitation
by others, although the average score on WRM and the lack of elevation on BOR-N
suggests that, to some extent, this strategy has been effective in maintaining impor-
tant relationships. An inspection of the self-esteem indicators may reveal that this
submissiveness could have taken a toll on the respondents self-image.

DOM very low, WRM average (with BOR-N 2 60T)

A person displaying this pattern can be characterized as self-effacing and lack-
ing confidence in social interactions. Such people tend to have difficulty having
their needs met in personal relationships and, instead, will subordinate their own
interests to those of others in a manner that may seem self-punitive. This failure to
assert oneself may result in mistreatment or exploitation by others, and the eleva-
tion on BOR-N suggests that this interpersonal strategy has not been effective in
maintaining the person’s most important relationships.

DOM very low, WRM low

A person displaying this interpersonal style is passive and somewhat uncom-
fortable in social situations. Such people tend to take a submissive, withdrawn
stance when dealing with others; they may feel little interest in or need for inter-
acting with others (particularly with SCZ-S above average), and they are unlikely
to initiate most relationships. This passivity may be accompanied by some feelings
of resentment when others request cooperation in some matter that the person
does not fully support (especially with PAR-R above average). It would be expected
that such people might attempt to avoid social interactions, rather than take cer-
tain risks that are implicit in relationships.

DOM very low, WRM very low

This is an interpersonal style characterized by marked discomfort in social sit-
uations. Such people have little interest or desire for interacting with others and,
for the most, part they take a passive, submissive stance when dealing with others.
If PAR-R is above average, this passivity probably leads to feelings of resentment
when others attempt to secure cooperation. It would be expected that they would
attempt to avoid or flee from most social interactions, rather than risk being forced
to make an active commitment to a relationship.

240
 Evaluating Specific Psychological Issues

Assessment of Perception of Environment

Personality assessment instruments have typically focused on identifying inter-
nal aspects of the individual, yet it is clear that a person’s behavior is substantially
influenced by aspects of the environment. Unfortunately, assessment of the situa-
tional environment is very difficult, as situations can vary in nearly infinite ways.
The constructs underlying the two PAI scales assessing environmental influences
were selected on the basis of the research literature studying these influences on
physical and mental health, as well as examining the factor structure of the rela-
tively few scales available for the assessment of the environment. Two aspects of
environment emerged from these investigations. One aspect involved the pre-
dictability, organization, and structure of the person’s surroundings, ranging from
fairly predictable environments to highly changeable and very stressful kinds of
environments. The second aspect involved the availability and quality of supports
in the environment.
The STR and NON scales provide an assessment of respondents’ perception of
their environment. The following sections provide a description of these two scales
individually, followed by a consideration of the scales in combination for the
assessment of the respondents’ view of their environment.

Stress (STR)
The STR scale provides an assessment of life stressors that respondents are cur-
rently experiencing or have recently experienced. Item content includes problems
in family relationships, financial hardships, difficulties related to the nature or sta-
tus of their employment, or major changes that have recently occurred or are about
to occur in their lives. The stress scale correlates moderately well with life events
checklists such as the Holmes and Rahe (1967) Schedule of Recent Events. How-
ever, unlike these checklists, the PAI items are not specific about the precise nature
of the stressors; they merely indicate the presence of many changes (i.e., day-to-day
circumstances in the person's life are not predictable). Although item content is not
specific, it appears that the majority of these changes have not been perceived as
being for the better, as correlations between STR and most indicators of depression
are quite high.
Average scores on STR (i.e., < 60T) reflect a person who describes life as sta-
ble, predictable, and uneventful. Scores between 60T and 70T are indicative of a
person who may be experiencing a moderate degree of stress as a result of diffi-
culties in some major life area. With scores above 70T, these difficulties are likely

241
PAI Interpretive Guide

to be having a significant impact on the respondent, a review of current work sit-

uation, family and close relationships, or financial status will probably reveal cir-
cumstances that are a source of worry, rumination, and unhappiness. Such indi-
viduals are at risk for development of a number of adjustment or reactive
disorders: scores on the clinical scales should be reviewed to determine the sever-
ity and nature of any such symptomatology.
STR scores that are markedly elevated (i.e., > 857) indicate that respondents
perceive themselves as surrounded by crises; nearly all major life areas are reported
to be in turmoil. They feel that they are powerless to control a series of undesirable
events that are happening to them. They see themselves as ineffectual, dependent,
and at the mercy of those around them, a situation that may lead to some bitter-
ness. Levels of stress in this range make the respondent vulnerable to many differ-
ent clinical disorders, and scores on the clinical scales should be examined to deter-
mine the precise nature of the individual’ reactions to stresses of this magnitude.

Nonsupport (NON)
The NON scale provides a measure of a perceived lack of social support, tap-
ping both the availability and quality of the respondent’ social relationships. Item
content addresses the level and nature of interactions with acquaintances,
friends, and family members. The scaling of NON is such that low scores reflect
high perceived social support, whereas elevations indicate a perception of the
social environment as unsupportive. The scale is a measure of the perception of
social support, rather than an objective measure (e.g., a count of frequency of con-
tact with family members). This is because one’s perception tends to be more
important, in terms of looking at the impact of social support as a moderator of
stress, than the actual amount of support received.
Average scores on NON (i.e., < 6OT) reflect a person who reports close, gener-
ally supportive connections with family and friends. Scores between 60T and 70T
are indicative of a person who may have few close interpersonal relationships, or
one who is perhaps dissatisfied with the nature of these relationships. With scores
above 70T, the respondent is reporting that social relationships offer little support;
family relationships may be either distant or combative, whereas friends are gen-
erally seen as unavailable or not helpful when needed.
NON scores that are markedly elevated (i.e., > 88T) indicate that respondents
perceive that they have little or no social support system to help them through sig-
nificant events in their lives. They tend to be highly critical of themselves as well

242
 Evaluating Specific Psychological Issues

as of other people, whom they perceive as uncaring and rejecting. Such

individu-
als have few emotional resources for dealing with crises and are particularly prone
to severe reactions to stress.

Configurations of STR and NON

The following sections discuss some of the implications for the assessment of
the environment related to different configurations of STR and NON. Four score
ranges are presented for STR: Low (< 44T), Average (45T to 59T), Moderate (60
to 691), and High (2 70T). Five ranges are provided for NON: Low (< 441), Aver-
age (45T to 59T), Moderate (60T to 69T), High (70T to 847), and Very high (= 857).
Interpretations for these scales are sometimes supplemented with references to other
scales that can shed further light on the nature of any problems suggested by the
environmental scales.

STR average, NON average

Such individuals report that their recent level of stress and perceived level of
social support are about average in comparison to normal adults. For individuals
in a clinical setting, the reasonably low stress environment and the intact social
support system are both favorable prognostic signs for future adjustment. How-
ever, any problems noted on the clinical scales are most likely enduring rather than
the result of situational influences.

STR low, NON average

This pattern involves a person who reports having experienced very few stress-
ful events in the recent past. The perceived level of social support is about aver-
age in comparison to normal adults. The combination of a stable and relatively
stress-free environment with the reasonably intact social support system is a favor-
able prognostic sign for future adjustment.

STR average, NON low

This pattern represents a level of stress comparable to that of normal adults,
combined with the presence of many individuals to whom the person can turn for
support when needed. This highly developed system of social supports has likely
been quite effective in buffering the demands of the environment, and this com-
bination is a favorable prognostic sign for future adjustment. However, if scores
on DOM are low, there is the possibility that the person is somewhat too reliant
on the social support system and may be sacrificing autonomy to maintain these
supports.

243
PAI Interpretive Guide

STR low, NON low

This pattern involves a report of very few stressful events in the recent past.
Furthermore, the person also describes having a large number of individuals to
whom he or she can turn for support when needed. The combination of a stable
and relatively stress-free environment with the extensive social support system is
a quite favorable prognostic sign for future adjustment. However, when accompa-
nied by difficulties manifested on the clinical scales (particularly on DEP or ANX),
this pattern may suggest a person with a strong tendency to internalize blame for
problems.

STR moderate, NON average or low

This pattern suggests a mild degree of stress as a result of difficulties in some
major life area. However, there appear to be a number of supportive relationships
that are serving as some buffer against the effects of this stress. The relatively intact
social support system is a favorable prognostic sign for future adjustment, and dif-
ficulties observed on the clinical scales (particularly DEP and ANX) are likely to be
related to these situational stressors, rather than reflecting more enduring patterns.

STR high, NON average or low

This pattern suggests notable stress and turmoil in a number of major life
areas. A review of the current employment situation, financial status, and family
and close relationships will clarify the importance of these in the overall clinical
picture, although the latter are less likely to be a major source of turmoil. Fortu-
nately, the person is reporting a number of supportive relationships that serve as a
buffer against the effects of this stress. The intact and committed social support
system is a favorable prognostic sign for future adjustment, and difficulties
observed on the clinical scales (particularly DEP and ANX) are likely to be related
to situational pressures, rather than reflecting a more enduring pattern.

STR average or low, NON moderate

This pattern represents an individual who experiences his or her level of social
support as being somewhat lower than that of the average adult. Such people may
have relatively few close relationships (look for SCZ-S scores that are above aver-
age) or they may be dissatisfied with the quality of these relationships (indicated
by above average scores on BOR-N or PAR-R, or both). However, there appears to
be relatively little stress arising from this or other major life areas.

STR average or low, NON high

This pattern reflects a person who believes that social relationships offer little
support; family relationships may be somewhat distant, and friends may not be

244
 Evaluating Specific Psychological Issues

available when needed. Despite the lack of social support, the environment is
viewed as reasonably stable and predictable, with relatively little stress arising from
this or other major life areas. Low scores on WRM may suggest that withdrawal is
used as a satisfactory means of coping with the unsuppotrtive social environment.
An accompanying elevation on PAR indicates that this is part of a more pervasive
pattern of dissatisfaction with the behavior and intentions of others.

STR average or low, NON very high

This pattern suggests individuals who believe that they have little or no social
support system to help them through difficult events in their lives. Such people see
others as rejecting and uncaring, and they believe that there is hardly anyone in
their environment to whom they can turn for help. Remarkably, despite their lack
of social support, they describe their environment as reasonably stable and pre-
dictable, with relatively little stress arising from this or other major life areas. This
unusual combination suggests that this pattern of dissatisfaction may be enduring
and a facet of their typical interactions with others, particularly if PAR or BOR-N
show elevations.

STR moderate, NON moderate

This pattern reflects a mild degree of stress as a result of difficulties in some
major life area, or areas. Some of these stressors may involve relationship issues,
because they experience their level of social support as being somewhat lower than
that of the average adult. Such people may have relatively few close relationships,
or they may be dissatisfied with the quality of these relationships. Interventions
directed at any problematic relationships (e.g., those involving family or marital
problems) may be of some use in alleviating one potential source of dissatisfaction.

STR moderate, NON high

This pattern represents an individual who reports that social relationships offer
little support; family relationships may be somewhat distant or ridden with con-
flict, and friends are not seen as available when needed. These relationship issues
are likely to be a major source of stress for the respondent. Interventions directed
at any problematic relationships (e.g., those involving family or marital problems)
may be of some use in alleviating other reported problems; however, if there are
elevations on PAR or BOR-N, this dissatisfaction with social relationships may be
more enduring and generalized than situational.

STR moderate, NON. very high

Such individuals report they have little or no social support system to help
them through difficult events in their lives. They see others as rejecting and

245
PAI Interpretive Guide

uncaring and believe that there is hardly anyone in their environment to whom
they can turn for help. These relationship issues appear to be a major source of
stress and concern. Interventions directed at these problematic relationships (e.g.,
those involving family or marital problems) may be of some use in alleviating a
major source of current stress. However, elevations on PAR or BOR-N (if present)
could suggest that this dissatisfaction with social relationships may be chronic and
related to personality problems.

STR high, NON moderate

This pattern involves a report of notable stress and turmoil in a number of
major life areas. A review of current employment situation, financial status, and
family and close relationships will clarify the importance of these in the overall
clinical picture. Some of these stressors may involve relationship issues, because
the level of social support is described as being somewhat lower than that of the
average adult. Such people may have relatively few close relationships, or they may
be dissatisfied with the quality of these relationships. Interventions directed at any
problematic relationships (e.g., those involving family or marital problems) may be
of some use in alleviating one potential source of dissatisfaction, although addi-
tional sources of stress are likely.

STR high, NON high

Individuals with this combination of elevations are likely to be experiencing
notable stress, chaos, and turmoil in a number of major life areas. A review of cur-
rent employment situation, financial status, and family and close relationships will
clarify the importance of these in the overall clinical picture. A primary source of
stress may involve relationship issues, because social relationships are described as
unsupportive; family relationships may be somewhat distant or ridden with con-
flict, and friends are not seen as available when needed. Interventions directed at
key problematic relationships (e.g., those involving family or marital problems)
may be of some use in alleviating what may be a major source of dissatisfaction.

STR high, NON very high

This pattern reflects individuals who believe that they have virtually no sup-
port system to help with the difficult life events that are besetting them. Such peo-
ple see others as rejecting and uncaring, and they believe that there is nobody they
can turn to for help. These relationship issues are part of a general life situation
that is viewed as unstable and unpredictable; a review of the current employment
situation, financial status, and family and close relationships will probably reveal
that all are areas of concern. Interventions directed at attempting to rebuild prob-
lematic relationships (e.g., those involving family or marital problems) may be of

246
 Evaluating Specific Psychological Issues

some use in alleviating a major source of current stress. However, elevations on

PAR or BOR-N (if present) suggest that this dissatisfaction with social relationships
may be chronic and related to personality problems.

;
:

a “a —_—

. -

e '

Pa

“th

4 FR Soul: “ ‘

“ater vas r
=s RrTPy ( “i
- -_"

rT We ckerit, mati

re » praviste — E he ow .
< ;
pee, ‘rps
r aac stie
Bind os “is oS

7 A _

® i4 ’+ . .}

as -9 jee oe
tess gine. dal haw Touhy
ape _ =
. Re
 CHAPTER 10
TREATMENT PLANNING
AND MONITORING

‘Treatment planning is a critical issue for psychological assessment, yet it is a

challenging one, because there is little empirical evidence to definitively support
specific treatments for specific problems or patient types. However, the PAI has
particular promise for refining treatment-related decision-making (Morey &
Henry, 1994), as it provides important information relevant to the treatment
process: choice of setting, need for medications, suitability for psychotherapy,
selection of therapeutic targets, and assessment of change. This chapter offers
guidelines to help the clinician use PAI data to make many commonly faced treat-
ment-related decisions. Because of the subscale structure of the PAI and its articu-
lation with current diagnostic nomenclature, the PAI is also useful in answering
many common referral questions in the context of psychological testing.
This chapter is organized into three sections. First, a description of the RXR
scale is provided, as this scale represents a variable critical at the entry point of
treatment. Then, specific issues related to determining prognosis, identifying treat-
ment obstacles, and considering differential treatment strategies are discussed. The
final section addresses issues involved in using the PAI in the assessment of
change.

Treatment Rejection (RXR)

For many years, it has been presumed that one of the most important deter-
minants of treatment outcome is the person’s motivation for treatment. Although
different authors have somewhat differing views on the nature of this motivation,
it is generally agreed that a dissatisfaction with current behavior patterns and a
willingness to make an effort to change these patterns are important components
of treatment motivation (Sifneos, 1987; Strupp & Binder, 1984). These compo-
nents can serve as important determinants of treatment outcome, no matter what
specific type of treatment is involved. Sifneos identified seven criteria for the eval-
uation of treatment motivation for his studies of short-term psychotherapy:

249
PAI Interpretive Guide

1. A willingness to participate actively in the diagnostic evaluation.

2. Honesty in reporting about oneself and one’s difficulties.
3. Ability to recognize that the symptoms experienced are psychological in
nature.
Introspectiveness and curiosity about one’s own behavior and motives.
Openness to new ideas, with a willingness to consider different attitudes.
Realistic expectations for the results of treatment.
SS Willingness to make a reasonable sacrifice in order to achieve a successful
Si
outcome.

RXR items were written to indicate attitudes that were not consistent with these
characteristics of treatment motivation. In other words, they were designed to
identify individuals who would not be motivated for treatment, but rather would
be at risk for noncompliance and early termination. Items were written to be
applicable across different therapeutic modalities. Broad content areas that were
sampled included (a) a refusal to acknowledge problems, (b) a lack of introspec-
tiveness, (c) an unwillingness to participate actively in treatment, and (d) an
unwillingness to accept responsibility for change in one’s life.
In interpreting scores on RXR, it must be remembered that T scores are refer-
enced against a community sample, not a treatment sample; hence, scores that are
typical of normal respondents actually represent little motivation for treatment.
Thus, even T scores that appear to be within the average range can have quite neg-
ative implications for treatment motivation when working within a clinical setting.
If working in other, nonclinical settings (e.g., a preemployment screening) scores
of 50T may be typical, but they are not typical when working with clinical popu-
lations. In the clinical standardization sample, the mean score on RXR was 40T.
Another aspect of RXR that is critical in its interpretation is that it is related to
treatment motivation, not prognosis. Motivation is a perhaps necessary, but cer-
tainly not sufficient, condition for successful treatment. Merely because a person
recognizes the need to make changes does not mean that accomplishing those
changes will be easy. In fact, very low scores on RXR are often somewhat of a “cry
for help,” indicative of overwhelming distress and beseeching mental health pro-
fessionals to do something to alleviate the individual's suffering. For example, indi-
viduals with Borderline Personality Disorder who are in acute distress will often
score quite low on this scale, presumably indicating very high motivation for treat-
ment. And, in fact, such patients are experiencing so much turmoil that they truly
do desperately want their lives to change. However, because such patients are

250
 Treatment Planning and Monitoring

extremely difficult to work with for other reasons, the prognosis for treatment is
not necessarily favorable.
The scaling of RXR is such that low scores reflect high motivation for treat-
ment, whereas elevations indicate little motivation for treatment. Low scores
on
RXR (.e., < 431) suggest a person who acknowledges major difficulties in his or
her functioning and who perceives an acute need for help in dealing with these
problems; scores below 20T indicate a desperate quality to these needs. Average
scores on RXR (i.e., 43T to 537) reflect a person who acknowledges the need to
make some changes, has a positive attitude toward the possibility of personal
change, and accepts the importance of personal responsibility. However, scores in
the upper portion of this range are higher than expected in respondents where
available information (e.g., from the history or from other PAI scales) suggests
some impairment; in such circumstances, the possibility of defensiveness, rigidity,
or lack of insight must be considered. Scores between 53T and 63T are indicative
of people who are generally satisfied with themselves as they are and see little need
for major changes in their behavior. Individuals scoring in this range would gen-
erally have little motivation to enter into psychotherapy and might be at risk for
early termination if they did enter treatment. RXR scores above 63T reflect indi-
viduals who admit to few difficulties and who have no desire to change the status
quo. Such individuals are not likely to seek therapy on their own initiative and will
likely be resistant if they do begin treatment; they will probably dispute the value
of therapy and have little, if any, involvement in any therapeutic attempts.

Predicting Treatment Process:

Impediments and Assets
Although motivation for treatment is an important factor in determining treat-
ment outcome, it is certainly not sufficient by itself to insure that the treatment will
be successful. There are countless patient, treatment, and interaction variables that
can potentially affect treatment outcome (Beutler, 1991). Patient predisposing vari-
ables, in isolation, will have a limited ability to predict outcome, because different
types of patients can and do respond differently to diverse forms of treatment
(Frances, Clarkin, & Perry, 1984). Some of these interactions and their implica-
tions for PAI interpretation will be discussed in a later section. Nonetheless, there
are a number of patient features that suggest a difficult treatment process, regard-
less of the type of treatment offered.
For example, a number of theorists have offered suggestions about factors
influencing amenability to various types of therapeutic approaches. Table 10-1

251
PAI Interpretive Guide

Table 10-1
Indicators of Suitability for Exploratory Therapy
Characteristic Low suitability High suitability

1. Friendliness Hostile Amiable

2. Likability Unlikable Likable
3. Intelligence Low High
4. Motivation Indifferent Motivated
5. Psychological-minded Low High
6. Conscience factors Deceittul Moral sense
7. Self-discipline Chaotic Disciplined
8. Impulse control Impulsive Self-control
9. Defensive style Alloplastic Autoplastic
10. Internalization Projecting Admits fault
11. Empathy Entitlement Empathy
12. Parental factors Abusive/Indifferent Supportive
13. Social supports Few Many

presents a list of variables offered as predictors of suitability for exploratory ther-

apy (Stone, 1985; Strupp & Binder, 1984; Waldinger @ Gunderson, 1987). How-
ever, a close examination of these features reveals that patients with numerous
indicators of “low suitability” for exploratory therapy probably are less likely to
respond to any form of intervention than patients who would be considered of
“high suitability” according to this table. For example, deceitful, impulsive, hostile
patients from an unsupportive and abusive environment are less than ideal candi-
dates for any treatment; they are unlikely to comply with pharmacotherapy, behav-
ior therapy, or group therapy, as well as exploratory therapy. Thus, this list of indi-
cators is a reasonable starting point for estimating the degree of difficulty likely
to be encountered as part of the treatment process. With the exception of “Intel-
ligence,” each of these indicators can be assessed using PAI profile information.
The following sections describe the assessment of these indicators of treatment
difficulty.
Friendliness. Individuals who are reasonably effective interpersonally are better
able to make use of any form of helping relationship, regardless of the techniques
used to achieve change. Individuals who are hostile are unlikely to cooperate with
treatment; the process of treatment is constantly at risk for deteriorating into a
struggle for control. For any individual to be considered amiable, some degree of
warmth is essential. Hence, extremely low scores on WRM (i.e., < 30T) would be
a negative indicator of friendliness. Similarly, overt indicators of hostility are also

Lode
 Treatment Planning and Monitoring

negative signs, and are probably most directly gauged by PAR-R or AGG-A eleva-
tions above 7OT.
Likability. Although friendliness and likability are likely to be empirically related,
they are separate constructs. Some people can be friendly in an overbearing or
ingenuine way and, hence, are not well liked: others can be rather hostile, but
(perhaps because, forexample, their hostility is expressed in a humorous way) still
can be reasonably likable. In general, individuals with personality disorders (par-
ticularly those in “Cluster B”) are the least likable of individuals presenting for
treatment, they tend to be manipulative, disagreeable, and egocentric. Thus, scores
on BOR and ANT, which tap the features of two of these disorders, are probably
the best indicators of likability on the test; individuals scoring above 70T on either
of these scales are typically not likely to be well liked by many other people.
Intelligence. The PAI is not an intelligence test, and, for this reason, intelligence
is the only item in Table 10-1 that cannot be estimated from the PAI profile. How-
ever, if the individual has at least the requisite fourth-grade education and cogni-
tive and/or intellectual impairment appears to be interfering with the valid com-
pletion of the PAI, this is likely to be a negative indicator for a smooth treatment
process.
Motivation. As discussed previously, motivation for treatment is perhaps a nec-
essary, although not sufficient, condition for successful interventions. The RXR
scale was constructed to yield information relevant to this construct, and scores
greater than 60T are a sign of very low motivation for treatment. However, elevated
scores on PIM can also indicate a level of rigidity and defensiveness that suggests
that motivation for personal change will be lacking; scores above 60T on this scale
should also be considered an indicator of inadequate interest in treatment.
Psychological-minded. For most forms of psychological therapy, the patient must
be willing to consider the psychological origin of problems, if only to allow the
individual to participate willingly in such treatments. Even in pharmacotherapy,
some capacity to self-monitor is necessary to enable the person to comply with the
medication regimen. Several PAI scales are suggestive of difficulties with intro-
spection and self-awareness. Marked impulsivity and acting-out tendencies are
negative indicators of introspection; thus, scores above 70T on BOR-S or ANT-A
suggest little capacity for reflection. If SOM exceeds 70T, a patient’s underlying
conflicts are prone to be expressed somatically, and such individuals may be
resistant to considering themselves in need of psychological intervention. If
ANT-E is above 7OT the patient may not have sufficient empathic capacity to
consider others’ experiences or viewpoints. Any of these features suggests lim-
ited psychological-mindedness.

253
PAI Interpretive Guide

Conscience factors. In general, a clearly established system of values and a good

moral sense are assets that are favorable prognostic features for therapy. In con-
trast, deceitful, vengeful, or antisocial types of individuals are likely to have con-
siderable difficulties working within a therapy relationship. Scores on ANT-E that
exceed 7OT indicate a willingness to deceive others for personal gain, a character-
istic that portends an arduous treatment process.
Self-discipline. Individuals with the capacity for order and discipline tend to
have smoother courses of treatment than those who have little discipline, who act-
out behaviorally, and who lead chaotic and uncontrolled lives. These problems
may lie in the realm of substance abuse (ALC or DRG above 70T), behavioral indis-
cretions (BOR or ANT above 7OT), or in a chaotic approach to life (NIM above
(OT):
Impulse control. Most psychosocial treatments require some capacity for reflec-
tion and delay. Individuals who act-out, rather than reflect on, their emotional
experience tend to have more difficulty with treatment in general. Impulsivity can
lead to compliance problems, even with treatments in which insight and intro-
spection are minimally important. On the PAI, elevations on BOR-S, ANTA,
ANT-S, or AGG are signs of poor capacity for delay and heightened impulsivity,
both of which make treatment difficult.
Defensive style. Stone (1985) uses Alexander's terms of alloplastic as opposed to
autoplastic defensive styles to refer to the nature of the patient's approach to his or
her symptoms and problems. This concept refers to whether the core problems
experienced by the person are central to the self-structure and part of the ingrained
personality (autoplastic), as opposed to problems that are viewed as ego-alien and
seen as a change from the person’s normal functioning (alloplastic). Individuals
with an autoplastic defensive style are often unable to identify the aspects of their
lives that cause them repeated difficulties, because these aspects are, in their own
minds, simply “the way they are,” rather than a disorder that they have. The char-
acterological aspects of personality represented by BOR and ANT represent the
essence of this defensive style. Similar defensive strategies are often found among
substance abusers as well, leading to concerns when ALC or DRG are elevated.
Internalization. In many clients, the internalization of blame and fault is often
both excessive and a source of distress. However, it is generally considered to be a
favorable prognostic sign within the context of psychotherapy. Individuals who
externalize blame for all their troubles, projecting responsibility outwards rather
than accepting some role in their problems, often are unwilling to make the per-
sonal changes needed in therapy. The pattern of externalization is likely to repeat in
the context of therapy, with the patient eventually coming to blame the therapist for

ZO*
 Treatment Planning and Monitoring

treatment impasses due to the clinician's unwillingness to accept the patient’s

world view. Such individuals often do not place sufficient trust in others to estab-
lish a helping relationship; eventually, they have difficulty with the treating pro-
fessional as an authority figure, and they may react to the therapist in a hostile or
derogating manner. Scores above 7OT on PAR are generally a sign that marked
externalization is part of the clinical picture.
Empathy. The establishment of an alliance with the treating professional is a
critical ingredient in therapeutic success, regardless of treatment modality, and the
ability to care about and establish rapport with others is central in forging this
alliance. Individuals who approach relationships with an entitled, exploitative, and
contemptuous attitude tend to have difficulty working within the therapy context.
Elevated scores on MAN-G or ANT-E are particularly related to problems in the
empathic realm. If DOM is greater than 7OT, the patient’s need for control over the
therapist may also make collaboration difficult.
Parental factors. Individuals who come from a background where caretakers
have been abusive, indifferent, or exploitative tend to have great difficulty placing
trust in helping professionals. In particular, they will become resistant and may
terminate treatment as issues become increasingly sensitive. Elevations on ARD-T
or NON, or both, can serve as cues to difficulties in this area.

Social supports. Research has shown that patients who have an adequate social
support network tend to make better and more rapid progress in psychotherapy.
NON scores below 7OT indicate that a patient’s perceived social supports are gen-
erally within normal limits, and STR scores in that range suggest that the support
system is reasonably stable and predictable. An adequate and predictable support
system is considered a favorable sign, whereas elevations on NON or STR, or both,
reflect problem areas that can serve as both an obstacle and a target for treatment.

The PAI Treatment Process Index (TPD

Table 10-2 presents the operationalization of these predictors of treatment
amenability into a cumulative index known as the Treatment Process Index (TPI).
Each feature of the TPI is considered present if any of its indicators (listed in the
second column of Table 10-2) are present. The features on the TPI tap a wide array
of different psychological problems, and, in general, respondents with globally ele-
vated profiles will obtain high scores. However, certain PAI scales appear repeat-
edly in the calculation of the TPI, and in general the greater the degree of charac-
terological problems, the higher the predicted degree of disruptions in treatment
process.

255
PAI Interpretive Guide

Table 10-2
Operationalization of the Items of the
PAI Treatment Process Index (TPI)
PAI Frequency Frequency
problem in community in clinical
Characteristic indicators sample® sample?

1. Friendliness PAR-R > 70T 7% 27%

AGG-A > 70T
WRM < 30T
2. Likability BOR > 70T 7% 36%
ANT > 70T
3. Motivation RXR > 60T 23% 12%
PIM > 60T
4. Psychological-minded BOR-S > 70T 13% 46%
ANT-E > 70T
SOM > 70T
ANT-A > 70T
5. Conscience factors ANT-E > 70T 5% 9%
6. Self-discipline BOR > 70T 12% 55%
ANT > 70T
ALC > 70T
DRG an
NIM > 70T
7. Impulse control BOR-S > 70T 10% 37%
AGG > 70T
ANT-A > 70T
ANT-S > 70T
8. Defensive style BOR > 70T 10% 52%
ANT > 70T
ALC > 70T
DRG > 70T
9. Internalization PAR > 70T 3% 18%
10. Empathy MAN-G > 70T
DOM > 70T
ANT-E > 70T 8% 13%
11. Parental factors ARD-T > 70T T% 40%
NON > 70T
12. Social supports NON > 70T 1% 41%
STR > 70T
Total score M Vode 3.86
Total score SD 1.90 3.22

aN = 1,000. °N = 1,246.

The PAI Treatment Process Index (TPI) is scored by counting the number of
positive features in Table 10-2. As the table demonstrates, each feature in isolation
is seen with reasonable frequency in a general clinical population. Table 10-3

256
 Treatment
Planning and Monitoring

Table 10-3
T-Score Equivalents for the PAI Treatment Process Index (TP)
Standardized Against Community and Clinical Normative Samples

T-score equivalent, T-score equivalent,

TPI score community norms? clinical norms?
0 44 38
1 49 41
2 55 44
3 60 47
4 65 50
5 70 54
6 76 57
7 81 60
8 86 63
9 91 66
10 97 69
11 102 72
42 107 7S
aN = 1,000. "N= 1,246.

presents T-score conversions for the Treatment Process Index (TPI), standard-
ized against the means for the community and clinical samples. Scores on the
TPI will be elevated in individuals who have refractory problems that will tend to
complicate treatment process, regardless of the specific modality used. TPI scores
below 4 indicate the presence of numerous personal assets that may assist the
treatment process. If presenting for treatment, such people may be experiencing
transient distress, perhaps associated with current circumstances, rather than
chronic difficulties. As the TPI begins to elevate (i.e., 7-10 items positive), there
are many and varied obstacles to a smooth treatment process. Problems tend to be
more refractory and chronic in nature, and therapy will likely be difficult and have
many reversals. Marked elevations (i.e., 11 or 12 items positive) suggest a very dif-
ficult treatment process. Because of the complexity of these problems and their
enduring nature, considerable efforts will be needed to establish any form of
alliance needed to maintain the person in treatment. Such individuals are likely to
be among the most challenging of any patients to treat.
Numerous correlates for the PAI Treatment Process Index (TPI) are listed in
the Appendix, of which selected results are presented in Table 10-4. This table
reveals that the TPI is positively correlated with various indicators of character
pathology and of an alienated, hostile detachment and withdrawal from others. It

257
PAI Interpretive Guide

Table 10-4
Selected Correlates of the
PAI Treatment Process Index (TPI) Total Score
Neen ee eee ee ea

Correlation with
Variable description TPI score

PAI BOR aT,

PAI ANT .68
PAI PAR .66
PAI RXR —.46
MMPI Sc .56
Mississippi PTSD scale A
Diagnostic Interview for Personality Disorder—Antisocial personality 46
Diagnostic Interview for Personality Disorder—Schizotypal personality .40
Hare Self-report Psychopathy .67

Note. MMPI = Minnesota Multiphasic Personality Inventory; PTSD = Posttraumatic Stress Disorder.

is important to point out that the TPI displays a moderate negative correlation with
RXR. This serves to underscore the observation that low RXR scores should NOT
be considered to be a predictor of a smooth treatment process. Although it is true
that individuals with high scores on RXR would be expected to refuse or reject
treatment, many individuals with very low RXR scores tend to have problems of
the sort reflected on the TPI, which can be very disruptive to treatment for differ-
ent reasons. The following sections describe how these two attributes in combina-
tion can be helpful in evaluating prognosis and in anticipating obstacles that may
arise in treatment.

The Interaction of Treatment Motivation and Difficulty

As mentioned earlier, the motivation of an individual for psychological treat-
ment and his or her suitability for treatment are somewhat independent issues
that are likely to interact to influence outcome. The following sections provide
interpretations on various combinations of these factors, using RXR scores to gauge
treatment motivation and the Treatment Process Index to assess treatment suit-
ability. Motivation is divided according to four levels of RXR: Scores above 62T are
markedly elevated, indicating likely treatment rejection; scores between 53T and
62T are moderately elevated, indicating probable treatment resistance; scores
between 43T and 527 are in the high average range for clinical groups, raising the
question of some resistance to treatment; and scores below 43T indicate accep-
tance of a need for treatment. With respect to treatment difficulty, Treatment Process
Index scores above 8 indicate many and varied obstacles to a smooth treatment

258
 Treatment Planning and Monitoring

process, with a degree of difficulty placing them among the top 10% of clinical
patients; scores between 4 and 8 are typical for clinical groups; and index scores
below 4 indicate the presence of numerous personal assets that may assist the treat-
ment process. The following sections make reference to these ranges of scores.

Treatment Process Index high, RXR markedly elevated

This pattern suggests a level of interest and motivation for treatment that is
below average in comparison to adults who are not being seen in a therapeutic set-
ting, and the treatment motivation is substantially lower than is typical of individ-
uals being seen in treatment settings. Such a pattern suggests that the individual is
not willing to take responsibility for the many difficulties in his or her life, and
there appears to be no desire to make personal changes, despite the fact that sev-
eral life areas do not seem to be going well at this time. Such people are very
unlikely to seek therapy on their own initiative, and they are likely to be highly
resistant if they do begin treatment. If treatment were to begin, the combination of
problems that is being described suggests that treatment would be an uphill strug-
gle and that the treatment process is likely to be arduous, with numerous rever-
sals. Given the level of resistance to treatment, the respondent is not likely to
weather many of these reversals before attempting to terminate treatment.

Treatment Process Index high, RXR moderately elevated

This level of interest and motivation for treatment is somewhat below average
in comparison to adults who are not being seen in a therapeutic setting. Further-
more, the treatment motivation is substantially lower than is typical of individuals
being seen in treatment settings. Such responses suggest persons who are satis-
fied with themselves as they are and who see little need for changes in their
behavior, despite the recognition that several life areas are not going well at this
time. The combination of problems being reported suggests that treatment would
be quite challenging and that the treatment process is likely to be marked by sig-
nificant turmoil.

Treatment Process Index high, RXR high average

The respondents interest in and motivation for treatment is comparable to that
of adults who are not being seen in a therapeutic setting, although they are some-
what lower than is typical of individuals being seen in treatment settings. Despite
the recognition that several life areas are not going well at this time, there may be
resistance to the idea that personal changes are needed; problems may be blamed
on external circumstances, rather than recognized as personal issues in need of
improvement. The combination of problems that are reported suggests a difficult
treatment process; after the precipitating crisis resolves, it is likely that greater

Zee)
PAI Interpretive Guide

resistance to treatment will emerge and the client may then be at risk for prema-
ture termination.

Treatment Process Index high, RXR below average

This pattern suggests an individual who appears to have substantial interest in
making changes in life, with satisfactory motivation for treatment. Such responses
indicate an acknowledgement of important problems, a perception of a need for
help in dealing with these problems, and a positive attitude toward personal
responsibility in pursuing treatment. Despite these favorable signs, the combina-
tion of problems that are reported suggests that treatment is likely to be difficult,
with numerous character and environmental obstacles to a smooth treatment
process. The current level of motivation will be a valuable asset in working
through the many reversals in treatment that are likely to occur.

Treatment Process Index average, RXR markedly elevated

This level of interest and motivation for treatment is below average in com-
parison to adults who are not being seen in a therapeutic setting and markedly
lower than is typical of individuals being seen in treatment settings. These responses
are typical of individuals who are quite satisfied with themselves as they are at pre-
sent, and who see little need for changes in their behavior. They are acknowledg-
ing that a number of life areas are not going well at this time, although the blame
for these problems appears to be placed externally. Treatment is not likely to be
sought voluntarily. The nature of some of these problems suggests that treatment
would be fairly challenging even if a commitment to treatment were made, and
that the treatment process is likely to be difficult. Setbacks in treatment are likely
and should be anticipated, with the respondent likely to attempt to terminate
treatment during such times.

Treatment Process Index average, RXR moderately elevated

This level of treatment motivation, which is somewhat below average in com-
parison to adults who are not being seen in a therapeutic setting, is substantially
lower than is typical of individuals being seen in treatment settings. Such people
are generally self-satisfied and see no need for major changes in their approach to
life, even though they acknowledge that certain life areas are not going well at this
time. Responsibility for most of these problems appear to be attributed externally,
and much of the initial treatment phases may need to focus on what role the
respondent has played in these difficulties. The nature of the clinical problems pre-
sented suggests that treatment would be fairly challenging, with the treatment
process likely to be hampered by the respondent's ambivalent commitment to
treatment.

260
 Treatment Planning and Monitoring

Treatment Process Index average, RXR high average

This level of interest and motivation for treatment is comparable to that of
community adults, although somewhat lower than is typical of individuals being
seen in treatment settings. Despite a recognition that a number of life areas are not
going well at this time, the pattern suggests possible resistance to the idea that per-
sonal changes are needed to address these problems. Individuals presenting such
a pattern may accept treatment in an effort to cope with an immediate crisis, but
their commitment to treatment may wane as the crisis resolves, and there may be
resistance to dealing with longer-term issues. The nature of some of the presenting
problems suggests that the treatment process may be difficult, and reversals should
be expected.

Treatment Process Index average, RXR below average

The respondents interest in and motivation for treatment is comparable to or
stronger than that of most individuals being seen in treatment settings. There is an
acknowledgement of important problems and a perception of a need for help in
dealing with these problems. Such people report a positive attitude toward the
possibility of personal change, the value of therapy, and the importance of personal
responsibility. However, the nature of some of the presenting problems suggests
that treatment will involve some challenges, with a number of character issues that
must be addressed if long-term changes are to be realized. The current level of
motivation should be an important asset in dealing with the setbacks in treatment
that are likely to occur as these issues rise to the surface.

Treatment Process Index low, RXR markedly elevated

The respondent’ interest in and motivation for treatment is below average in
comparison to adults who are not being seen in a therapeutic setting and a great
deal lower than is typical of individuals being seen in treatment settings. Such peo-
ple are adamant about being satisfied with themselves as they are; they are not
experiencing significant distress, and, as a result, they see little need for changes
in their behavior. The respondent does report a number of strengths that augur
well for a relatively smooth treatment process if he or she were willing to make a
commitment to treatment, but it does not appear likely that such a commitment
would be made voluntarily at this time.

Treatment Process Index low, RXR moderately elevated

This level of treatment motivation is substantially lower than is typical of indi-
viduals being seen in treatment settings and is even below that of adults who are
not being seen in a therapeutic setting. These responses suggest a person who is
satisfied with his or her approach to life and not interested in making major

261
PAI Interpretive Guide

changes in this approach. Because such people are not experiencing marked dis-
tress, they see little need for such changes at this time. However, this pattern does
indicate a number of strengths that would predict a relatively smooth treatment
process if the person made a commitment to treatment at some future point.

Treatment Process Index low, RXR high average

The respondent’ interest in and motivation for treatment is comparable to that
of adults who are not being seen in a therapeutic setting, although somewhat lower
than is typical of individuals being seen in treatment settings. Because there does
not appear to be marked distress at this time, the person does not report needing
to make major changes in his or her behavior or approach to life. However, the
individual seems responsive to the importance of personal responsibility and self-
improvement and reports a number of strengths that are positive indications for a
relatively smooth treatment process, should he or she decide that some form of
psychological treatment might be needed.

Treatment Process Index low, RXR below average

This level of treatment motivation is comparable to or better than the majority
of individuals seen in treatment settings. These responses suggest an acknowledg-
ment of important problems and a perception of the need for help in dealing with
these problems. The respondent reports a positive attitude toward the possibility
of personal change, the value of therapy, and the importance of personal responsi-
bility. The respondent seems interested and willing to engage in some introspec-
tion in order to bring about self-improvement; because there are relatively few
clinical problems being reported, this desire appears to be independent of any
necessities brought about by immediate crisis. In addition, he or she reports a
number of other strengths that are positive indications for a relatively smooth
treatment process and a reasonably good prognosis.

Differential Treatment Planning

Treatment selection is a difficult task in the mental health field; among psy-
chosocial interventions alone, there are at least 130 different approaches from
which to select (Smith, Glass, & Miller, 1980). There is also frustratingly little evi-
dence to suggest that a specific treatment is unequivocally indicated for a particu-
lar disorder. Unfortunately, the realities of clinical practice dictate that many criti-
cal treatment selection decisions must be made despite the limited information
that can be brought to bear upon these questions. Obviously, making treatment
recommendations based upon the PAI results is hampered by this limited database,

262
 Treatment Planning and Monitoring

but some conclusions can be drawn: this section offers some suggestions for this
purpose.

Perry, Frances, and Clarkin (1988) have divided mental health treatments
according to five intervention parameters:
I. Setting, such as inpatient hospitalization, outpatient therapy, or halfway
house placements.
2. Format, referring to whether treatment should involve individual sessions,
group therapy, and/or family or marital therapy.
3. Time, involving the length and frequency of sessions, and the total duration
of treatment.
4. Approach, involving the use of different techniques based upon different
theoretical perspectives.

5. Somatic, involving the use of psychopharmacologic medications or other

somatic forms of treatment.
The following sections are organized according to these five parameters of
treatment and the resulting treatment decisions that the clinician often faces
related to these parameters. Each common question for which the PAI may pro-
vide guidance is followed by a list of topics or areas that are important to assess in
answering the question. Each area is, in turn, followed by the specific sources of
PAI data most relevant to that area. It should be stressed that these suggestions are
to be treated as guidelines to aid in the clinical decision-making process and are
not offered as firm rules.

Choice of Treatment Setting

One frequent function of psychological assessment involves determining
whether inpatient treatment is required and, if the patient is already in an inpatient
setting, to provide recommendations about the continued necessity of such treat-
ment. The following areas should be considered:

Functional Impairment
Is the patient’s current level of overall functioning or ability to meet role
responsibilities impaired to such an extent that hospitalization is warranted? Such
problems can be manifest in a number of areas tapped by the PAI, particularly
with extreme scores on the clinical scales that are at or above the profile “skyline”
in the absence of any indication of negative distortion of the profile due to malin-
gering or exaggeration. Chronic and severe somatic complaints and accompanying

263
PAI Interpretive Guide

dysfunction or fatigue can compromise functional capacity, and extreme scores on

SOM can reflect such issues. Anxiety may be so overwhelming that the patient may
be unable to meet daily tasks, and mild stressors might precipitate a major crisis
in such an individual; ANX scores above the skyline would be expected in such
respondents. Extreme scores on DEP are usually accompanied by a crippling level
of fatigue, loss of motivation, social withdrawal, and helplessness that may make
outpatient treatment unfeasible. Individuals with extreme MAN scores may display
a level of impulsivity, inability to delay gratification, and flight of ideas that can
render them unable to meet role expectations. With extreme PAR scores, particu-
larly elevations on PAR-P, the possibility of paranoid delusions that interfere with
social and occupational functioning should be explored. Similarly, extreme scores
on SCZ are typically associated with an active schizophrenic episode requiring hos-
pitalization, and even the more moderate elevations on the SCZ-P subscale should
be investigated, as this subscale measures psychotic signs unique to schizophrenia.

Potential for Self-Harm

Is the patient an imminent risk to himself or herself due to suicidality or
impulsive self-damaging behaviors? Obviously, suicidality is a critical indication of
the need for inpatient treatment, and this issue is discussed in detail in chapter 7.
The SUI scale is an important tool for such assessments; individuals on suicidal
precautions display an average score of 84T on that scale. Marked elevations on
SUI are particularly worrisome when accompanied by the risk factors represented
by the Suicide Potential Index. However, impaired judgment and recklessness can
place an individual at risk for self-harm in the absence of overt suicidal ideation.
Scores above 75T on MAN represent a degree of behavioral impulsivity that may
increase the risk of self-damaging behaviors. Elevations above 7OT on either BOR-S
or ANT-S represent long-standing characterological features that do not necessar-
ily indicate suicidality, but do suggest impulsivity that heightens the risk for self-
harm, particularly when combined with other clinical indicators. The BOR-S ele-
vation suggests a pattern of impulsive behavior with high potential for negative
consequences (e.g., reckless spending, sexual behavior or substance abuse). An
ANT-S elevation indicates a tendency toward reckless and dangerous behavior and
a craving for excitement and stimulation.

Danger to Others
Does the patient require hospitalization because he or she is an immediate dan-
ger to others? Obviously, assaultive behavior indicates a need for inpatient treat-
ment; this issue is discussed in detail in chapter 8.

264
 Treatment Planning and Monitoring

Chemical Dependency
The choice between an inpatient and an outpatient setting for the treatment of
chemical dependency is an increasingly common and important decision. Often
this decision is based on whether or not the patient has the ability to control sub-
stance use On an outpatient basis or can be detoxified safely as an outpatient. If
ALC is greater than 84T or DRG is greater than 80T, then the patient is increas-
ingly likely to qualify for a diagnosis of substance dependence and may require
detoxification in an inpatient setting, particularly if there are emotional complica-
tions such as suicidality or danger to others. It should be remembered that the PAI
drug and alcohol scales are straightforward measures of what the patient reports;
various PAI indicators (as described in chapters 2 and 4) should be checked for
evidence of denial.

Traumatic Stress Reaction

Evidence on ARD-T of extreme preoccupation with past traumatic events when
accompanied by high levels of anxiety (i.e., ANX > 90T) may indicate the need for
crisis hospitalization. In cases where no obvious stressors are known, this pattern
has sometimes been observed to indicate the imminent emergence of suppressed
memories of childhood abuse. On occasion, the ARD-T subscale may be elevated,
even in cases where the patient cannot currently report specific traumatic memo-
ries. In extreme cases, the patient may be in temporary need of a protected envi-
ronment. This is particularly true if there is evidence of recent passively self-
damaging behaviors, such as car accidents. Signs of thought disturbance also will
exacerbate such a clinical picture.

Choice of Treatment Format

Individual treatment remains the most prevalent format for mental health
treatment, and it is difficult to imagine situations in which some individual con-
tact with a patient would be contraindicated. Nonetheless, the increasing acknowl-
edgment of interpersonal factors in personal problems has led, in recent years, to
a growing use of group and family/marital interventions.
Group-based treatments come in many forms, ranging from self-help groups to
psychotherapy groups with heterogeneous members. These different forms share
a number of critical mechanisms that emphasize the importance of interpersonal
feedback, confrontation, and support within an environment of peers. Such
interventions are particularly effective for individuals with poor social skills, dis-
tortions in their view of others and themselves, problems with empathy, or social

265
PAI Interpretive Guide

anxiety. A number of PAI scales are global indicators of social ineffectiveness of the
type that might be amenable to group intervention, including low scores on WRM
and high scores on SCZ-S (suggesting social awkwardness) and ARD-P (potentially
indicating social anxiety). Other indicators of problems that may be helped with
group interventions include marked distrust (elevated PAR scores or any of the
PAR subscales), rigid needs for interpersonal control (high scores on DOM), or fail-
ures in empathy (ANT-E). Although these latter problems present considerable
hurdles for any form of therapy, group-based interventions may be helpful in dif-
fusing the problems with authority (in the form of resistance or hostility toward
the therapist) that such people often manifest.
Family and/or marital therapy is particularly effective in ameliorating issues
that lie primarily within a family system, and even interventions focused upon par-
ticular emotional problems may be more effective if made within a family therapy
context. On the PAI, marital and family issues are most evident on NON and, to a
lesser extent, on STR. Elevations on NON that are 10T points above any of the clin-
ical scales are particularly indicative that the respondent views the primary con-
cerns as existing within the marriage and/or the family. In interpreting the NON
elevation in this manner, the clinician should pay particular attention to elevations
on PAR and/or BOR, which may indicate a generalized pattern of interpersonal bit-
terness, of which the reported family difficulties are merely an instance.

Choice of Treatment Length

As cost containment becomes an ever-increasing consideration in health care,
efforts to predict and even to limit length of treatment have become important
concerns. Unfortunately, in the mental health field it is quite difficult to predict in
advance how long treatments should last. Length of treatment also is confounded
with treatment approach, with some treatments (e.g., certain behavioral treat-
ments) tending to be briefer, whereas others (e.g., psychoanalysis, or maintenance
medication) can last for years. Finally, over the course of treatment, both patient
and therapist will reconsider whether the frequency of sessions should change and
whether further treatment is necessary.
One rather global guide to the likely duration of treatment is the Treatment
Process Index, which will be elevated in individuals who have refractory problems
that will require treatments of greater intensity. Persons presenting for treatment
with 4 or fewer items on this index are likely to be experiencing transient distress,
perhaps associated with current circumstances. A relatively brief intervention
with such individuals can have a significant impact, relative to other patients. As
the index begins to elevate (i.e., 7-10 items positive), the refractory nature of the

266
 Treatment Planning and Monitoring

problems makes it unlikely that a brief intervention will be effective in ameliorat-

ing the issues that are probably driving the observable level of distress, and treat-
ments of greater duration and intensity may be required to effect lasting change.
Marked elevations (i.e., 11 or 12 items positive) suggest a need for highly inten-
sive treatments. Because of the complexity of the problems and their enduring
nature, brief interventions are likely to involve crisis intervention, and consider-
able efforts will be needed to establish any form of the alliance needed to maintain
the person in more intensive treatment.
In the course of clinical practice, decisions about length of treatment are usu-
ally part of the treatment process, rather than fixed at the beginning of treatment.
As improvements are noted, the intensity of treatment may be lessened, or formal
treatment may be terminated. The scale and subscale structure of the PAI make it
particularly useful for charting patient changes and for making decisions about
modifications in treatment intensity based on those changes. For example, in the
inpatient treatment of severe depression, the relative changes in the affective, cog-
nitive, and physiological components can be measured separately by readminis-
tering the test in order to better understand the specific effects of treatment; this
will facilitate decisions about the need for continued inpatient care. A reduction in
suicidal ideation may be noted, and changes in the patient’s openness to treatment
(RXR), negativity of world view (NIM), and perceived balance of external stress
(STR) versus available support (NON) may all be useful for judging the patient’s
progress and updating treatment plans as needed.
Multiple administrations of the PAI during treatment can be useful in identify-
ing critical elements of the treatment process that might indicate a need for alter-
ing the treatment intensity. For example, for clients presenting with RXR scores
suggestive of treatment rejection, it would be anticipated that initial efforts in treat-
ment might need to be directed at potential resistance. Alternatively, clients receiv-
ing an interpersonally-based treatment might be expected to show changes in the
interpersonal scales as a prerequisite to addressing any distress that would be evi-
dent from the clinical scales. Similarly, clients receiving cognitive therapy for
depression might be expected to show the most rapid improvements on DEP-C,
with improvements in somatic and affective aspects of the syndrome contingent
upon this change. If anticipated changes are not observed, revisions in treatment
intensity or treatment approach might be needed.

Choice of Differential Treatment Approach

As noted earlier, the research literature provides little evidence to support the
selection of specific therapies for specific problems. However, PAI data may be

267
PAI Interpretive Guide

Table 10-5
Selective Patient Variables for Psychodynamic Therapy

Patient selection variable PAI indicators

Chronic sense of emptiness and elevated BOR-/, suppressed MAN-G

underestimation of self-worth

Loss or long separation in childhood elevated ARD-T

Conflicts in past relationships elevated STR, ARD-T, BOR-N

(e.g., with parent or sexual partner)

Capacity for insight low RXR, normal to low BOR-S

Ability to modulate regression normal to low BOR, AGG

Access to dreams and fantasy SCZ > 45T

Little need for direction and guidance DOM > 45T

Stable environment STR < 80T

coupled with guidelines offered in the literature, as well as with common “clinical
wisdom,” to provide some general guidance for treatment planning. For example,
Karasu (1990a, 1990b) has offered a comparison of psychodynamic, cognitive,
and interpersonal approaches along a variety of theoretical and technical dimen-
sions. Using the syndrome of depression as an example, Karasu delimits patient
variables that would either call for or contraindicate each of these psychothera-
peutic approaches. Although the model is presented in the context of depression,
the concepts are equally applicable to many other clinical problems. Tables 10-5,
10-6, and 10-7 present Karasu’s selective patient variables for the psychodynamic,
cognitive, and interpersonal strategies, respectively, and also list various PAI mark-
ers for these selective variables.
The psychodynamic or exploratory approach focuses on insight, understand-
ing, and resolution of internal conflict, taking a developmental approach toward
understanding the individual’ present difficulties. This approach is particularly
suited for individuals with difficulties that are developmental in nature; hence, the
issue of conflicts in past relationships is especially salient. However, use of this
approach requires the individual to be reasonably psychologically minded, have
the capacity for trust, and be able to handle the anxiety resulting from a con-
frontation of his or her defenses. Karasu (1990b) suggests that individuals with
more focused interpersonal problems or social deficits, particularly those pertain-
ing to present-day relationships, might be better treated with an interpersonal

268
 Treatment Planning and Monitoring

Table 10-6
Selectiv
eo e Patient Variable
ed
s for Cognitive Therapy
en
Patient selection variable PAI indicators
Obvious distorted thoughts about self, world, elevated DEP-C, ANX-C
and future ‘
Pragmatic (logical) thinking ARD-O > 55T, WRM < 60T
Moderate-to-high need for direction and DOM < 45T
guidance

Responsiveness to behavioral training ARD-O > 55T

and self-help (high degree of self-control) BOR-S, ANT-A, AGG-P < 60T
ALC, DRG < 60T

approach. Finally, the cognitive approach is particularly well suited to individuals

with negative distortions of the self, because there is less need for introspection
and insight during the course of treatment.
There are a variety of other psychosocial approaches in addition to the psy-
chodynamic, cognitive, and interpersonal approaches described by Karasu (1990a,
1990b). For example, many treatments are supportive in nature, aiming to shore
up a patient's defenses and restore them to a more functional level. Such treatments
are particularly important when there is evidence that the patient is extremely
overwhelmed, has highly disorganized thought processes, or is quite vulnerable
due to traumatic stress reactions (e.g., Frances et al., 1984). Approaches utilizing
behavioral or environmental manipulation procedures may be optimal for difficul-
ties involving circumscribed phobias (look for ARD-P elevations), somatization
(SOM-S or SOM-H), assertiveness training, or lack of impulse control (see the fol-
lowing sections below). Conjoint family or marital therapy should be considered
in cases of extreme functional impairment, or when the patient reports a marked
lack of support by others, as suggested by elevated scores on NON.

Choice of Somatic Treatments

In many outpatient settings, the clinician often has to make the important deci-
sion of whether or not to refer the patient for a medication consult. In inpatient
settings, the test results can help the physician choose between medications based
on the relative prominence of depression, anxiety, mania, psychosis, or other
symptomatology that is amenable to pharmacologic treatment. For example, Karasu

269
PAI Interpretive Guide

Table 10-7
Selective Patient Variables for Interpersonal Therapy
ee

Patient selection variable PAI indicators

Recent, focused dispute with spouse or elevated NON, STR, BOR-N

significant other

Social or communication problem elevated SCZ-S, ARD-P; WRM < 40T

Recent role transition or life change elevated STR

Abnormal grief reaction elevated STR

Modest to moderate need for direction DOM 40T-50T

and guidance
Responsiveness to environmental NON < 75T
manipulation (available support network)

(1990b), in addition to the indications for different psychotherapy approaches

already described, offered a number of indications for pharmacotherapy of depres-
sion; these guidelines and related PAI markers are presented in Table 10-8.

In addition to the DEP scale and related markers of depression described in

Table 10-8, other PAI scales can suggest the possible need for somatic treatment.
A variety of scale elevations can serve as general markers for medical evaluation
and/or intervention. With respect to anti-anxiety medications, the ANX and ARD
scales are particularly informative. Marked elevations on ANX suggest intense pre-
occupation and rumination that may be intrusive enough to place the patient at
risk for inadequate occupational or social functioning and sufficient to interfere
with the progress of psychotherapeutic interventions. Also, very high STR scores
suggest that nearly all major life areas are in turmoil and that the patient feels sur-
rounded by crises. Severe scores on ARD-P can indicate multiple phobias, panic
disorder, and/or agoraphobia, disorders which may benefit from a combination of
medical and psychosocial treatment.
Various PAI markers also can indicate the need to consider antipsychotic med-
ications. Marked elevations on PAR (particularly PAR-P) indicate a need to evaluate
for systematic paranoid delusional systems that may benefit from antipsychotic
medication. If the full SCZ scale is markedly elevated, or even if the SCZ-P subscale
displays a more modest elevation, the patient may require neuroleptic medication.
Noteworthy elevations on SCZ-T indicate marked confusion and concentration

270
 Treatment Planning and Monitoring

Table 10-8
Indications for Pharmacotherapy of Depression
Patient selection variable PAI indicators
Marked vegetative signs; extreme or elevated DEP-P, DEP-A, BOR-A
uncontrolled mood

Anhedonia; loss of libido; impaired sexual elevated DEP-P

function or performance

Significant weight loss, early morning awakening elevated DEP-P

Hyperactivity or motor retardation elevated DEP-P, suppressed MAN-A
Depressive stupor elevated DEP-P
Nihilistic or self-deprecatory delusions, elevated SCZ-P
self-berating auditory hallucinations

Loss of control over thinking, obsessive elevated SCZ-T, ARD-O

rumination, inability to focus or act

Acute, episodic, and uncontrolled suicidal acts markedly elevated SU/

or plans

Panic attacks or phobias; persecutory delusions; elevated ARD-P, ANX-P

pseudodementia; physical symptoms or elevated PAR-P, SOM, SOM-C
somatic delusions

Other mental disorders, such as schizophrenia, elevated SCZ, ALC

alcoholism, anorexia

problems that may benefit from medication; however, even without elevations on
other SCZ subscales, SCZ-T also may reflect severe depression. Finally, elevations
on MAN above the profile skyline raise the possibility of a full-blown manic
episode (i.e., medication should be considered).

Specifying Therapeutic Targets

The PAI also can be a useful source of data for isolating specific targets for ther-
apeutic work, regardless of approach and/or diagnosis, and may help order the pri-
orities for intervention. Morey and Henry (1994) have described a number of such
targets. The following list, derived from the Morey and Henry guidelines, is not
exhaustive, but it does cover some commonly observed areas of difficulty that
cause people to seek treatment:

EN
PAI Interpretive Guide

Poor Impulse Control

The most obvious priorities for intervention are impulsive and potentially dan-
gerous behaviors, chemical dependency, and maladaptive anger expression. Eleva-
tions on any of the following scales and subscales are associated with poor impulse
control: ALC, DRG, MAN, BOR (particularly BOR-A and BOR-S), ANT (particularly
ANT-S), and AGG. Treatment may involve medical management in the case of a
manic episode, or it may require direct limit-setting, therapeutic contracts (con-
ditions under which therapy will or will not proceed), or anger management
training. The more numerous the indicators, the greater the problem and the
poorer the prognosis. There is some research evidence (e.g., Sloane et al., 1975) to
suggest that behavioral approaches may be somewhat more effective with these
types of acting-out and antisocial problems.

Anger Repression
Some patients experience problems with overinhibition of impulses (e.g., an
inability to appropriately express angry feelings, resulting in maladaptive strategies
to contain anger). This may be due to fear of rejection, fear of loss of control, the
unacceptability of angry feelings, and so forth. Repressed anger may express itself
as timidity and lack of assertion (very low AGG), compulsive rigidity (elevated
ARD-O), or as physical symptoms (SOM elevations). Those patients with a history
of abuse (observed on ARD-T) may also have difficulty expressing anger directly,
even though there may be deep underlying anger. In such cases, encouragement
of the more direct expression of anger may be a useful first step. It should be noted
that the mere expression of anger (e.g., “cathartic” treatment) has not usually been
shown to be of lasting benefit in and of itself as the only therapeutic procedure.

Excessive Dependency
Excessive dependency may be a problem for a number of reasons. Patients may
be unable to leave abusive relationships, may sacrifice their own needs for those of
others, or may be so eager to please and fearful of rejection that they are exploited.
Above average emphasis on attachment relationships (high WRM), marked sub-
missiveness (low DOM), and indications of borderline features (high BOR) are
often associated with a pathological need for acceptance.

Interpersonal Distrust
Problems related to the ability to trust others, experience and tolerate genuine
intimacy, and relinquish some control to others are among the most difficult to
address therapeutically. The PAR scale is the most obvious indicator of such dis-
trust, but there are many indicators that can be related to a self-protective stance
and relational ambivalence or rejection that is based on minimal expectations of
 Treatment Planning and Monitoring

others and fears of exploitation. Elevations on ARD (particularly ARD-T), SCZ-S,

BOR (particularly BOR-N), ANT, AGG-A, and/or NON all raise the possibility that
establishing trust should be considered to be a treatment goal as well as a treat-
ment obstacle. Group therapy may be of particular benefit as a conjoint therapy
for such patients.

Constriction-Rigidity
A rigid, inflexible, perfectionistic, or constricted style, such as those suggested
by an elevated ARD-O, may cause a host of problems deserving therapeutic
attention. These include overreaction or stress response to unexpected events and
change in routine, inability to experience pleasure, disrupted interpersonal rela-
tionships, fear of loss of impulse control (which may manifest itself in panic dis-
order symptoms), inefficient work habits, indecisiveness, and so forth. These traits
also may indicate the effects of an abusive or traumatic history. Problems related to
these obsessional features are exacerbated by a high need for interpersonal control
(suggested by an elevated DOM) that interferes with the ability to make necessary
compromises, and leads others to see the individual as overbearing.

Lack of Self-Confidence—Assertiveness
Lack of self-confidence, difficulty having needs met in relationships, self-
doubt, inability to act assertively, excessive preoccupation with pleasing others,
submissiveness, and inhibitions concerning expressing negative feelings to others
may be associated with any number of pathological conditions. However, if these
problems are not extreme and are not accompanied by a complex and polysymp-
tomatic clinical picture, they are quite amenable to therapeutic intervention.
Typically a behavioral deficit, rather than a behavioral excess, is involved. Any vari-
ety of therapeutic approaches, from behavioral to psychodynamic, might be appro-
priate, and short-term therapy is often effective. Indicators include elevations on
DEP-C and ARD-P or suppression on AGG, DOM, or MAN-G, particularly when
coupled with a relative lack of elevations on other scales.

Cognitive Distortions
Most psychopathology, almost by definition, involves some manifestation of
cognitive distortion. However, certain extremely negative evaluations of self, others,
and situations might profitably be explored and challenged as an early step in ther-
apy. The PAI contains a number of indicators that suggest a world view that might
impede therapeutic efforts. These cognitions could be confronted with straight
cognitive or rational-emotive therapy or with cognitive techniques integrated into
other theoretical approaches. A high NIM score indicates that an individual tends
to think in extreme and categorical terms. Substantial NIM elevations in the

273
PAI Interpretive Guide

absence of malingering indicate that the patient is reporting a profoundly negative

evaluation of self and life. If this elevation is accompanied by elevated DEP-C and
low DOM, the patient likely has a very long-standing, fixed negative self-image
that is not likely to yield to brief therapy. ANX-C elevations indicate that such
respondents are prone to experience considerable tension and worry over events
they cannot control, but which they feel they should be able to control. The DEP-C
scale, when elevated, suggests unrealistic feelings of worthlessness, failure, self-
blame, and hopelessness. PAR, or any of its subscales, can indicate a fixed belief
system involving distorted views and expectations of others. Such respondents
may distort their experience in order to attribute their misfortune to the neglect of
others, and they may see others’ successes as luck or favoritism.

The PAI in the Evaluation of Change

In addition to the applicability of the PAI for treatment planning, the instru-
ment also has many characteristics that make it well suited for the evaluation of
treatment efficacy. Newman and Ciarlo (1994) have described 11 criteria for the
selection and use of instruments as treatment outcome measures; the following
paragraphs discuss these criteria as they pertain to the PAI.
1. The outcome measure should be relevant to the target group. The PAI contains
a number of scales relevant to a wide variety of clinical conditions, and use
of the test as a pre-post measure can provide information about client
improvement in several critical areas. However, the utility of the PAI as a
treatment outcome measure will obviously vary across different target
populations; for example, little information about improvements in eating
disorders or sexual dysfunction can be gleaned from the instrument. How-
ever, the broad range of symptomatology tapped by the PAI still would pro-
vide useful information in studies with such groups, as this information
could assist in (a) identifying potentially associated problems in such
groups (e.g., depression, anxiety, or anger), and (b) allowing for increased
homogeneity for classification in such groups (e.g., differentiating within
such groups according to levels of depression, psychotic features, sub-
stance abuse, personality problems, etc.).
2. The method should be simple and teachable. The implementation of the PAI as
a treatment outcome instrument would be quite simple in most settings.
The test is self-administered, and it can also be administered by computer.
Hand-scoring the test requires no templates, and it can be accomplished by
clerical personnel in 10 minutes; optically scanned computer scoring also

Li
 Treatment Planning and Monitoring

is available. It is available for use by both English- and Spanish-speaking

individuals. Interpretation of the test is reasonably straightforward for any
clinician trained in the basics of psychometric assessment, as well as in
descriptive psychopathology. PAI interpretation is aided by the information
provided in the Professional Manual, as well as the information presented
in this interpretive guide.
The method should have objective referents. The PAI provides a number of ref-
erents against which the clinician can compare a given respondent. The T
scores are referenced against a census-matched community sample; addi-
tional transformations are available based on norms for clinical respon-
dents, college students, African Americans, and older adults. In addition,
profile data for many different diagnostic or evaluation groups are pre-
sented in this guide or in the PAI Professional Manual (Morey, 1991).
Use of multiple respondents is encouraged. A number of writers have noted
that different stake-holders (e.g., patient, therapist, spouse, independent
evaluator) can give differing portrayals of treatment outcome. The PAI was
designed as a self-report instrument intended to capture the experience of
the individual completing it; as such, it is primarily useful in capturing the
respondents perspective. The test includes validity scales that seek to
identify any systematic distortions in self-representation, but such scales
cannot substitute for the nature of information that can be obtained from
collateral informants and from clinical impressions. Thus, self-reported
improvements on the PAI, as gauged by reductions of posttreatment clini-
cal scale scores, should be supplemented with information from other
sources whenever possible.
Outcome measures should ideally identify the processes by which treatment is
producing positive effects. Newman and Ciarlo (1994) note that this criterion
is fairly controversial, as researchers often do not agree on the extent to
which treatment processes and treatment outcomes should correspond.
However, repeated administrations of the PAI could be useful in docu-
menting the process of change associated with a particular treatment. For
example, in treating depression with cognitive therapy, it is assumed that
alterations in the attribution system of the respondent will produce effects
on other types of depressive symptoms. This theoretically anticipated pat-
tern of change could be mapped by repeated administrations of the DEP
scale; initial changes on DEP-C should be observed, with changes on
DEP-A and DEP-P occurring later in the treatment process. Similarly, efforts
at establishing interpersonal trust that might be leading to personal distress

275
PAI Interpretive Guide

could be mapped by comparing the temporal pattern of changes observed

on PAR and ANX.
The measure should meet minimum criteria of psychometric adequacy. The psy-
chometric characteristics of the PAI have been described in some detail ear-
lier in this volume, and they reflect one of the primary strengths of the
instrument. The reliability of the instrument is very good, leading to stan-
dard errors of measurement that are sufficiently small to reliably detect
even small changes that might be associated with treatment. The validity of
the instrument has been documented with respect to widely used measures
of treatment-associated changes, including self-administered (e.g., Beck
Depression Inventory, State-Trait Anxiety Inventory) and clinician-rated
(e.g., Hamilton Rating Scale for Depression, Brief Psychiatric Rating Scale)
instruments.

The measure should have low costs relative to its utility. The costs associated
with a pre—post administration of the PAI for treatment outcome evaluation
are relatively minor. As a self-report instrument, it requires no professional
time to administer or score the instrument. Scoring can be accomplished
by hand in 10 minutes; alternatively, an unlimited-use computer scoring
and interpretation program is available at a one-time cost.
The measure should be easily understood by nonprofessional audiences. The
scale names and scaling procedures used in the PAI are easily understood
by most individuals. PAI scale names such as Depression or Anxiety, are
straightforward descriptions of the types of questions contained on these
scales, and the concurrent validity data support the conclusion that the scales
measure what their names imply. The linear T score is easily interpreted by
nonprofessionals, and these scores can also be expressed as percentile scores
referenced against a variety of different groups (e.g., census-matched com-
munity sample, clinical sample, or various demographic or diagnostic
groups). Although the multiple dimensions assessed by the PAI often pre-
sent a complex picture for a given respondent, the use of profiles in pre-
senting these data often render them comprehensible, even to the client.
The instrument should provide easy feedback and uncomplicated interpretation.
In many respects, this criterion is the result of meeting many of the crite-
ria described previously. In particular, ease of interpretation is precisely
what the concept of psychometric strength is designed to ensure; a test that
is reliable and valid is quite easy to interpret. In particular, the focus on
discriminant validity in the construction of the PAI was designed to facil-
itate interpretation. Many of the difficulties in interpreting measures of

276
 Treatment Planning and Monitoring

psychopathology stem from inadequate discriminant validity; it can be

quite challenging to interpret a scale that was intended to measure schizo-
phrenia if there are dozens of other factors that can lead to scale elevations.
Thus, interpreting the PAI is more straightforward than interpreting other
instruments with lower discriminant validity. In addition, the computer
interpretive report and accompanying graphical display of detailed profile
information also assists interpretation of the PAI.
10. The measure should be useful in clinical services. From its inception, the PAI
was designed to be of maximum utility in a wide variety of clinical settings.
As a pretreatment measure, the instrument provides a comprehensive
assessment of different functional areas, and also provides information crit-
ical in making diagnostic assignments. The treatment consideration scales
provide information specifically geared to determining treatment intensity
(e.g., inpatient vs. outpatient treatment) by providing an assessment of
potential for immediate crisis (e.g., suicide or assaultive behavior) as well
as the respondent's motivation for treatment and the likelihood of compli-
ance with treatment. As a posttreatment measure, the instrument provides
empirically defined normal ranges for each scale. Also, scales such as those
measuring environmental stress and social support levels provide valuable
data for determining the risk of problem relapse.
bie The instrument should be compatible with clinical theories and practices. The
development of the individual PAI scales was based on a systematic review
of the extant theories and supportive empirical research surrounding each
construct measured. Key theoretical elements that have received research
support were included in scale construction; these elements included
aspects from many different theories. Examples include cognitive mecha-
nisms in depression (DEP-C), identity disturbance in borderline personality
(BOR-D), or sensation-seeking in antisocial personality (ANT-S). Thus, rather
than adopting one theoretical approach and applying it to several different
disorders, the PAI was constructed to tap specific theoretical elements that
have received empirical support as they pertain to specific disorders.

Application of the PAI in Outcome Assessment

At a global level, a successful intervention should have the effect of moving the
respondents PAI scores in the direction of the norm for a community sample (i.e.,
50T). For most scales, this improvement would be reflected in reductions in
scores, although there are exceptions to this rule. For example, MAN-G is often
abnormally low in clinical samples, revealing very poor self-esteem; thus, increases

Pas
PAI Interpretive Guide

on MAN-G would be desirable if the score fell substantially below 50T. Increases
on RXR would also be expected over the course of a successful treatment, as many
of the motivating sources for treatment (e.g., distress or interpersonal difficulties)
would be gradually ameliorated.
PAI scores have been found to be quite stable over 1-month periods in non-
treatment samples (Morey, 1991); the reliability of the instrument would be
expected to be even higher over shorter intervals. It should be noted that most of
the scales represent constructs in a way that would not be expected to fluctuate
from moment to moment; for example, the ANX scale demonstrates a somewhat
greater correlation with “trait” anxiety than with “state” anxiety. Thus, researchers
interested in measuring momentary mood states would be better served by instru-
ments designed for that purpose. The PAI can profitably be used as a measure of
change over periods of longer duration,.and the instrument was designed to be
able to detect changes that might occur from week to week.
Determining the significance of changes in PAI scores can be accomplished
using the standard error of measurement (SEM) estimates calculated from various
reliability studies. The SEM provides an index of variability in measurement that
would be expected strictly from random fluctuations in scores; thus, changes in
scores that are less than 1 SEM cannot be interpreted with any confidence as
reflecting true change. For each of the PAI full scales, the SEM is 3 to 4 T-score
points, meaning that the 95% confidence interval for these scale scores is 5 to 6
points. As a result, changes in T-scores that are 2 SEMs (i.e., 6-8 T-score points)
in magnitude can serve as a conservative threshold for detecting statistically reli-
able change in a given client. For treatment studies where group comparisons are
involved, the statistical significance of any group difference will obviously depend
upon sample size, and, with large samples, even quite small differences might
attain statistical significance. When the PAI is used for such purposes, any group
differences should certainly be larger than the SEM for the scale before being inter-
preted as clinically meaningful.
It should be recognized that although the test-retest reliability of the PAI is
high, and, hence, scores tend to be stable, these reliability estimates were derived
from untreated samples. This does not imply that the PAI is not sensitive to change.
This was demonstrated in a study by Friedman (1995), who performed a pre—post
administration of the PAI with 22 patients during outpatient psychotherapy that
had a median duration of 3 months. Friedman reported that 19 of the 21 scales of
the PAI (excluding ICN) demonstrated statistically significant changes. However,
Friedman’ study also is valuable in that it demonstrated that the PAI scales are dif-
ferentially sensitive to the changes observed in psychotherapy, with some scales

278
 Treatment Planning and Monitoring

Table 10-9
Pre—Post Changes on PAI Scales
Effect size Direction Significance of
PAI Scale (SD change) of change change
STR OP 4 reduced p< .0001
BOR 1.74 reduced p< .0001
PIM 1.46 increased p< .0001
RXR 1.46 increased p< .0001
ANX eA reduced p< .0001
DEP 1.39 reduced p< .0001
ARD 1.34 reduced p< .0001
NIM 123 reduced p< .0001
WRM nie increased p< .0001
SCZ ks reduced p< .0001
NON 1.08 reduced p< .0001
PAR 1.05 reduced p< .0001
ANT 0.98 reduced p< .001
SOM 0.87 reduced p< .001
AGG 0.74 reduced p< .01
SUI 0.62 reduced (ars (Oh
DOM 0.62 increased p<.01
DRG 0.55 reduced (Os WS
ALC 0.50 reduced Da05
MAN 0.18 reduced ns
INF 0.07 reduced ns
Note. From Change in Psychotherapy: Foundation for Well Being Research Bulletin 106, Table 4,
by P. H. Friedman, 1995, Plymouth Meeting, PA: Foundation for Well Being. Copyright 1995 by
P. H. Friedman. Adapted with permission.

demonstrating changes that were quite substantial and others showing smaller
changes. Friedman's results are summarized in Table 10-9, with changes expressed
as effect sizes presented in units of standard deviation; thus, on the STR scale, the
reduction observed following treatment amounted to over 2 standard deviations
from the mean of pretreatment scores. This table reveals that most of the PAI scales
demonstrated large changes during treatment; generally, effect sizes of greater than
.70 are considered large effects, whereas those between .50 and .70 are considered
moderate effects (Cohen & Cohen, 1985). The pattern of changes suggests that the
largest impact of psychotherapy could be observed in reduction of negative affect
(ANX, DEP, ARD), improvement of self-esteem (PIM, RXR, BOR), and reduction of
interpersonal and environmental turmoil (STR, BOR). Although the changes in
substance abuse scales ALC and DRG were significant, only moderate effects were

279
PAI Interpretive Guide

observed. This could be expected for two reasons: first, this was not a substance
abuse treatment, and there were few significant problems of this nature in the
sample; and second, the historical nature of many of the ALC and DRG items
makes these scales somewhat less sensitive to change. For example, if someone
has ever lost a job due to alcohol abuse, this item may be endorsed even if the
person has not had a drink in 10 years. Nonetheless, the significance of changes
on the substance abuse scales demonstrates that ALC and DRG are sensitive to
treatment effects.
In the Friedman (1995) study, the only PAI scale (other than INF, which would
not be expected to change with treatment) that did not demonstrate a treatment
effect was MAN. However, this result is somewhat misleading, because, in fact, sig-
nificant changes on MAN subscales did take place. The MAN-G subscale increased
0.59 standard deviations on average, and MAN-I decreased 0.87 standard devia-
tions (no significant changes were observed on MAN-A). Thus, the opposing changes
in these two subscales canceled each other at the full scale level.
The Friedman (1995) study demonstrates that the PAI can be used to assess
improvement in a group of patients. However, the test also has been used in the
literature to study change in a particular patient. One interesting application of the
PAI as an outcome measure was reported by Saper, Blank, and Chapman (1995),
who described the treatment of a patient with visual and auditory hallucinations
that were refractory to conventional pharmacotherapy. This patient had continu-
ous auditory hallucinations, including command hallucinations, and intrusive
visions occurring roughly 10 times per day. In addition, she reported experiencing
flashbacks of traumatic events that included repeated rapes. This patient had been
treated unsuccessfully with all classes of neuroleptic medication, as well as tricyclic
antidepressants, serotonin reuptake inhibitors, lithium, carbemazepine, and ECT.
The authors described a treatment that combined an imaginal exposure (implo-
sion) treatment for the posttraumatic stress symptoms with fluphenazine medica-
tion. Saper et al. used the 11 clinical scales of the PAI and two treatment scales,
SUI and AGG, as outcome measures. They reported two measures of treatment
success: number of clinical scales reduced below 7OT, and number of scales that
decreased following treatment. Significance testing was conducted in this case
study by examining the binomial probability of each of these events occurring. In
the study, 12 of the 13 scales examined displayed decreased scores, and none of the
7 scales that had been elevated pretreatment were elevated above 70T following the
intervention. The binomial probability of either of these outcomes occurring by
chance was less than .01. These PAI changes were corroborated by a mental status
examination and staff observations at discharge. This use of the PAI is a valuable
demonstration of how decisions about outcome and improvement can be made
using a solid empirical foundation, even in the context of a case study.

280
 REFERENCES

Abramson, L. Y., Seligman, M. E. P, & Teasdale, J. D. (1978). Learned helplessness in

humans: Critique and reformulation. Journal of Abnormal Psychology, 87, 49-74.
Adams, H. B. (1964). “Mental illness” or interpersonal behavior? American Psychologist,
19, 191-197.
Akiskal, H. S., Hirschfeld, R. M., & Yerevanian, B. I. (1983). The relationship of per-
sonality to affective disorders: A critical review. Archives of General Psychiatry, 40, 801-810.
Akiskal, H. S., Yerevanian, B. I., Davis, G. C. (1985). The nosological status of border-
line personality: Clinical and polysomnographic study. American Journal of Psychiatry, 142,
192-198.
Alpher, V. S. (1995). Personality Assessment Inventory and response sets in the assess-
ment of dissociative disorder. Unpublished manuscript.
Alterman, A. I., Zaballero, A. R., Lin, M. M., Siddiqui, N., Brown, L. S., Rutherford,
M. J., & McDermott, P. A. (1995). Personality Assessment Inventory (PAI) scores of
lower-socioeconomic African American and Latino methadone maintenance patients.
Assessment, 2, 91-100.
American Psychiatric Association. (1952). Diagnostic and Statistical Manual of Mental
Disorders. Washington, DC: Author.
American Psychiatric Association. (1968). Diagnostic and Statistical Manual of Mental
Disorders (2nd ed.). Washington, DC: Author.
American Psychiatric Association. (1980). Diagnostic and Statistical Manual of Mental
Disorders (3rd ed.). Washington, DC: Author.
American Psychiatric Association. (1987). Diagnostic and Statistical Manual of Mental
Disorders (3rd ed., Rev.). Washington, DC: Author.
American Psychiatric Association. (1994). Diagnostic and Statistical Manual of Mental
Disorders (4th ed.). Washington, DC: Author.
Andraesen, N. C. (1985). Positive vs. negative symptoms: a critical evaluation. Schizo-
phrenia Bulletin, 11, 380-389.
Angrist, B., Rotrosen, J., @ Gershon, S. (1980). Differential effects of amphetamine and
neuroleptics on negative vs. positive symptoms in schizophrenia. Psychopharmacology, 72,
17-19.
Ban, T. A. (1982). Chronic schizophrenias: A guide to Leonhard’ classification. Com-
prehensive Psychiatry, 23, 155-169.
Ban, T. A., Fjetland, O. K., Kutcher, M., & Morey, L. C. (1993). CODE-DD: Devel-
opment of a diagnostic scale for depressive disorders. In 1. Hindmarch & P. Stonier
(Eds.), Human psychopharmacology: Measures and methods. (Vol. 4, pp. 73-86). Chichester,
England: Wiley.

281
PAI Interpretive Guide

Bandura, A. (1977). Self-efficacy: Toward a unifying theory of behavioral change. Psy-

chological Review, 84, 191-215.
Beck, A. T. (1967). Depression: Clinical, experimental, and theoretical aspects. New York:
Harper & Row.
Beck, A. T., & Emery, G. (1979). Cognitive therapy of anxiety and phobic disorders. New
York: Guilford.
Beck, A. T., & Steer, R. A. (1987). Beck Depression Inventory manual. San Antonio: The
Psychological Corporation.
Beck, A. T., & Steer, R. A. (1988). Beck Hopelessness Scale manual. San Antonio: The Psy-
chological Corporation.
Beck, A. T., & Steer, R. A. (1990). Beck Anxiety Inventory manual. San Antonio: The Psy-
chological Corporation.
Bell, M. J., Billington, R., & Becker, B. (1985). A scale for the assessment of object rela-
tions: Reliability, validity, and factorial invariance. Journal of Clinical Psychology, 42, 733-
Teale
Bell, M. J., Billington, R., Cicchetti, D., & Gibbons, J. (1988). Do object relations
deficits distinguish BPD from other diagnostic groups? Journal of Clinical Psychology, 44,
511-516.
Bell-Pringle, V. J. (1994). Assessment of borderline personality disorder using the MMPI-2
and the Personality Assessment Inventory. Unpublished doctoral dissertation, Georgia State
University, Atlanta.
Benjamin, L. S. (1974). Structural analysis of social behavior. Psychological Review, 81,
392-425.
Bernadt, M. W,, Mumford,J.,Taylor, C., et al. (1982). Comparison of questionnaire and
laboratory tests in the detection of excessive drinking and alcoholism. Lancet, 325-328.
Beskow, J. (1979). Suicide and mental disorder in Swedish men. Acta Psychiatrica Scan-
dinavica Supplementum, 277, 1-138.
Billings, A. G., @ Moos, R. H. (1984). Coping, stress, and social resources among adults
with unipolar depression. Journal of Personality and Social Psychology, 46, 877-891.
Bishop, E. R., & Holt, A. R. (1980). Pseudopsychosis: A reexamination of the concept
of hysterical psychosis. Comprehensive Psychiatry, 21, 150-161.
Blashfield, R. K., @ McElroy, R. A. (1987). The 1985 journal literature on the person-
ality disorders. Comprehensive Psychiatry, 28, 536-546.
Bleuler, E. (1950). Dementia Praecox or the group of Schizophrenias (J. Zinkin, Trans.).
New York: International Universities Press. (Original work published 1923)
Bongar, B. (1991). The suicidal patient: Clinical and legal standards of care. Washington,
DC: American Psychological Association.
Boyle, G. J., & Lennon, T. J. (1994). Examination of the reliability and validity of the
Personality Assessment Inventory. Journal of Psychopathology and Behavior Assessment, 16,
173-188.
Buss, A. H., @ Durkee, A. (1957). An inventory for assessing different kinds of hostil-
ity. Journal of Counseling Psychology, 21, 343-349.

282
 References

Campbell, D. T., & Fiske, D. W. (1959). Convergent and discriminant validatio

n by the
multitrait-multimethod matrix. Psychological Bulletin, 56, 81-105.
Carlson, G. A., & Goodwin, F K. (1973). The stages of mania: A longitudinal analysis
of the manic episode. Archives of General Psychiatry, 28, 221-228.
Cashel, M. L., Rogers, R., Sewell, K., & Martin-Cannici, C. (1995). The Personalit
y
Assessment Inventory and the detection of defensiveness. Assessment, 2, 333-342.
Cherepon, J. A., &@Prinzhorn, B. (1994). The Personality Assessment Inventory profiles
of adult female abuse survivors. Assessment, 1, 393-400.
Clark, D. C., & Fawcett,J.(1992). Review of empirical risk factors for evaluation of the
suicidal patient. In B. Bongar (Ed.), Suicide: Guidelines for assessment, management, and
treatment (pp. 16-48). New York: Oxford University Press.
Cleckley, H. (1941). The mask of sanity. St Louis, MO: Moseby.
Colligan, R. C. & Offord, K. P (1990). MacAndrew versus MacAndrew: The relative
efficacy of the MAC and SAP scales for the MMPI in screening male adolescents for sub-
stance misuse. Journal of Personality Assessment, 55, 708-716.
Costa, P T., @ McCrae, R. R. (1985). The NEO Personality Inventory manual. Odessa, FL:
Psychological Assessment Resources.
Costa, P T., @ McCrae, R. R. (1989). The NEO-PI/FFI manual supplement. Odessa, FL:
Psychological Assessment Resources.
Costa, P T., &@ McCrae, R. R. (1992). Normal personality in clinical practice: The NEO
Personality Inventory. Psychological Assessment: A Journal of Consulting and Clinical Psychol-
ogy, 4, 5-13.
Cronbach, L. J. (1951). Coefficient alpha and the internal structure of tests. Psychome-
trika, 16, 297-334.
Crowne, D. P, & Marlowe, D. (1964). The approval motive: Studies in evaluative depen-
dence. New York: Wiley.
Cull, J. G., & Gill, W. S. (1982). Suicide Probability Scale manual. Los Angeles: Western
Psychological Services.
Deffenbacher, J. L. (1992). Trait anger: Theory, findings, and implications. In C. D.
Spielberger & J. N. Butcher (Eds.), Advances in Personality Assessment (Vol. 9, pp. 177-
201). Hillsdale, NJ: Erlbaum.
Deisinger, J. A. (1994). Relationship between coping style and psychopathology in a com-
munity sample. Doctoral Dissertation, Illinois Institute of Technology, Chicago.
Edwards, G., Arif, A., @ Hodgson, R. (1982). Nomenclature and classification of drug-
and alcohol-related problems: A shortened version of a WHO memorandum. British Jour-
nal of Addiction, 77, 259-273.
Edwards, G., & Gross, M. M. (1976). The alcohol dependence syndrome: Provisional
description of a clinical syndrome. British Medical Journal, 1, 1058-1061.
Endler, N. S., & Edwards,J.M. (1988). Personality disorders from an interactional per-
spective. Journal of Personality Disorders, 2, 326-333.
Epstein, S., & O’Brien, E. J. (1985). The person-situation debate in historical and cur-
rent perspective. Psychological Bulletin, 98, 513-537.

283
PAI Interpretive Guide

Fals-Stewart, W. (1996). The ability of individuals with psychoactive substance use dis-
orders to escape detection by the Personality Assessment Inventory. Psychological Assess-
ment, 8, 60-68.
Faravelli, C., Albanesi, G., & Poli, E. (1986). Assessment of depression: a comparison
of rating scales. Journal of Affective Disorders, 11, 245-253.
Finney, D. J. (1971). Probit Analysis. Cambridge, England: Cambridge University Press.
Frances, A., Clarkin, J., & Perry, S. (1984). Differential therapeutics in psychiatry: The art
and science of treatment selection. New York: Brunner/Mazel.
Friedman, P. H. (1995). Change in psychotherapy: Foundation for Well Being Research Bul-
letin 106. Foundation for Well Being, Plymouth Meeting, PA.
Fuller, R. K., Lee, K. K., & Gordis, E. (1988). Validity of self-report in alcoholism
research: Results of a Veterans Administration cooperative study. Alcoholism: Clinical and
Experimental Research, 12, 201-205.
Gaies, L. A. (1993). Malingering of depression on the Personality Assessment Inventory.
Doctoral Dissertation, University of South Florida, Tampa.
Greenstein, D. S. (1993). Relationship between frequent nightmares, psychopathology, and
boundaries among incarcerated male inmates. Unpublished doctoral dissertation, Adler
School of Professional Psychology, Chicago, IL.
Goodwin, F K., & Jamison, K. R.. (1990). Manic-depressive illness. New York: Oxford
University Press.
Govia, J. M., & Velicer, W. F (1985). Comparison of multidimensional measures of
aggression. Psychological Reports, 57, 207-215.
Grinker, R. R., Werble, B., & Drye, R. C. (1968). The borderline syndrome. New York:
Basic Books.
Gunderson, J. G., & Elliott, G. R. (1985). The interface between borderline personality
disorder and affective disorder. American Journal of Psychiatry, 142, 277-288.
Gunderson, J.G., & Singer, M. (1975). Defining borderline patients: An overview.
American Journal of Psychiatry, 132, 1-10.
Guze, S. B., Woodruff, R. A., & Clayton, P. J. (1971). A study of conversion symptoms
in psychiatric outpatients. American Journal of Psychiatry, 128, 643-647.
Hamilton, M. (1960). A rating scale for depression. Journal of Neurology, Neurosurgery,
and Psychiatry, 23, 56-62.
Hare, R. D. (1985). Comparison of procedures for the assessment of psychopathy. Jour-
nal of Consulting and Clinical Psychology, 53, 7-16.
Harpur, T. J., Hare, R. D., & Hakstian, R. (1989). A two-factor conceptualization of psy-
chopathy: Construct validity and implications for assessment, Psychological Assessment, 1,
6-17.
Hart, S. D., Kropp, P R., & Hare, R. D. (1988). Performance of male psychopaths fol-
lowing conditional release from prison. Journal of Consulting and Clinical Psychology, 56,
221-232,
Hathaway, S. R., & McKinley, J. C. (1967). MMPI manual (Rev. ed.). New York: Psy-
chological Corporation.

284
 References

Helmes, E. (1993). A modern instrument for evaluating psychopathology: The Person-

ality Assessment Inventory Professional Manual. Journal of Personality Assessment, 61, 414-
417.
Herman) Jel Perry, J; C.,.So Vanider Kolk, B. A»(1989).. Childhood trauma in border-
line personality disorder. American Journal of Psychiatry, 146, 490-495.
Hesselbrock, M. N., Babor, T. F, Hesselbrock, V, Meyer, R. E., @ Workman, K. (1983).
“Never believe an alcoholic”? On the validity of self-report measures of alcohol depen-
dence and related constructs. International Journal of the Addictions, 18, 563-609.
Hollon, S. D., & Mandell, M. (1979). Use of the MMPI in the evaluation of treatment
effects. In J. Butcher (Ed.), New developments in the use of the MMPI. Minneapolis: Univer-
sity of Minnesota Press.
Holmes, T. H., & Rahe, R. H. (1967). The Social Readjustment Rating Scale. Journal of
Psychosomatic Research, 11, 213-218.
Horney, K. (1945). Our Inner Conflicts. New York: Norton.
Hyler, S. E., Rieder, R. O., Williams, J. B. W, Spitzler, R. L., Hendler, J., & Lyons, M.
(1988). The Personality Diagnostic Questionnaire: Development and preliminary results.
Journal of Personality Disorders, 2, 229-237.
Jackson, D. N. (1970). A sequential system for personality scale development. In C. D.
Spielberger (Ed.), Current topics in clinical and community psychology (Vol. 2, pp. 62-97).
New York: Academic Press.
Jackson, D. N. (1971). The dynamics of structured personality tests. Psychological
Review, 78, 229-248.
Jellinek, E. M. (1960). The disease concept of alcoholism. New Haven, CT: College and
University Press.
Kagan, J., & Moss, H. A. (1962). Birth to maturity: A study in psychological development.
New York: Wiley.
Karasu, T. B. (1990a). Toward a clinical model of psychotherapy for depression: I.
Systematic comparison of three psychotherapies. American Journal of Psychiatry, 147,
133-147.
Karasu, T. B. (1990b). Toward a clinical model of psychotherapy for depression: II. An
integrative and selective treatment approach. American Journal of Psychiatry, 147, 269-278.
Keane, T. M., Caddell, J. M., & Taylor, K. L. (1988). Mississippi scale for combat-
related posttraumatic stress disorder: Three studies in reliability and validity. Journal of
Consulting and Clinical Psychology, 56, 85-90.
Kenrick, D. T., & Funder, D. C. (1988). Profiting from controversy: Lessons from the
person-situation debate. American Psychologist, 43, 23-34.
Kernberg, O. F (1975). Borderline conditions and pathological narcissism. New York: Jason
Aronson.
Kiesler, D. J. (1983) The 1982 interpersonal circle: A taxonomy for complementarity in
human transactions. Psychological Review, 90, 185-214.
PAI Interpretive Guide

Koksal, F, & Power, K. G. (1990). Four systems anxiety questionnaire (FSAQ): A self-
report measure of somatic, cognitive, behavioral, and feeling components. Journal of Per-
sonality Assessment, 54, 534-545.
Kurtz, J. E., Morey, L. C., & Tomarken, A. J. (1993). The concurrent validity of three
self-report measures of borderline personality. Journal of Psychopathology and Behavioral
Assessment, 15, 255-266.
Lambert, M. J. (1991). Introduction to psychotherapy research. In L. E. Beutler &
M. Crago (Eds.), Psychotherapy research: An international review of programmatic studies.
Washington, D.C.: American Psychological Association.
Lang, P J. (1971). The application of psychophysiological methods. In S. Garfield &
A. Bergin (Eds.), Handbook of psychotherapy and behavior change (pp. 75-125). New York:
Wiley.
Lazarus, R. S., & Cohen, J. P (1977). Environmental stress. In I. Altman @ J. EF
Wohlwill (Eds.), Human behavior and the environment: Current theory and research. New
York: Plenum.
Leary, T. (1957). Interpersonal Diagnosis of Personality. New York: Ronald.
Lenzenweger, M. FE, Dworkin, R. H., & Wetherington, E. (1989). Models of positive
and negative symptoms in schizophrenia: An empirical evaluation of latent structures.
Journal of Abnormal Psychology, 98, 62-70.
Linehan, M. M., Chiles, J. A., Egan, K. J., @ Devine, R. H. (1986). Presenting problems
of parasuicides versus suicide ideators and nonsuicidal psychiatric patients. Journal of Con-
sulting and Clinical Psychology, 54, 880-881.
Loeber, R. & Dishion, T. (1983). Early predictors of male delinquency: A review. Psy-
chological Bulletin, 94, 68-99.
Loevinger, J. (1957). Objective tests as instruments of psychological theory. Psychologi-
cal Reports, 3, 635-694.
Louks, J., Hayne, C., & Smith, J. (1989). Replicated factor structure of the Beck Depres-
sion Inventory. Journal of Nervous and Mental Disease, 177, 473-479.
MacAndrew, C. (1965). The differentiation of male alcoholic outpatients from non-
alcoholic psychiatric patients by means of the MMPI. Quarterly Journal of Studies on Alco-
hol, 26, 238- 246.
Maris, R., Berman, A., Maltsberger, J. T., & Yufit, R. 1. (1992). Assessment and prediction
of suicide. New York: Guilford.
McCrae, R. R., & Costa, P. T. (1984). Emerging lives, enduring dispositions: Personality in
adulthood. Boston: Little-Brown.
McGlashan, T. H. (1983). The borderline syndrome. I. Testing three diagnostic systems.
Archives of General Psychiatry, 40, 1311-1318.
McLemore, C. W. & Benjamin, L. S. (1979). Whatever happened to interpersonal diag-
nosis? A psychosocial alternative to DSM-III. American Psychologist, 34, 17-34.
McNiel, D. E., & Binder, R. L. (1994). The relationship between acute psychiatric
symptoms, diagnosis, and short-term risk of violence. Hospital and Community Psychiatry,
45, 133-137.

286
 References

Meehl, P_E., & Rosen, A. (1955). Antecedent probability and the efficiency of psycho-
metric signs, patterns, or cutting scores. Psychological Bulletin, 52, 194-216.
Megargee, E. I. (1970). The prediction of violence with psychological tests. In C. D.
Spielberger (Ed.), Current topics in clinical and community psychology (Vol. 2, pp. 98-156).
New York: Academic Press.
Montag, I., & Levin,.J. (1994). The five factor model and psychopathology in nonclin-
ical samples. Personality and Individual Differences, 17, 1-7.
Moran, P. W,, & Lambert, M. J. (1983). A review of current assessment tools for mon-
itoring change in depression. In M. J. Lambert, E. R. Christensen, & S. S. DeJulio (Eds.),
The measurement of psychotherapy outcome in research and evaluation (pp. 263-303). New
York: Wiley.
Morey, L. C. (1985). An empirical comparison of interpersonal and DSM-III approaches
to classification of personality disorders. Psychiatry, 48, 358-364.
Morey, L. C. (1991). The Personality Assessment Inventory Professional Manual. Odessa,
FL: Psychological Assessment Resources.
Morey, L. C., &@Henry, W. (1994). Personality Assessment Inventory. In M. Maruish
(Ed.), The use of psychological testing for treatment planning and outcome assessment (pp. 185-
216). Hillsdale, NJ: Erlbaum.
Morey, L. C., Waugh, M. H., & Blashfield, R. K. (1985). MMPI scales for DSM-III per-
sonality disorders: Their derivation and correlates. Journal of Personality Assessment, 49,
245-251,
Motto, J. A. (1989). Problems in suicide risk assessment. In D. G. Jacobs & H. N.
Brown (Eds.), Suicide: Understanding and responding. Harvard Medical School perspectives on
suicide (pp. 129-142). Madison, CT: International Universities Press.
National Center for Health Statistics. (1992). Advance report of final mortality statis-
tics, 1989. NCHS Monthly Vital Statistics Report, 40, 8 (suppl).
Newman, F L., & Ciarlo, J. A. (1994). Criteria for selecting psychological instruments
for treatment outcome assessment. In M. Maruish (Ed.), The use of psychological testing for
treatment planning and outcome assessment (pp. 98-110). Hillsdale, NJ: Erlbaum.
Osborne, D. (1994, April). Use of the Personality Assessment Inventory with a medical pop-
ulation. Paper presented at the meetings of the Rocky Mountain Psychological Association,
Denver, CO.
Overall, J. E., & Gorham, D. R. (1962). The Brief Psychiatric Rating Scale. Psychologi-
cal Reports, 10, 799-812.
Parker, J. G., & Asher, S. R. (1987). Peer relations and later personal adjustment: Are
low-accepted children at risk? Psychological Bulletin, 102, 357-389.
Perry, S., Frances, A., & Clarkin, J. (1988). A DSM-III-R casebook of treatment selection.
New York: Brunner-Mazel.
Pinel, P (1801). Traite medico-philosophique sur Valienation mentale, ou la manie. Paris:
Caille et Ravier. ;
Pope, H. G., Jonas, J. M., Hudson, J. L., Cohen, B. M., & Gunderson, J. G. (1983).
The validity of DSM-III borderline personality disorder. Archives of General Psychiatry, 40,
23-30.

287
PAI Interpretive Guide

Procidano, M. E., & Heller, K. (1983). Measures of perceived social support from
friends and from family: Three validation studies. American Journal of Community Psychol-
ogy, tinti=24i
Rachman, S. J., @ Hodgson, R. J. (1980). Obsessions and compulsions. Englewood Cliffs,
NJ: Prentice-Hall.
Reynolds, W. M. (1982). Development of reliable and valid short forms of the Marlowe-
Crowne Social Desirability Scale. Journal of Clinical Psychology, 38, 119-125.
Rich, C. L., Young, D., & Fowler, R. C. (1986). San Diego suicide study: I. Young vs.
old subjects. Archives of General Psychiatry, 43, 577-582.
Riley, W. T., & Treiber, F A. (1989). The validity of multidimensional self-report anger
and hostility measures. Journal of Clinical Psychology, 45, 397-404.
Robins, E. (1981). The final months: A study of the lives of 134 persons who committed sut-
cide. New York: Oxford University Press.
Robins, E., Murphy, G. E., Wilkinson, R. H., Gassner, S., & Kayes,J.(1959). Some clin-
ical considerations in the prevention of suicide based on a study of 134 successful sui-
cides. American Journal of Public Health, 49, 888-897.
Robins, L. N. (1966). Deviant children grown up: A sociological and psychiatric study of
sociopathic personality. Baltimore: Williams & Wilkins.
Rogers, R., Flores, J., Ustad, K., & Sewell, K. W. (1995). Initial validation of the Per-
sonality Assessment Inventory-Spanish Version with clients from Mexican American com-
munities. Journal of Personality Assessment, 64, 340-348.
Rogers, R., Ornduff, S. R., & Sewell, K. (1993). Feigning specific disorders: A study of
the Personality Assessment Inventory (PAI). Journal of Personality Assessment, 60, 554-560.
Rogers, R., Sewell, K. W,, Morey, L. C., & Ustad, K. L. (in press). Detection of feigned
mental disorders on the Personality Assessment Inventory: A discriminant analysis. Jour-
nal of Personality Assessment.
Saper, Z., Blank, M. K., & Chapman, L. (1995). Implosive therapy as an adjunctive
treatment in a psychotic disorder: A case report. Journal of Behavior Therapy and Experi-
mental Psychiatry, 26, 157-160.
Schinka, J. A. (1995). Personality Assessment Inventory scale characteristics and factor
structure in the assessment of alcohol dependency. Journal of Personality Assessment, 64,
101-111.
Schinka, J. A., & Borum, R. (1993). Readability of adult psychopathology inventories.
Psychological Assessment, 5, 384-386.
Schinka, J. A., & Vanderploeg, R. D. (1995, August). Validity of the NST in a drug-depen-
dent sample. Paper presented at the meetings of the American Psychological Association,
New York.
Schlosser, B. (1992). Computer assisted practice. The Independent Practitioner, 12,
12-15.
Selzer, M. L. (1971). The Michigan Alcoholism Screening Test: The quest for a new
diagnostic instrument. American Journal of Psychiatry, 127, 1653-1658.

288
 References

Serin, R., Peters, R. D., & Barbaree, H. E. (1990). Predictors of psychopathy and release
outcome in a criminal population. Psychological Assessment, 2, 419-422.
Shaffer, C. E., Waters, W. F, & Adams, S. G. (1994), Dangerousness: Assessing the
risk
of violent behavior. Journal of Consulting and Clinical Psychology, 62, 1064-1068.
Shneidman, E. S., Farberow, N. L., & Litman, R. E. (1970). The psychology of suicide.
New York: Jason Aronson.
Siegel, J. S. (1985). The measurement of anger as a multidimensional construct. In M.
A. Chesney & R. H. Rosenman (Eds.), Anger and hostility in cardiovascular and behavioral
disorders. New York: Hemisphere.
Sifneos, P. E. (1987). Short-term dynamic psychotherapy: evaluation and technique (2nd
ed.). New York: Plenum.
Skinner, H. A. (1982). The drug abuse screening test. Addictive Behaviors, 7, 363-371.
Sloane, R. B., Staples, FE R., Cristol, A. H., Yorkston, N. J., &@Whipple, K. (1975). Short-
term analytically-oriented psychotherapy vs. behavior therapy. Cambridge, MA: Harvard Uni-
versity Press.
Smith, M. L., Glass, G. V,, & Miller, T. I. (1980). The benefits of psychotherapy. Baltimore:
Johns Hopkins University Press.
Sobell, L. C., Sobell, M. B. (1975). Outpatient alcoholics give valid self-reports. Journal
of Nervous and Mental Disease, 161, 32-42.
Soloff, P H., George, A., Nathan, R., & Schulz, P M. (1989). Amitriptyline versus
haloperidol in borderlines: Final outcomes and predictors of response. Journal of Clinical
Psychopharmacology, 9, 238-246.
Spielberger, C. D. (1983). Manual for the State-Trait Anxiety Inventory. Palo Alto, CA:
Consulting Psychologists Press.
Spielberger, C. D. (1988). State-Trait Anger Expression Inventory. Odessa, FL: Psycho-
logical Assessment Resources.
Strupp, H. H., & Binder, J. (1984). Psychotherapy in a new key. New York: Basic Books.
Sullivan, H. S. (1953). The Interpersonal Theory of Psychiatry. New York: Norton.
Trapnell, P. D., & Wiggins, J. S. (1990). Extension of the Interpersonal Adjective Scale
to include the big five dimensions of personality. Journal of Personality and Social Psychol-
ogy, 59, 781-790.
Trull, T. J. (1995). Borderline personality disorder features in nonclinical young adults:
1. Identification and validation. Psychological Assessment, 7, 33-41.
Trull, T. J., Useda, J. D., Conforti, K., @ Doan, B. T. (1995, August). Two-year outcome
of subjects with borderline features. Paper presented at the meeting of the American Psycho-
logical Association, New York.
Wahler, H. J. (1983). Wahler Physical Symptoms Inventory. (1983). Los Angeles: Western
Psychological Services.
Waldinger, R. J., & Gunderson, J. G. (1987). Effective psychotherapy with borderline
patients: Case studies. New York: MacMillan.

289
PAI Interpretive Guide

Wiggins, J. S. (1966). Substantive dimensions of self-report in the MMPI item pool.

Psychological Monographs, 80, 22 (whole No. 630).
Wiggins, J. S. (1973). Personality and prediction: Principles of Personality Assessment.
Reading, MA: Addison-Wesley.
Wiggins, J. S. (1982). Circumplex models of interpersonal behavior in clinical psy-
chology. In P. Kendall & J. Butcher (Eds.), Handbook of Research Methods in Clinical Psy-
chology. New York: Wiley.
Wiggins,J.S. (1987, August). How interpersonal are the MMPI personality disorder scales?
Paper presented at the meeting of the American Psychological Association, New York.
Wiggins, J. S. (1995). Interpersonal Adjective Scales. Odessa, FL: Psychological Assess-
ment Resources.
Wiggins, J. S., Trapnell, P, & Phillips, N. (1988). Psychometric and geometric charac-
teristics of the revised Interpersonal Adjective Scales (IAS-R). Multivariate Behavioral
Research, 23, 517-530.
Wolpe, J., & Lang, P. (1964). A fear survey schedule for use in behavior therapy. Behav-
ior Research and Therapy, 2, 27-30.
Zanarini, M. (1987). Diagnostic Interview for Personality Disorders—Revised. Belmont, MA:
McLean Hospital.
Zelin, M. L., Adler, G., & Myerson, P. G. (1972). Anger self-report: An objective ques-
tionnaire for the measurement of aggression. Journal of Consulting and Clinical Psychology,
39.340:
Zung, W. W. (1965). A self-rating depression scale. Archives of General Psychiatry, 12,
63-70.
 APPENDIX
INDEX CORRELATIONS
WITH CLINICAL, PERSONALITY,
AND VALIDITY INDICATORS

291
PAI Interpretive Guide

Table A-1
Index Correlations With PAI Full Scale Scores

PAI scales MAL DEF CDF SAEst_ TPI VPI SPI RDF

ICN .07 = 58) 01 16 18 19 MS .28

INF .00 —.08 18 04 02 .03 .08 7
NIM 61 ale .26 .48 .64 .66 .69 .09
PIM = 82) .56 .06 = 0) =157/ —.54 — 67 10
SOM 35 20 13 aif ES 32 .58 .10
ANX 46 —A49 mly/) .28 .50 46 82 .07
ARD .50 = 510 (EE 38 .60 Be .80 OZ
DIE: 40 —.47 alii 2 .50 42 82 Ks)
MAN 33 24 .40 47 44 53 .30 =A)
PAR OO ais) .26 49 .66 .68 .65 35
SCZ MN, =o 7 45 BE .65 A) 16
BOR 45 —.48 .28 .67 AT 74 83 = 01
ANT 25 —.06 07 94 .68 74 37 .02
ALC —.03 0S = [622 .39 33 39 ei = 10
DRG .05 —.22 10 54 46 fos .30 .03
AGG oT —.24 PA ls Ane) (35 44 .08
SUI 32 —.34 ae 38 47 44 .65 .06
STR 33 = —.14 42 DNS) 53 .64 =|
NON 34 —.24 W .39 .56 158 .66 Af
RXR = 21 52 .26 Sy —.46 —A1 =e .26
DOM =,09 49 10 .06 (VE .00 = 87 —.23
WRM ee 40 .06 =U S83 See, Ol —.33
Note. Sample of clinical patients, n = 447. MAL = Malingering Index; DEF = Defensiveness Index;
CDF = Cashel Discriminant Function; SA Est = estimated Substance Abuse scale scores; TP! =
Treatment Process Index; VPI = Violence Potential Index; SPI = Suicide Potential Index; RDF =
Rogers Discriminant Function.

292
 Index Correlations With Clinical, Personality, and Validity Indicators

Table A-2
index Correlations With PAl Subscale Scores

PAI
subscales MAL DEF CDF SAEst_ TPI VPI SPI RDF
SOM-C .36 —1(3;<" AB 23 .39 .38 52 =07
SOM-S 31 —.26 Ao 11 32 26 56 sl7A
SOM-H .26 15 10 11 27 23 A7 .16
ANX-C 43 —.42 13 25 48 A 77 02
ANX-A 40 ~.50 14 7 A6 43 as — 01
ANX-P 45 ~.39 .20 BR 46 44 73 .22
ARD-O .39 12 21 .20 37 40 54 Sl
ARD-P 34 —.45 11 aly, 36 25 61 .28
ARD-T 44 —.29 19 44 63 64 72 —.04
DEP-C 38 = on Aly .30 50 42 res a2,
DEP-A .38 —.46 10 25 48 AN 78 .09
DEP-P 32 =33 .09 12 .38 .29 68 NZ
MAN-A .29 07 32 .40 .37 44 38 Salts!
MAN-G = 03 52 .23 12 04 nie ane —21
MAN-I 54 =i4 34 58 59 62 56 =—.05
PAR-H A <i 19 48 61 65 .60 (ie
PAR-P 65 —.08 124 43 55 .60 49 .26
PAR-R 34 —.29 23 35 53 48 58 43
SCZ-P 46 =10 18 36 .40 49 43 13

SCZ-S 26 —.36 05 .30 45 48 64 30

SCZ-T 51 —.33 .20 A 53 55 76 —.06

BOR-A 42 —.45 28 55 .66 65 79 02

BOR-I .37 —.39 223 At 59 49 Al —.09

BOR-N 35 =,34 2a) 46 62 60 .67 21

BOR-S 33 —.26 19 81 69 .69 56 -.18

ANT-A .08 —14 —.09 79 58 .60 .29 —.04

ANT-E 35 10 .08 .70 53 55 24 il

ANT-S 25 —.08 23 82 57 .67 a7, .03

AGG-A 32 =19 16 63 56 57 50 .09

AGG-V 13 he, .20 48 31 36 .07 .06

AGG-P 34 2s 18 79 63 72 50 .05
a E TESST Tan
c
OT a a
DEF = Defensiveness Index;
Note. Sample of clinical patients, n = 447. MAL = Malingering Index;
Est = estimate d Substan ce Abuse scale scores; TPI =
CDF = Cashel Discriminant Function; SA
VPI = Violence Potential Index; SPI = Suicide Potential Index; ale =
Treatment Process Index;
Rogers Discriminant Function.

293
PAI Interpretive Guide

Table A-3
With
Index Correlations8 MMPI Clinical and Validity Scales
88
5

MMPI
scales MAL DEF CDF SAEst _ TPI VPI SPI RDF

L —.08 55 .04 a =.23 =F —.40 .02

F .39 aie .24 .33 55 52 63 13
K awe 25 —29 —.42 —.46 —.49 —.59 07
Hs .06 =10 —12 —03 21 10 25 18
D 22 = 80 A WW 36 22 53 26
Hy —=02 aid —.05 SAF 10 —.06 2 14
Pd .08 =e —.15 ot A7 33 .38 .04
Mf 21 25 .00 .40 26 .29 01 —.03
Pa 38 =20 14 18 45 .39 53 .05
Pt 48 =23 ~.09 28 AI 33 49 .00
Sc 26 —.24 .09 .39 56 51 61 .09
Ma 19 .20 ae 41 40 45 32 —.25
Si .28 —.45 .23 .00 25 21 49 Si
Note. Sample of clinical patients, n = 91. MMPI = Minnesota Multiphasic Personality Inventory;
MAL = Malingering Index; DEF = Defensiveness Index; CDF = Cashel Discriminant Function;
SA Est = estimated Substance Abuse scale scores; TPI = Treatment Process Index; VPI = Violence
Potential Index; SPI = Suicide Potential Index; RDF = Rogers Discriminant Function.

294
 Index Correlations With Clinical, Personality, and Validity Indicators

Table A-4
Index Correlations With Wiggins MMPI Content Scales*
MAL DEF CDF SAEst_ TPI VPI SPI RDF
HEA 38 —41 eee 22 44 40 57, sf
DEP 39 =e .40 .60 70 .63 83 16
ORG 33 =i, 36 m2 36 36 49 ne
FAM GE, aS a 49 .50 .40 54 .01
AUT 42 14 aL 43 .39 42 47 al
FEM 01 =18 .08 .03 .04 .02 .03 ee
REL .23 00 .08 .03 .04 02 .01 .09
HOS 40 —.14 48 .68 .62 .63 .63 202
MOR 43 —49 .26 48 61 aye ATA .08
PHO 13 ob 33 .03 19 18 30 SIS)
PSY 49 a0 .63 40 9 .65 .69 a2
AYE .27 fee 61 40 oH .40 39 = 09
SOC aA —.28 .09 .02 V2 US .26
.28
E
ieee ee ern E
Note. Sample of clinical patients, n = 91. MMPI = Minnesota Multiphasic Personality Inventory;
MAL = Malingering Index; DEF = Defensiveness Index; CDF = Cashel Discriminant Function;
SA Est = estimated Substance Abuse scale scores; TPI = Treatment Process Index; VPI = Violence
Potential Index; SPI = Suicide Potential Index; RDF = Rogers Discriminant Function.
4J. S. Wiggins, 1966.

295
 PAI Interpretive Guide

Table A-5
Index Correlations With Indicators of Clinical Symptomatology
MAL DEF CDF SAEst_ TPI VPI SPI RDF
Beck Depression
Inventory (BDI)* 39 —.28 —.08 01 35 21 .63 ALS
Beck Anxiety
Inventory (BAI)? .38 —.26 01 10 .30 .25 eyed .09
Hamilton Rating
Scale for
Depression
(HAM-D)° .05 —.47 19 04 .30 —.11 silt ih
Mississippi
Combat-related
PTSD scale? 46 a2 48 49 wie, 63 .56 a2
Self-report
Psychopathy
scale (SRP)° 25 —.36 EU 70 .50 .67 .59 =.07
Suicide Probability
Scale (SPS),
Hopelessness 24 19 —.10 .06 45 .39 .64 —12
Suicide Probability
Scale (SPS),
Suicidal Ideation 14 .01 —.16 .03 42 ES .65 -—.19
Suicide Probability
Scale (SPS),'
Negative Self-
evaluation .14 =—.01 —.19 —.17 ro 18 48 105
Suicide Probability
Scale (SPS),‘
Hostility 19 215 —.11 P| 44 .40 39 —.08
Suicide Probability
Scale (SPS),'
Total score .24 We —.10 .07 .46 .39 .63 —.10
Wahler Physical
Symptoms
Inventory (WPSI)9 —.16 —.79 —.23 —.07 Pp) —.08 50 29
Note. Sample of clinical patients, n = 91. MAL = Malingering Index; DEF = Defensiveness Index;
CDF = Cashel Discriminant Function; SA Est = estimated Substance Abuse
scale scores; TPI =
Treatment Process Index; VPI = Violence Potential Index; SPI = Suicide
Potential Index; RDF =
Rogers Discriminant Function.
°A. T. Beck & R. A. Steer, 1987. °A. T. Beck & R.A. Steer, 1990. °M. Hamilton,
1960. °T. M. Keane,
J. M. Caddell, & K. L. Taylor, 1988. °R. D. Hare, 1985. J. G. Cull & W. S.
Gill, 1982. 9H. J. Wahler,
1983.

296
 Index Correlations With Clinical, Personality, and Validity Indicators

Table A-6
Index Correlations With MMPI Personality Disorder Scales*
MMPI scale MAL DEF CDF SA Est TPI VPI SPI RDF
Histrionic —.06 41 04 12 =f Queene2 a4 95
Narcissistic 01 45 20 =01 =07— = 06 ~.34 — 25
Borderline .29 — 30 36 64 55 53 68 SO
Antisocial 16 20 20 59 53 52 53 = 06
Dependent 42 =e 417 36 46 44 67 23
Compulsive es: =a 6 34 pe, 40 44 57 —.16
Passive-
Aggressive .60 2107 33 49 57 60 81 .06
Paranoid .58 00 51 39 56 59 66 sli
Schizotypal 44 = 23 37 20 44 49 61 20
Avoidant .36 = 41 21 15 37 39 63 29
Schizoid 14 Salo 21 —.08 a 15 31 32
e
ee e.
eee eee ee .. eee
Note. Sample of clinical patients, n = 91. MMPI = Minnesota Multiphasic Personality Inventory,
MAL = Malingering Index; DEF = Defensiveness Index; CDF = Cashel Discriminant Function;
SA Est = estimated Substance Abuse scale scores; TPI = Treatment Process Index; VPI = Violence
Potential Index; SPI = Suicide Potential Index; RDF = Rogers Discriminant Function.
aL. C. Morey, MH. Waugh, & R. K. Blashfield, 1985.

297
PAI Interpretive Guide

Table A-7
Index Correlations With Personality Disorder Questionnaire—
Revised? DSM-III-R Personality Disorder Scale Scores
MAL DEF CDF SAEst_ TPI VPI SPI RDF

10 .07 Py .23 14 .22 .20

Histrionic PT
.30 —.16 45 45 .28 9 “he
Narcissistic 32
—.37 .06 .36 49 .24 19 aS
Borderline [25
.03 —.34 —.32 58 Sit 25 NS Ae
Antisocial
—.27 —.03 .05 Ald 01 .05 ME
Dependent .09
Obsessive-
.25 .20 10 .33 .28 .28 “ATE whe
Compulsive
Passive-
.20 —.33 —11 33 30 Ue Sut .16
Aggressive
.39 —.31 —.17 BO so) 22 .20 27
Paranoid
roi -.17 19 DE 54 41 2 ESO
Schizotypal
oS —.34 —.15 02 Be .24 23 16
Avoidant
01 —.12 —.08 —.17 aii 16 .00 —.02
Schizoid
.30 —.25 .04 .30 ESif, AQ 103 .05
Self-defeating
01 .09 —.26 Foo) 21 BZ —.12 —.08
Sadistic
A
Note. Sample of clinical outpatients, n = 64. MAL = Malingering Index; DEF = Defensiveness Index;
CDF = Cashel Discriminant Function; SA Est = estimated Substance Abuse scale scores; TPI =
Treatment Process Index: VPI = Violence Potential Index; SPI = Suicide Potential Index; RDF =
Rogers Discriminant Function.
4S. E. Hyler, R. O. Rieder, J. B. W. Williams, R. L. Spitzler, J. Hendler, & M. Lyons, 1988.

298
 Index Correlations With Clinical, Personality, and Validity Indicators

Table A-8
Index Correlations With Diagnostic Interview for
Personality Disorder? DSM-III-R Personality Disorder Diagnoses
MAL DEF CDF SA Est TPI VPI SPI RDF
Histrionic —.08 43 103 36 NO) .06 -.01 16
Narcissistic .00 01 —.10 41 33 AS 01 ld
Borderline —.12 —.04 —.10 23 .06 = (05 03 alte
Antisocial 42 —.43 —.09 ih 46 .68 37 24
Dependent .09 = 28 A —.19 01 .0O .00 13
Compulsive 24 .30 —.04 16 24 16 24 01
Paranoid 135 = 39 —.18 —.05 30 .20 Aye 14
Schizotypal 24 .0O 14 td .40 18 16 =—08
Avoidant .16 —.31 —.25 —.09 21 .06 a3 .05
Schizoid —.01 -.13 .08 —.10 SUS PA 15 10
nn eee
Note. Sample of clinical outpatients, n = 72. MAL = Malingering Index; DEF = Defensiveness Index;
CDF = Cashel Discriminant Function; SA Est = estimated Substance Abuse scale scores; TPI =
Treatment Process Index; VPI = Violence Potential Index; SPI = Suicide Potential Index; ADS =
Rogers Discriminant Function.
4M. Zanarini, 1987.

299
PAI Interpretive Guide

Table A-9
Index Correlations With Brief Psychiatric
Rating Scale (BPRS)* Clinician Ratings
BPRS ratings MAL DEF CDF SAEst_ TPI VPI SPI RDF

Somatic
concern —.10 —.07 —.12 —.18 .00 —.17 —.19 —.02

Anxiety —.16 —.29 —.09 .20 25 .03 NT —.13

Emotional
withdrawal —.03 .03 —.17 —.18 —.23 —.19 —.04 .08
Conceptual
Disorganization 10 .30 WZ —02 02 .00 =12 —.15
Guilt feelings —.09 —.39 —.20 14 .20 10 20 —.12
Tension 19 16 oe Lif 35 .20 2 —.01
Mannerisms
and posturing .08 sli .00 01 —.01 .02 .05 —.18
Grandiosity JS 39 male: 16 .20 16 .03 —.14
Depressive mood -.01 —.49 01 13 22 .02 16 AG,
Hostility —.09 —.21 Pf 19 .26 —.07 —.12 .07
Suspiciousness .O7 —.20 19 13 Lis =02 —.09 10
Hallucinatory
behavior —.17 —.08 14 —.08 01 —.26 —.33 —.09
Motor retardation 02 —.34 —.19 —.08 .02 —.04 15 .22
Uncooperative —.08 01 .29 LO 16 —.15 —.24 .00
Unusual thought
content .08 18 10 10 .07 01 —.07 WE
Blunted affect 01 —.06 —.14 —.10 —.22 —.10 .02 .23
Excitement —.18 les. 21 —.11 —.02 —.09 —.07 .05
Disorientation ao —.27 —.02 —.16 —.21 —.18 —.17 —.09

Note. Sample of psychiatric inpatients, n = 72. MAL = Malingering Index; DEF = Defensiveness
Index; CDF = Cashel Discriminant Function; SA Est = estimated Substance Abuse scale scores; TPI =
Treatment Process Index; VPI = Violence Potential Index; SPI = Suicide Potential Index; RDF = Rogers
Discriminant Function.
&J. E. Overall & D. R. Gorman, 1962.

300
 Index Correlations With Clinical, Personality, and Validity Indicators

Table A-10
Index Correlations With Indicators of Personality and Environment
MAL DEF CDF SAEst_ TPI VPI SP! RDF

Social
Readjustment
Rating Scale 05 sl7/ 05 32 42 33 B60) e245
Perceived Social
Support Rating
Scale—Friends® —.14 .20 —.01 —.25 —.29 —.27 —.44 —.15
Perceived Social
Support Rating
Scale—Family” OM =15 als} —.21 —13 —.12 —.22 .06
Marlowe-Crowne
Social Desirability
Scale° —.19 .28 D7 ae a AD ieee A0 ~.36 09
Interpersonal
Adjective
Scales—Warmth? -.21 —.03 =,00) gy—-45. ge33) yy = 44 34 a8
Interpersonal
Adjective Scales—
Dominance* .03 53 ts 32 21 ict =19) ce 18
a a a Ne NA ce i ee
Note. Sample of community adults, n = 85. MAL = Malingering Index; DEF = Defensiveness Index;
CDF = Cashel Discriminant Function; SA Est = estimated Substance Abuse scale scores; TPI =
Treatment Process Index; VPI = Violence Potential Index; SP! = Suicide Potential Index; RDF =
Rogers Discriminant Function.
aT_H. Holmes & R. H. Rahe, 1967.°M. E. Procidiano & K. Heller, 1983. °D. P. Crowne & D. Marlowe,
1964. YJ. S. Wiggins, 1995.

301
PAI Interpretive Guide

Table A-11
Index Correlations With Indicators of Clinical Symptomatology
MAL DEF CDF SAEst TPI VPI SPI RDF

Self-report
.03 —.08 .28 81 {a7 .69 35) Pe
Psychopathy Scale
Maudsley Obsessive-
2) —.24 NS —.06 —.06 mili .60 .00
Compulsive Index?
Fear Survey
Schedule® 16 —.14 19 —.13 —.29 -.26 20° °=.11

Beck Hopelessness
Scale® tke Say —.06 01 06— 07 — 9499 rae
State-Trait Anxiety
Inventory—State® —.09 —.32 —.06 —.11 12 -.04 855 139

State-Trait Anxiety
Inventory—Trait' 24 -37 =03 —.05 =16 '=07 2.66 403
State-Trait Anger
Expression Inventory—
State Anger! -07 —35 14 .08 06" le =ae4a se eats
State-Trait Anger
Expression Inventory—
Trait Anger! 07 -.19 18 .38 (02 ee 4G oom meCG
State-Trait Anger
Expression Inventory—
Angry Temperament! .09 .06 31 30 12D ~MAQIERAD WELOO
State-Trait Anger
Expression Inventory—
Angry Reaction‘ 05 -—.16 —12 18 24 24 19 -—.02
State-Trait Anger
Expression Inventory—
Anger In! 19 -.24 —.04 EP O88 2749935 205
State-Trait Anger
Expression Inventory—
Anger Out! —11 -—16 .00 46 Oe eeOc meme
State-Trait Anger
Expression Inventory—
Anger Control ' 34 39 att =19 0G 3 1065 0D eaco
State-Trait Anger
Expression Inventory—
Anger Expression! —09 -.40 —.08 BF 08 5a O eco
Note. Sample of college students, n = 42. MAL = Malingering Index; DEF = Defensiveness Index;
CDF = Cashel Discriminant Function; SA Est = estimated Substance Abuse scale scores; TPI =
Treatment Process Index; VPI = Violence Potential Index; SPI = Suicide Potential Index; RDF =
Rogers Discriminant Function.
@R. D. Hare, 1985, °S. J. Rachman & R. J. Hodgson, 1980. °J. Wolpe & P. Lang, 1964. 9A. T. Beck
& R. A. Steer, 1988. °C. D. Spielberger, 1983. 'C. D. Spielberger, 1988.

302
 Index Correlations With Clinical, Personality, and Validity Indicators

Table A-12
Index Intercorrelations in Clinical and Community Samples
MAL DEF CDF SA Est TPI VPI SPI RDF
MAL — oO a 34 40 42 A7 26
DEF 10 32 -07 .08 04 -18 -.07
CDF 24 sis = 22 AQ 10 27
SA Est Be SY 19 _ 79 61 31
Wiz) 44 SEE 16 78 16 65 .30
VPI 48 —17 19 83 — 65 21
SPI 50 Sele) 13 54 16 AE — .36
RDF ali —.06 19 07 10 .09 ae —_
Note. Values above diagonal from community normative sample, N = 1,000; values below diagonal
from clinical normative sample, N = 1,246. MAL = Malingering Index; DEF = Defensiveness Index;
CDF = Cashel Discriminant Function; SA Est = estimated Substance Abuse scale scores; TPI =
Treatment Process Index; VPI = Violence Potential Index; SPI = Suicide Potential Index; RDF =
Rogers Discriminant Function.

303
 ~

jhe

~ vo Oe
mn mor
Sion? oiuet) 4a) 80D
\ ‘ rin) 23 n*| inendiseiT
iin’ | neat) wreagort
 INDEX

Academic potential, 16
Adjustment disorder, 93, 146, 148, 150-151, 168
Adjustment reactions, 82
Affective disorder, 53, 57, 151, 202 (table), 205 (table)
Agitation, 86, 87, 95, 151 (table), 165, 215, 218 (table)
Alcohol abuse, 53, 75, 78, 85, 91, 102, 190, 280
Alcohol dependence, 4 (table), 15, 25, 72-74, 79, 85, 88, 91, 94, 96, 100, 102, 190,
192-193, 195, 201, 271
Ambivalence, 60, 62, 87, 90 95, 98, 198-199, 272
Anger, 16, 57, 62-64, 68, 84, 86, 87, 89, 92, 93, 98-99, 101, 125, 140, 158, 167, 177, 183,
184.187,188..194 202 (table), 209-215, 218 (table), 219, 225, 226, 237, 239,271,274
Apathy, 38-39, 43, 55, 150, 204
Assertiveness, 4 (table), 7 (table), 18, 140, 212-213, 237, 239, 269, 273
Attachment, 58, 59, 184, 232-234, 272
Autonomy, 237, 243
Avoidant personality, 184-186, 189 (table)

Beck Depression Inventory (BDI), 13, 38, 114, 151, 206, 207 (table), 276, 296
Beck Anxiety Inventory, 13, 296
Beck Hopelessness Scale, 13, 14, 207 (table), 302
Brief Psychiatric Rating Scale (BPRS), 15, 142, 143 (table), 276
Behavior therapy, 252
Bias, 8
Bipolar disorder, 84, 90, 96, 99, 103, 165, 172 (table)
Borderline Personality, 16, 47, 56, 57, 59-65, 86, 87, 89, 90, 93, 94, 96, 98, 101, 102, 113,
173-179, 189 (table), 199, 250, 277

Chronicity,23; 25226541, 975 1499150 6163-165; 1689,195,246, 247,257, 205, 2608

Cognitive therapy, 30, 267, 269 (table), 275
Complementarity, 232-233
Compulsion(s), 155, 186
Conduct disorder, 67, 71
Conversion, 3 (table), 5 (table), 24-28, 163, 164 (table)

Defensiveness, 47, 74, 76, 78, 129-143, 146, 196, 207, 219-220, 221 (table), 251, 253,
292, 293, 294, 295, 296, 297, 298, 299, 300, 301, 302, 303
Delusions, 6 (table), 44-49, 52, 54, 55, 123, 164-165, 172, 184, 270
Demographic, 8, 10-11, 197, 215, 276

305
PAI Interpretive Guide

Denial, 32, 73, (4. 75, 78779) 102,129,130 713) Sis O01 csi20e
Dependency, 31, 58, 85, 87, 90, 93, 98, 163, 174, 177, 1862187). 226)2 3092315 252254,
235,230, 239,242 200m (2
Depression, 2, 3 (table), 5 (table), 9, 13-14, 16, 18, 19, 30, 38-42, 43, 45, 53, 56, 57, 66,
82-8385 80: 00/0192) 03.94. 111-117, 126,129) 13 Viola oe alot
179, 188, 195, 201, 202 (table) 204, 205 (table), 206, 207 (table), 241, 267, Petey Sicilw
Dit oe LOe ino
Depressive personality, 187-188
Detachment, 6 (table), 52-53, 156, 170, 184, 257
Dissociative disorders, 113, 194
Drug abuse, 16, 62, 72, 74, 80, 85, 94, 96, 102, 180, 190, 202, 218 (table)
Drug dependence, 4 (table), 80, 97, 99, 101, 103, 192-193
Dysthymic Disorder, 41, 86, 88, 89, 90, 146, 149-151, 188

Egocentricity, 6 (table), 61, 66 (table), 67-71, 98, 101, 102, 103, 180, 182-183, 234, pe ey
230.23
Ego-syntonic, 175
Environment, 6 (table), 17, 34, 36, 38, 39, 48, 49, 67, 113, 125, 148, 153, 166, 194, 201,
223, 225, 226, 229, 241-247, 252, 265, 268 (table), 301
Exploratory therapy, 252
Extraverted, 235

Family of origin, 59

Generalized anxiety, 30, 115, 118, 158, 160

Guilt, 36, 40, 63, 64, 66, 68, 70, 71, 92, 94, 155-156, 300 (table)

Hamilton Rating Scale for Depression (HAM-D), 13, 38, 276

Helplessiess, 36,09, 95, 149, 155,187, 237, 239,264
Histrionic personality, 183, 188 (table)
Hopelessness, 4 (table), 5 (table), 17, 39-41, 86, 89, 92-93, 179, 202, 203 (table),
205 (table), 274
Hostilitye bolo, 1, 44, 45,49" 64.68. 83,84, 86, 89, 92 05007908 00RD a lo:
183-184, 188, 204, 207 (table), 209-215, 219, 221 (table), 236, 252, 253, 266
Hypersensitive, 48, 50, 89
Hypochondriasis, 14, 24, 25, 27, 155
Hysteria, 24

Idiosyncratic, 106-107, 113

Inhibition, 34, 70, 165, 185, 187, 210
Inpatient(s), 2, 20, 74, 97, 177, 263-269, 277
Interpersonal Adjective Scales (IAS), 15
Introversion, 238

306
 Index

Major Depressive Disorder, 9, 19, 0) 9302,,03,,85, 86588289 £90, 91-94, 112, 115, 146,
149-151
Mania, 4 (table), 6 (table), 15, 42-47, 53, 124, 136, 164, 269
Marlowe-Crowne Social Desirability Scale, 136 (table), 143 (table), 207 (table), 221 (table)
Medical 45 25-29,°85" 161416392700272
Medication(s), 40, 42, 51-54, 73, 84, 85, 170, 179, 249, 253, 263, 266, 269-271, 280
Methadone, 10, 17, 190, 192
Minnesota Multiphasic Personality Inventory (MMPI), 13, 121 (table), 127 (table),
136 (table), 143 (table), 207 (table), 221 (table), 258 (table), 294 (table), 295 (table),
297 (table)
Mood swings, 57, 62, 63, 64, 92
Motivation, 4 (table), 17, 40, 74, 78, 88, 124, 239, 249-264, 277
Motivation to change, 124

Narcissism, 39, 44, 45, 68, 182, 189 (table), 297, 298
Negative affect, 13-14, 31, 57, 161, 204-205, 206 (table), 279
Neurosis, 13-14, 16, 33, 56, 61, 92, 93, 148-150, 168, 180, 184, 198

Obsessive-compulsive, 3 (table), 5 (table), 13, 33, 140, 175, 186, 189 (table)
Obsessive-Compulsive Disorder, 29, 155, 186
Orpanicy 4. 22. 24620. 2 Aig B58), Ui61h05
Organic mental disorders, 83, 85, 195
Outpatient treatment, 111, 114, 168, 170, 177, 263-265, 269, 277, 278

Panic attacks, 151, 154, 202 (table), 205 (table), 271 (table)
Paranoid delusions, 167, 264, 270
Paranoid personality, 4 (table), 15, 47, 183-184, 189 (table)
Passive-aggressive, 27, 49, 187-188, 189 (table), 210, 232
Pharmacotherapy, 252, 253, 269-270, 271 (table), 280
Phobia(s), 3 (table), 5 (table), 29, 34, 36, 37, 38, 69, 153, 154, 158, 160 (table), 185, 269,
270, 271 (table)
Physical abuse, 210
Positive affect, 204-205, 206 (table)
Posttraumatic stress, 13,35, 61, 81, 84, 87, 89. 90, 92, 93, 94, 99. 148 155, 158, 280
Posttraumatic Stress Disorder, 13, 34, 60, 87, 88, 92, 94, 99, 148, 155, 160 (table),
258 (table)
Projection, 7; 49; 71, 93
Psychodynamic treatment, 268-269, 273
Psychological-minded, 252 (table), 256 (table)
Psychopathy, 14, 16, 66 (table), 67, 69, 180, 221 (table), 258 (table)
Psychosis; 6 (able), 15, 15,.42,.47, 52; 54.55, 56, 61, 88, LIS, 123. 125, 126, 136, 164,
167.168.170.172 (table), 185, 218, 264, 269, 274
Psychotherapy, 20, 27, 31, 249, 251, 254, 255, 266, 270, 278-279

307
PAI Interpretive Guide

Recklessness, 6 (table), 31, 34, 60, 69-70, 96, 97, 102-103, 177, 180, 264
Rejection, 4 (table), 18, 59, 60, 62, 64, 87, 90, 93, 97208-175..20% 239, 249; 258,257,
DIA. 2
Reliability, 1.3, lOMLIE SlE27%o,2 76
Rigidity, 5 (table), 33-34, 37, 138, 140, 186, 202 (table), 205 (table), 251, 253, 266, 272,
MAS
Ruminative, 5 (table), 27, 29-31, 34, 36, 41, 62, 82, 85, 89-90, 153

Schizoaffective, 83, 87, 89, 92, 93, 97, 164, 170

Schizoid personality, 53, 184-185, 189 (table)
Schizophrenia, 14, 17, 47, 51-55, 82, 84, 87, 93, 97, 114-116, 118, 164, 166-172, 184-185,
PLONE het Rear ath
Schizotypal personality, 56, 184-185, 189, 258 (table)
Self-effacing, 238, 240
Self-efficacy, 39, 223-224, 229, 231
Self-esteem, 5 (table), 19, 39, 41, 42-47, 69, 86, 88, 111, 124, 133, 140, 148, 149, 165,
POMS 065 41876225231, 289, 240 27 1 ee/9
Self-harm, 6 (table), 60, 86, 89, 92-94, 98, 99, 498, 206, 264
Self-mutilation, 60, 63, 65, 199
Sexual abuse, 35, 59, 177
Sidationalproplems, 13, 80,83, 129, 131,154) [8ayz0s 224, 2412240
Sociability, 13
Social phobia, 153-154, 160 (table), 185
Somatic delusions, 25-26, 82-83, 166, 172, 271 (table)
Somatoform, 24, 27, 82-85, 161-163, 164 (table), 183
Splitting, 60, 64, 113
Standard error of measurement, 11, 81, 278
State-Trait Anger Expression Inventory (STAXI), 17
State-Trait Anxiety Inventory (STAI), 13, 121 (table), 276, 302
Submissiveness, 4 (table), 18, 186-187, 196, 232, 234, 239, 240, 272-273
Suicide, v, 17, 60-61, 149, 197-199, 201-203, 205, 206 (table), 277

Termination of treatment, 254, 259-260

Trauma, 36, 156, 218 (table)
Traumatic stress, 3 (table), 5 (table), 34-35, 155-158, 177, 265, 267
Treatment intensity, 267, 277
Treatment format, 265
Treatment setting, 263

Violence, v, 7 (table), 155, 194, 197, 209, 212, 214-221, 292, 293, 294, 295, 296, 297,
298, 299, 300, 301, 302, 303

308
 About the Author
Dr. Leslie C. Morey is Associate Profes-
sor of Psychology at Vanderbilt Univer-
sity and Visiting Associate Professor of
Psychology at the Harvard Medical
School. He has also held academic
appointments at the Yale University
School of Medicine and the University of
Tulsa. He received his PhD in Clinical
Psychology from the University of
Florida and completed his clinical
internship at the University of Texas
Health Sciences Center at San Antonio.
Dr. Morey is the author of the Personal-
ity Assessment Inventory and has pub-
lished extensively in the area of the
assessment and diagnosis of personality
and mental disorders. In addition to his
general work in the field of assessment,
he is well known for his research into
personality disorders and alcoholism.
He is on the editorial boards of numer-
ous journals in the areas of assessment
and psychopathology.
 ISBN 0-911907-22-X
90000

|| PAR Psychological Assessment Resources, Inc.

9"780911'"907223 mums P.(). Box 998/Odessa, Florida 33556/ Toll-Free 1-800-331-TEST