SECTION

3 :
SINGLE-
CASE
DESIGNS
What You Are Going
To Learn :
1. Use Reversal/Withdrawal designs.

2. Use Multi-element/Alternating
Treatments
designs.

3. Use Multiple Baseline designs.

4. Interpret and make accurate

determinations
when comparing Phase Change Designs.

5. Explain the Dependent Variable and how

it is
used.

6. Explain the Independent Variable and

how it is
used.

7. Explain what defines Functional

3.1. Use Reversal/Withdrawal Designs

Reversal Design
• Repeated measures of behavior in given
setting.

• Requires at least 3 consecutive phases:

– Initial baseline (A)

– Intervention (B)
– Return to baseline (A)

A-B-A-B Reversal
• A-B-A-B preferred over A-B-A as
stronger
demonstration.

•Most powerful within-subject design for

demonstrating a functional relation
between an environmental manipulation
and a behavior.
Operation and Logic of Reversal
Design
• Involves prediction, verification, and
Replication.

•Independent variable is
responsible for behavior change if
repetition of baseline and
treatment phases approximate the
original phases.

Variations of the A-B-A-B

Design
• Repeated reversals
• B-A-B reversal design
• Multiple treatment reversal
designs
• NCR reversal technique
• DRO reversal technique
• DRI/DRA reversal technique

Repeated Reversals
• A-B-A-B-A-B
• Replications present more
convincing
B-A-B Reversal
Design
• Doesn’t enable assessment of effects
on
preintervention level

• Possible sequence effects

• May be appropriate with dangerous

behaviors

• Ethics of withholding effective

treatment

Multiple Treatment Reversal

Designs

•To compare effects of two or more

experimental conditions with each
other or Baseline

• Can make design decisions based

on ongoing
assessment of data

• Vulnerable to sequence effects

• i.e., A-B-A-B-C-B-C, A-B-C-B-C-B-C

NCR Reversal
Technique
• Noncontingent reversal

• Demonstrates effects of contingent

Reinforcement

• Useful when not possible to

eliminate
activity used as contingent
reinforcement

• Deliver NCR on fixed or variable

schedule
independent of the behavior

DRO/DRI/DRA Reversals
• DRO: Deliver reinforcement
following any
behavior other than the target
behavior

• DRI: Reinforcement following

behavior
that’s incompatible with target
behavior

• DRA: Following an alternative

The Appropriateness of
the Reversal Design
• Advantages:

– Clear demonstration of functional

relationship
– Quantifies amount of behavior change
– Shows need to program for
maintenance

• Disadvantages:

– Irreversibility
– Social, educational, and ethical
concerns

Irreversibility
• Reversal design not appropriate when
independent variable cannot be
withdrawn

•Level of behavior from earlier

phase cannot be reproduced
again under the same
conditions

• If suspected, consider DRO or

DRI/DRA
Withdrawing Effective
Interventions
• Social concerns

– Get full support of everyone involved

• Educational and clinical issues

– Reversal phases can be very short

– For ethical reasons, withdrawal of

intervention
may not be appropriate in harmful
situations
Links
:
Reversal Design
Graphing
3.2. Use
Multi-element/Alternating
Treatments Designs

Alternating Treatments Design

• Efficient for comparing effects of 2 or
more
Treatments

• Also known as:

– Multi-element baseline design
– Multiple schedule design
– Concurrent schedule design
– Simultaneous treatment design

Operation and Logic of

Alternating Treatments Design
• Alternated in a variety of ways

• A distinct stimulus is often associated

with each treatment

• Involves prediction, verification, and

replication
Operation and Logic of
Alternating Treatments Design
•Experimental control demonstrated
with different levels of response in
different Treatments

• Allows for quick comparison

• Stress importance of evaluating

individualized treatments

Variations of Alternating
Treatments Design
• Single phase without no-treatment
control
Condition

• Single phase with one no-

treatment
control condition

• Two phase with initial baseline

• Three phase with baseline and

final best
treatment phase
Without No-Treatment Control
Condition
With No-Treatment Control
Condition

Copyright © 2007 by Pearson Education, Inc. All rights reserved

With Baseline and Final Best Treatment
Phase

Cooper, Heron, and Heward Copyright © 2007 by Pearson Education, Inc. All rights
reserved
Applied Behavior Analysis, Second
Edition
Advantages of
Alternating
Treatments Design
• Does not require treatment
withdrawal

• Speed of comparison

• Minimizes irreversibility problem

• Minimizes sequence effects

• Can be used with unstable data

• Can be used to assess generalization

of
Effects

• Intervention can begin immediately

Disadvantages of
Alternating Treatments
Design
• Multiple treatment interference

• Unnatural nature of rapidly

alternating
Treatments

• Limited capacity (max. of 4

conditions)
Articles/
Journals :

Examples Alternating Treatment Designs

Article Attached

1. Escape and avoidance hypothesis testing using an alternating treatment design. Mozzoni
& Hartnedy. Behavioral Interventions. July 2000

Abstract: Persons with acquired brain injuries (ABIs) are frequently confused, disoriented, and
frightened. Many of the procedures used in rehabilitation are experienced by the person with a
brain injury as aversive, frustrating, and intrusive. The therapeutic modalities of PT, OT, speech,
education, and medical procedures are frequently occasions for clients to emit escape and
avoidance behaviors, which interfere with therapy and result in the client redirecting the
therapy, teacher, or medical personnel from their treatment objective(s). Three treatment
methods were compared: (i) earned escape, where the child was able to earn short breaks for
treatment cooperation; (ii) praise for treatment cooperation; (iii) graphic representation of
progress, where the therapist helped the child to graph his performance. An alternating
treatment design (Barlow & Hersen, 1984) was used to compare the effectiveness of three
behavioral contingencies for escape and attention motivated behavior. Results clearly
demonstrated the differential effectiveness of the three as well as a general decrease in target
behaviors across all conditions.

2. Brief report: An extension of the sequential alternating treatment design using reversals
with subjects not available concurrently. Houlihan, Jones, Sloan, Cook. Behavioral
Interventions. 1992

Abstract: This brief report proposes an extension of the traditional Alternating Treatments Design
that would be acceptable to use in hospital and residential settings where subjects are often not
concurrently available. Concerns about the inability of traditional designs to work in these special
situations and the emergence of a “scientist-practitioner spli” are discussed. An example of how the
Extended Alternating Treatments Design might be used is provided

3. Applications of a sequential alternating treatments design. Wacker et al. JABA. 1990

Abstract We propose the use of a combined version of the alternating

treatments and multiple baseline designs in situations in which a traditional
baseline (no treatment) condition either does not provide an adequate
contrast condition or is not feasible or practical due to clinical constraints.
We refer to this design as a sequential alternating treatments design
because two treatments are initially implemented in a random or
counterbalanced fashion and are followed by a sequential change in one or
both treatments across settings, subjects, or tasks. The effects of the
independent variables are assessed first by analyzing the two series of data
points representing the different treatments (relative effects) and then by
assessing changes in one or both series, as application of the alternative
treatment is introduced sequentially. The sequential application of treatment
provides an analysis of control in the same manner as the multiple baseline
design; the initial alternating treatments phase provides a contrast condition
in much the same manner as a baseline condition. Applications of this
design to the assessment of peer training and self‐injurious behavior are
described.
Articles/
Journals :

Links
:
Alternating Treatment
Multi-Element
3.3. Use Multiple Baseline Designs.

Multiple Baseline Design

• Most widely used for evaluating
treatment
effects in ABA

• Highly flexible

• Do not have to withdraw treatment

variable

Operation and Logic

•Alternative to reversal design when
target behavior is likely to be
irreversible or when impractical or
unethical to reverse Conditions

• 3 basic forms:

– Multiple baseline across behaviors

– Multiple baseline across settings

– Multiple baseline across subjects

Prediction, Verification, and
Replication
•Apply independent variable to Behavior 1
when you can confidently predict that the
behavior would remain the same in
constant conditions

•If Behaviors 2 and 3 remain unchanged

after the application of the IV to Behavior
1, this verifies the prediction

• If the IV changes Behavior 2 like it did

Behavior 1,
the effect of the IV has been replicated

• The more replications, the more

convincing the
demonstration; most commonly 3-5 tiers

Multiple Baseline Across

Behaviors
• 2 or more different behaviors of same
subject

• Each subject serves as his/her own

control

•After steady state baseline responding,

independent variable is applied to 1st
behavior, while other behaviors are kept
in baseline Conditions
Multiple Baseline Across
Settings
• A single behavior is targeted in two or
more different settings or conditions

•After steady state baseline

responding, independent variable
is applied to 1st setting, while
other settings are kept in
baseline conditions

•When steady state responding is

reached for 1st setting, then IV is
applied to next setting

Multiple Baseline Across

Subjects
• One target behavior for 2 or more
subjects in the
same setting

•After steady state baseline

responding, independent variable is
applied to 1st subject, while other
subjects are kept in baseline
Conditions

• When steady state responding is

reached for 1st
Variations of Multiple
Baselines
• Alternative tactics for pursuing a
multiple
baseline analysis:

– Multiple probe design

– Delayed multiple baseline design

•When extended baseline

measurement is unnecessary,
impractical, too costly, or
unavailable

Multiple Probe Design

• Analyzes relation between
independent
variable and acquisition of skill
sequences

•Instead of simultaneous baselines,

probes provide basis for
determining if behavior change has
occurred prior to intervention

• Appropriate for analyzing a shaping

program
Delayed Multiple Baseline
Design
•Initial baseline and intervention begin
and subsequent baselines are added in
a delayed or staggered fashion

•Effective when reversal design is not

possible, limited resources preclude a
full-scale design, and when a new
behavior, subject, or setting becomes
available

• Limitations: shorter baselines and can

mask
interdependence of dependent variables
Assumptions and
Guidelines
1. Select independent, yet functionally similar
baselines

• Behaviors are functionally independent of one

another

• Behaviors share enough similarity that they

will change with the application of the same
independent Variable

2. Select concurrent and plausibly related

multiple Baselines

• Behaviors must be measured concurrently

• All relevant variables that influence one

behavior
must have opportunity to influence other
behaviors

3.Do not apply the independent variable

to the next behavior too soon

4.Vary significantly the lengths of

multiple Baselines

• The more baseline phases differ in

length, the
stronger the design

5. Intervene on most stable baseline

Multiple Baseline
Design
Advantages
• Does not require withdrawal of an
effective
Treatment

• Ideal for multiple behavior changes

sought
by many practitioners

• Useful in assessing occurrence of

generalization of behavior change

• Relatively easy to conceptualize

Limitations
• Does not demonstrate experimental
Control

• Provides more information about

effectiveness of treatment
variable than function of target
behavior

•Can require treatment being

withheld for some
behaviors/settings/subjects for a
long time
Changing Criterion
Design
•Used to evaluate effects of a
treatment that is applied in a
graduated fashion to a single
target behavior

• Initial baseline phase followed by

series of
treatment phases that serve as
baseline for increased criterion of
the next phase

Operation and Logic

• Prediction, replication, and
verification

• Variation of the multiple baseline

design

•Show repeated production of new

rates of behavior as function of
manipulations of independent
variable

• Flexibility of the design

Guidelines for Use

•Requires careful manipulation of 3
Length of
Phases
•Each phase serves as baseline for
next phase, so must be long
enough to achieve stable
responding

• Slower to change target

behaviors,
therefore, require longer phases

•Should vary considerably to

increase design’s validity

Magnitude of Criterion
Changes
•Varying size of changes gives more
convincing demonstration of
experimental Control

• Must be large enough to be

detectable,
but not so large as to be
unachievable

• Smaller changes can be used with

very
stable levels of responding
Number of Criterion
Changes
•The more times the behavior
changes to meet new criteria, the
more convincing the demonstration
of experimental control

• Interrelated with phase length and

magnitude of criterion changes

•If limited time for study, the

greater the number of phases, the
shorter each phase can be.

Appropriateness of
Changing Criterion
Design
• Does not require reversal of
improved
Behavior

• Only one target behavior is

required

•Only for use with behaviors that

are already in student’s
repertoire and lend themselves
to stepwise modification
Appropriatenes
s
• Not appropriate for shaping
behaviors

•Best suited for evaluating

effect of instructional
techniques on stepwise
changes in rate, frequency,
accuracy, duration, or latency
of single target behavior
Links
:
Multiple Baseline
Graphing
Multi-element (Alternating
Treatment), and Multiple
Baseline Designs
In this session :

• We will describe two types of experimental

analysis
tactics used by
applied behavior
analysts
- Alternating
Treatment Designs
-Multiple Baseline
Design

• How each design

incorporates the
steady state
strategy

• Discuss variations
between each
design

• Considerations
for using each
design

• Most behaviors
Alternating Treatments
Design :
• Efficient for comparing effects of 2 or more
treatments

• Also known as:

• Multi-element baseline design
• Multiple schedule design
• Concurrent schedule design
• Simultaneous treatment design

Operation and Logic of

Alternating Treatments
Design
• 2 or more treatments IV are rapidly
alternated in a
variety of ways

• Their effect on the DV are measured

•The different interventions are manipulated

independent of the level of responding. One need
not wait for steady responding.

•A distinct stimulus a sign, distinct areas,

different coloured worksheets, verbal
instructions) is often associated with each
treatment
Operation and Logic of
Alternating Treatments
Design

Involves prediction, verification,

and replication –
the three components of steady
state strategy.
Operation and Logic of
Alternating Treatments
Design
•Presence &
Degree of
experimental
control.

•Experimental control is defined

as objective, with evidence that
Treatment A & B produced
predictable and reliably
different levels of Responding.
Operation and Logic of Alternating
Treatments Design
•Involves prediction, verification, and
replication – the three components of steady
state strategy

•Here each successive data point of the treatment

plays all the three roles

a) Provides a basis for the prediction of future

levels of
responding

b)Potential verification of the previous

prediction of performance

c)Provides the opportunity for replication of

previous effects
A Comparison of Prompting Tactics To Establish
Intraverbals in
Children With Autism. Ingvarsson & Hollobaugh.

We compared the efficacy of tact-to-intraverbal (i.e., using

picture prompts) and echoic-to-intraverbal transfer-of-stimulus-
control procedures to establish intraverbal responding in 3 boys
(4 years old) with autism. For all 3 participants, the picture
prompts resulted in fewer trials to criterion, but both
prompting tactics were eventually effective.

Figure 1: The number of correct unprompted intraverbal responses

in training and generalization probes
Operation and Logic of Alternating
Treatments Design
• Experimental control demonstrated with
different
levels of response in different treatments

• Allows for quick comparison

• Stress importance of evaluating

individualized
treatments

Variations of Alternating
Treatments
Design
• Single phase without a no-treatment
control
condition

• Single phase with two or more conditions,

one of
which is a no-treatment control condition

• Two phase with initial baseline

• Three phase with baseline and final best

treatment
phase
With Initial Baseline
Condition Includes two-phase
exp’tal Design.

6 children – 2 teams
5 Word-Spelling test
Advantages of Alternating
Treatments Design

1. Does not require treatment withdrawal

2. Speed of comparison

3. Minimizes irreversibility problem

4. Minimizes sequence effects

5. Can be used with unstable data

6. Can be used to assess generalization of

effects

7. Intervention can begin immediately

1. Does not require treatment

withdrawal
•Effective treatment need not be
withdrawn to demonstrate a
functional Relationship

•Ethical issues arising out of an

withdrawing an effective
intervention can be avoided
2. Speed of
comparison
•The superiority of one treatment over another
can be studied quickly.

•It is a basic experimental tactic used in

functional behavior analysis for this
reason.

•It leaves the researcher with meaningful

data even if the experiment must be
terminated whereas a reversal design or
multiple baseline design must carry on till
a functional relationship can be observed.

3. Minimizes Irreversibility
problem
•Few behaviors do not return to baseline
when the intervention is withdrawn.

•Ex: Learning mands using PECS / Signs

4. Minimizes Sequence
Effects
•Sequence effects pose a threat to internal
validity. Alternating treatments design
minimizes the sequence effect confounds.

•The experimental condition remains in effect

for a long
period producing sequence effects.

•Extremely difficult to control in experiments using

reversal or multiple baseline tactics to compare 2
5. Can be used with Unstable
data
•At times difficulties with stable baselines, Ex:
academics Where repeated opportunities can
result in improved Performances.

•The effects of practice, maturation, task

difficulty will be equally represented in each
treatment condition and will not differentially
effect any one condition.

6. Can be used To Assess

Generalization Effects
•Singh & Winston (1985) is their study
on pica alternated different therapists
to study the effectiveness of the
overcorrection treatment when
presented by different therapists.

7. Intervention can begin

immediately
• The clinical necessity of beginning an
intervention immediately instead of taking
repeated measurement can be practiced by
using an alternating treatment design
without an initial baseline phase.
Disadvantages of Alternating
Treatments Design
1. Multiple treatment interference

2. Unnatural nature of rapidly

alternating treatments

3. Limited capacity (max. of 4

conditions)

4. Selection of treatments – should

be significantly
different from one another

1.Multiple treatment
interference
•Minimizes sequence effects but
raises questions on effects if each
treatment was implemented in
isolation.

•Refers to the confounding effects

of each treatment on participants
behavior.

•This can be overcome by following the

phase with most effective treatment
2. Unnatural Nature of rapidly
Alternating Treatments
•The back and forth switching is not
practiced by Clinicians.

•The detrimental effects of the rapid

alternation of conditions is an empirical
question.

•The rapid switching to identify an

efficacious treatment does outweigh
the participants endurance of an
ineffective treatment.

3. Limited Capacity
•A maximum of 4 different conditions
can be
compared effectively within a single
phase.

•The potentially competing aspects of

each treatment condition should be
counterbalanced during administration
such as time of day, therapists, order of
presentation etc.

•The designs effectiveness can be

reduced if the participant is unable to
discriminate between competing
4. Selection of
Treatments
•Less discrete treatments may impact the
design. Ex: Group size of 4-10-20 vs 6-7-8.

•Some interventions may not produce

important behavior change unless they
are implemented consistently over a
period of time.

•The designs effectiveness can be

reduced if the participant is unable to
discriminate between competing
treatments.
Multiple Baseline
Design :
•Most widely used for evaluating treatment
effects in ABA.

• Highly flexible.

• Do not have to withdraw treatment variable.

Operation and Logic:

•Alternative to reversal design when target
behavior is likely to be irreversible or when
impractical or unethical to reverse conditions.

• 3 basic forms:

• Multiple baseline across behaviors

• Multiple baseline across settings
• Multiple baseline across subjects
Prediction, Verification, and
Replication:
•Apply independent variable to Behavior 1 when you
can confidently predict that the behavior would
remain the same in constant conditions.

•If Behaviors 2 and 3 remain unchanged

after the application of the IV to Behavior
1, this verifies the prediction.

• If the IV changes Behavior 2 like it did

Behavior 1, the
effect of the IV has been replicated.

• The more replications, the more

convincing the
demonstration; most commonly 3-5 tiers.
Multiple Baseline Across
Behaviors
• 2 or more different behaviors of same
participant

• Each subject serves as his/her own control

•After steady state baseline responding,

independent variable is applied to 1st behavior,
while other behaviors are kept in baseline
conditions

• When steady state responding is reached for

1st
behavior, then IV is applied to next behavior
Operant procedures and the Comatose
Patient Boyle & Greer 1983, JABA

Abstract:

-Operant procedures were tested on three

patients who had been in vegetative coma for 6
months, 10 months, and 38 months.

-A discrete trial procedure was used to test

compliance to verbal requests for three
behaviors for each patient, including lateral head
movement or finger movement, eye focus or eye
squeeze, and mouth movement.

-The design for each patient consisted of multiple

baselines across three behaviors with a
withdrawal phase and a reinstatement phase
(ABAB) for one behavior.
Baseline phases were followed by contingent
music phases in which 15-second taped
excerpts of patients' preferred music
sounded immediately following the emission
of targeted behaviors.

-The contingent music treatment affected all

three behaviors for Patient 1,but was less
effective for two out of three behaviors for the
second and third patients who had been in
coma for more extensive periods of time.

-Results are discussed in terms of the potential

contributions of the use of operant assessment
Multiple Baseline Across
Settings:
• A single behavior is targeted in two or more
different
settings or conditions.

•After steady state baseline responding,

independent variable is applied to 1st setting,
while other settings are kept in baseline
conditions.

• When steady state responding is reached for

1st setting,
then IV is applied to next setting.
INCREASING THE VARIETY OF FOODS
CONSUMED BY A PICKY EATER:
GENERALIZATION OF EFFECTS ACROSS
CAREGIVERS AND SETTINGS:
Hildur Valdimarsdóttir, Yr Halldórsdóttir, Zuilma
Gabriela
Sigurådóttir, JABA , 2013

•A multiple baseline across settings was used to

evaluate the effects of differential reinforcement
of alternative behavior, nonremoval of the fork
(Hoch, Babbitt, Coe, Krell, & Hackbert, 1994),
and stimulus fading on consumption of food
rejected previously.

•The study was conducted in two separate

settings, and caregivers were trained in the
intervention technique to increase
generalization to natural settings. Food variety
Abstract:

The Impact of Verbal prompts on Child

Safety Belt Use in Shopping Carts. Baker,
Bailey, Lee. (2004) JABA

-Each year thousands of children are injured by

falling from shopping carts. Buckling children
into the seats of shopping carts could prevent
many of these injuries.

- A combined reversal and multiple baseline

across settings design was used to evaluate the
impact of verbal prompts on shopping cart
safety-belt use in two stores.

-Safety-belt use increased following

implementation of the verbal prompt. This
study extends the literature on the efficacy of
verbal prompting in the promotion of safety-
Multiple Baseline Across Subjects:
• One target behavior for 2 or more subjects
in the
same setting.

•After steady state baseline responding,

independent variable is applied to 1st
subject, while other subjects are kept in
baseline conditions.

• When steady state responding is reached for

1st
subject, then IV is applied to next subject.

• Most widely used multiple baseline design.

Effects of Video Modeling On Social
Initiations By CWA Nikolopolous &
Keenan. (2004) JABA

Abstract:

-We examined the effects of a video modeling

intervention on social initiation and play
behaviors with 3 children with autism using a
multiple baseline across subjects design.

-Each child watched a videotape showing a

typically developing peer, and the
experimenter engaged in a simple social
interactive play using one toy.

-For all children, social initiationand reciprocal

play skills were enhanced, and these effects
were maintained at 1- and 3-month follow-up
periods.
Evaluation of Behavioral Skills Training To
Prevent Gun Play in Children. Miltenberger
et al. (2004).
JABA

-This study evaluated behavioral skills training

(BST), in a multiple baseline across subjects
design, for teaching firearm safety skills to 6 6-
and 7-year-old children.

-Similar to previous research with 4- and 5-year-

olds, half of the children acquired the safety
skills following BST and half acquired the skills
following BST plus in situ training.
Variations of Multiple
Baselines
• Alternative tactics for pursuing a multiple
baseline
analysis:

• Multiple probe design.

• Delayed multiple baseline design.

• When extended baseline measurement is

unnecessary,
impractical, too costly, or unavailable.

Multiple Probe Design

• Analyzes relation between independent
variable and
acquisition of skill sequences.

•Instead of simultaneous baselines, probes

provide basis for determining if behavior
change has occurred prior to intervention.

• Appropriate for analyzing a shaping program

Graphic prototype of a multiple probe design. Square data
points represent results of probe sessions in which the entire
sequence or set of behaviors (1–4) are tested.

A multiple probe design showing the percentage of correct

responses for each trial on each component of a laundry task
by a young adult male with mental retardation. Heavy vertical
lines on the horizontal axis represent successive training
sessions; lighter and shorter vertical lines indicate trials
within a session.
Delayed Multiple Baseline
Design
• Initial baseline and intervention begin and
subsequent
baselines are added in a delayed or staggered
fashion

•Effective when reversal design is not possible,

limited resources preclude a full-scale design,
and when a new behavior, subject, or setting
becomes available

• Limitations: shorter baselines and can mask

interdependence of dependent variables
Assumptions and
Guidelines
1. Select independent, yet functionally similar
baselines:

• Behaviors are functionally independent of one

another.

• Behaviors share enough similarity that they will

change
with the application of the same independent
variable.

2. Select concurrent and plausibly related

multiple
baselines:

• Behaviors must be measured concurrently.

• All relevant variables that influence one

behavior must
have opportunity to influence other behaviors.

3.Do not apply the independent variable to

the next behavior too soon.

4. Vary significantly the lengths of multiple

baselines:

• The more baseline phases differ in length,

the stronger
the design.
Multiple Baseline
Design
Advantages:
• Does not require withdrawal of an effective
treatment

• Ideal for multiple behavior changes sought by

many
practitioners

• Useful in assessing occurrence of

generalization of
behavior change

• Relatively easy to conceptualize

Limitations:
• Does not demonstrate experimental
control

• Provides more information about

effectiveness of
treatment variable than function of target
behavior

• Can require treatment being withheld for

some
behaviors/settings/subjects for a long time

• Required time and resources

3.4. Interpret and make accurate
determinations when comparing
Phase Change Designs.

1. Phase line (Condition change

line)

Figure 1. Comparison of number of words produced between

Baseline and Intervention.
Phase Change
Line
•Phase change line (or condition
change line) indicates change in
intervention, treatment or
experimental condition.

• Phase change line allows the

identification of
condition in effect during each period.

• It is represented by either solid

vertical line or
dashed vertical line along the
horizontal axis, X axis.

Figure 1. Comparison of number of words produced

between Baseline and Intervention.
Figure 1. Percentage of correct trials identifying alphabets during the
first 20 sessions.
2. Phase label (Condition change
label)
Phase Change Label
Phase change labels and condition
change labels are placed at the top of
the chart to identify
the phase of condition in effect during
each period of the intervention of
experiment.

Figure 1.Percentage ofcorrecttrials identifying alphabets

during thefirst 20sessions.
3.5. Explain the Dependent
Variable and how it is used.

DEPENDENT VARIABLE :
• Dependent Variable (aka: Target Behavior)

• The target behavior of interest, being

measured in an
experiment.

•Some experiments have more than 1 DV

measured.

Reasons for multiple DVs:

-Identifying data patterns that are

treated as a control condition against which
you evaluate and replicate the outcomes of
an IV.

-Determining if there are any Collateral

Effects = when an IV effects behaviors other
than the target behavior.

-Assessing for confounding variables (e.g.

when instead of the IV, another individuals
behavior change impacts the target
individuals behavior change)
3.6. Explain the Independent
Variable and how it is used.

INDEPENDENTVARIABLE :

• Independent Variable (aka: Treatment;

Intervention; Experimental Variable)

•The particular aspect of the

environment that is manipulated to assess
whether it effects the target behavior.

• The treatment/intervention

•Example: Your IV = a DRA procedure to

target hitting aggressive behavior (the
DV)
CONFOUNDING
VARIABLES :
•Confounding Variables (aka :
Extraneous Variables; Unrelated
Variables)

•The terms , confounding variables and

extraneous variables, are often used
synonymously, but they also have a
distinction.

•Both refer to when unknown or unintended

variables have uncontrolled influence on
your study (you should try to control these)

•Extraneous Variables = any aspect of the

environment that must be held constant to
prevent unplanned environmental
variation, usually something the
experimenter is aware of and for which
he/she does his/her best to plan .

•Any uncontrolled factor known or

suspected to exert influence on the
dependent variable that is unexpected
and not considered in planning.

•Example : You passed your driver’s license

exam! The day before, your dad happened to
get a ticket for not using his blinker. The
question happened to come up on your exam
the next day (random confounding variable
that helped you pass)
3.7. Explain what defines
Functional Relationships.

FUNCTIONAL RELATIONSHIPS :

• Experimental Control is also known as Functional

Relationship.

• Experimental Control :- Two meanings :

a)the outcome of an experiment that

demonstrates convincingly a functional relation,
meaning that experimental control is achieved
when a predictable change in behavior ( the
dependent variable) can be reliably produced by
manipulating a specific aspect of the environment
( the independent variable); and

b)the extent to which a researcher maintains

precise control of the independent variable by
presenting it, withdrawing it, and/or varying its
value, and also by eliminating or holding
constant all confounding and extraneous
variables.
 EXPERIMENTAL
DESIGN
Outline:

•Goal of science

•Assumptions and attitudes shared by

scientists in all fields

•Defining features and assumptions about

behavior that guide the work of behavior
analysts

•Why applied behavior analysts conduct

experiments

•Components of ABA experiments

What Is Science?

•Systematic approach for seeking and

organizing knowledge about the natural
world.

•Goal/Purpose
- a thorough understanding of
phenomena
under study.
- discover natural facts and universal
Levels of Scientific
Understanding:
Different types of scientific
investigations contribute to three
types or levels of understanding:

1.description
2. prediction
3. control

 Each level contributes to the

overall knowledge
base in a given field of science.

1.Description - objective facts

derived from
observational studies:

•can be quantified, classified, &

examined for possible relations with
other known facts.

•descriptive studies often lead to

hypotheses or questions for additional
research.

2. Prediction - repeated observations

reveal a consistent correlation between
two events:

• when one event occurs another

event occurs (or doesnʼt) with
some known Probability.
3. Control - based on experimental
demonstrations of a functional relation
between events:

•A functional relation is evident when the

systematic manipulation of one event
reliably produces a predictiable change
in the other event and the change in the
dependent variable was unlikely to be
the result of extraneous factors.

•used to develop technologies (e.g.,

medicine, communications,
transportation, . . . socially significant
behavior change!)

How Science Gets Done:

Instruments improve our contact with

behavior and the variables of which it is
a function.

•There is no standard “scientific method”

•Scientists in all fields share

- an assumption about the nature

of the universe and
- a set of attitudes that guide their
practice

•This assumption and attitudes guide

Attitudes of
Science:
•Determinism - scientists presume the
universe is lawful and orderly.

- natural events do not occur in a

completely random, chaotic way.

- natural events are related to one

another
in systematic ways.

•Empiricism - scientists conduct

objective, systematic observations of
the phenomena of interest. Empirical
results (i.e., data) are …

- independent of the beliefs,

prejudices, and
opinions of the scientist or others.

- available for inspection and

interpretation
by others.

•Experimentation – scientists conduct

experiments. An experiment is a …

- controlled comparison of some

measure of the phenomenon of
•Replication - scientists repeat
experiments to

- confirm previous findings and

discover errors.

- determine the reliability and

believability of findings.

•Parsimony - scientists prefer simple,

logical explanations over complex or
abstract explanations. Parsimonious
interpretations …

- contain only those elements necessary

and
sufficient to explain the phenomenon.

- help scientists fit findings

within the existing knowledge
base.

• Philosophic doubt – scientists

continually question the
truthfulness of current knowledge.

•This healthy skepticism requires

scientists
to …

- regard all theory and knowledge

as
tentative.
Other Ways of
Knowing:
•Contemplation

•Common sense

•Logic authority

•Folklore

•Religious beliefs

•Political
campaigns

•Testimonials

•Pseudo-science
 Defining Features and
Assumptions about Behavior
that guide Research and
Practice in Behavior Analysis
How ABA Science Gets Done:
•ABA is a science devoted to
understanding and improving human
behavior.

•Research methods in ABA are guided

by how behavior analysts define their
subject matter (i.e., behavior) and the
working assumptions they hold about
it.

Defining Features of
Behavior:

•Behavior is an individual phenomenon.

•Behavior analysts define the subject

matter of their science as the
interaction between an organism and
its environment.

•Therefore, experimental method in

ABA focus
on within-subject analyses of behavior
environment
relations.
•Behavior is a dynamic, continuous
phenomenon.

•Behavior occurs in an ongoing, unbroken

stream.

- you can't hold it still

- a dead person canʼt do it

•Therefore, repeated measurement

over time is a hallmark of ABA
research and practice.

Working Assumptions About

Behavior:
•Behavior is determined

-like all natural phenomena,

behavior is functionally related to
other events.

•Therefore, functional relations

between environmental events and
behavioral improvements is the
primary goal of ABA experiments.
•Behavioral variability is extrinsic to the
organism.

•Behavioral variability observed in an

ABA experiment is considered a
function of one or a combination of
environmental factors:

- the independent variable under

investigation.

- uncontrolled factor(s) within the

experiment.

- uncontrolled or unknown events

outside of the
experiment.

•ABA researchers seek to bring behavior

change under experimental control of
the independent variable.

Role of Measurement in ABA:

•ABA is a scientific approach for

- discovering environmental variables

that reliably influence socially
significant behavior (i.e, improve
peopleʼs lives), and

- developing a technology of behavior

change
Why Practitioners Should
Measure Behavior:

Direct and frequent measurement enables

practitioners to detect their successes and,
equally important, their failures so they can make
changes to change failure to success. - Cooper,
Heron, & Heward (2007)

Measurement Defined: Johnston &

Pennypacker (1993)

•the process of assigning numbers and units to

particular features of objects or events . .

- It] involves attaching a number

representing the observed extent of a
dimensional property to an appropriate
unit.

- The number and the unit together

constitute the measure(e.g.,15 cm,50
kg,30 sec)

Dimensional Quantities of
Behavior:

•Repeatability Instances of a response class

can occur repeatedly through time (i.e.,
behavior can be counted)

•Temporal extent Each instance of behavior

Definitional
Measures:
•Topography - the physical form or shape of the
behavior can be measured (and modified).

•Magnitude - the force or intensity with which a

response is emitted can be measured (and
modified).

Measurable Dimensions of Behavior:

•Rate/Frequency - how many responses
occur in a given amount of time, usually
measured in number of responses per
minute.

• Duration - how long a response lasts.

•Latency - how much time elapses

between a stimulus change and onset
of a response.

•Topography -the correctness or quality

of the response.

• Magnitude - the strength or intensity

of the
Dimensions of Behavior Response
Card:
Which Dimension?
(Rate/Frequency, Duration,
Latency, Topography, or
Magnitude?)

Mrs. Gibson asks her fourth graders to get

their science books out and turn to page 48.
Mandy responds by getting out of her seat,
sharpening her pencil, throwing some paper
away in the trashcan, and talking to another
student on the way back to her seat. Mrs.
Gibson says, "Mandy, I told you to get your
science book out." Mandy replies, “Iam!"

Which Dimension?
(Rate, Accuracy, Duration, Latency,
or
Magnitude?)

•Mandy did not contact

reinforcement for getting out
her science book because her:

- latency was too long

- 1, 2, 8
Which Dimension?
(Rate, Accuracy, Duration, Latency,
or Magnitude?)

•Shobana complains to her parents that she

often must stay in the classroom and work
on math problems during morning recess.
When Shobanaʼs parents asked the teacher
about this they were told that their daughter
is a well-behaved, pleasant student who
answers problems accurately. But because
she does not finish all assigned problems in
the allotted time, Shobana has to finish the
problems during recess.

Which Dimension?
(Rate, Accuracy, Duration, Latency,
or Magnitude?)

• Shobana often misses morning recess

Because her:

- rate of answering math

problems is too slow

- 1, 5, 7
Which Dimension?
(Rate, Accuracy, Duration, Latency,
or
Magnitude?)

•Ana is working part-time at a t-shirt

store. His job involves sealing logos and
designs on t-shirts with a hot pressing
machine. To operate the machine, a lever
must be pulled down with significant
force and then lifted back up. Anaʼs
supervisor has complained that some
customers are returning the t-shirts
because the logos are falling off.

Which Dimension?
(Rate, Accuracy, Duration, Latency,
or Magnitude?)

•Customers are complaining because

Anaʼs –

- magnitude of lever pulling is too low

- 4, 7, 10
Which Dimension?
(Rate, Accuracy, Duration, Latency,
or
Magnitude?)

Neil is part of a four-person

assembly-line team at the Cracker
Packaging Company. The other
members of the team are
complaining that Neil is not getting
his items packaged fast enough,
causingthe team to miss out on
bonus pay for meeting the
production quota.

Which Dimension?
(Rate, Accuracy, Duration,
Latency, or
Magnitude?)

Neilʼs team is losing pay for meeting

production quotas because his

•rate/frequency of packaging
items is too slow

• 1, 3, 9
Which Dimension?
(Rate, Accuracy, Duration, Latency,
or
Magnitude?)

Ian has been learning how to wait

for his turn. He does all right in
fast moving activities and games,
but whenever the time between
turns is more than 30 seconds, Ian
begins talking out and grabbing
for the materials being used by
another child.

Which Dimension?
(Rate, Accuracy, Duration,
Latency, or
Magnitude?)

Ian has a problem taking turns

because the –

•duration that he waits is too

short

•2, 3, 5
Which Dimension?
(Rate, Accuracy, Duration, Latency,
or
Magnitude?)

Gita is a high school student with autism.

Her academic work is improving, but she
has few friends due to poor social skills. The
school counselor is teaching Gita how to
engage in informal conversations with
others. Gita has learned to start
conversations, and her classmates initially
respond to her openers. But when it is Gita's
turn to talk, she talks without stopping for a
very long time. Gita is not making any new
friends and her classmates are avoiding her.

Which Dimension?
(Rate, Accuracy, Duration, Latency,
or
Magnitude?)

Gitaʼs classmates may be avoiding her

because the

•duration of her conversational talk is

too long

• 7, 7, 10
Which Dimension?
(Rate, Accuracy, Duration, Latency, or
Magnitude?)

Raviʼs teacher is trying to help him

learn to initiate conversations with
peers during recess and lunch break
by saying, "Hi guys, how's it going?”
But Ravi goes up to the group and
immediately begins telling them
what he saw on TV the night before.
His peers usually ignore him or walk
away.

Which Dimension?
(Rate, Accuracy, Duration,
Latency, or
Magnitude?)

Raviʼs problem starting

conversations
may be because of the

•poor topography of his conversation openers--he

says the wrong things

• 2, 4, 8
Measurement: Necessary
but Not Sufficient
•Measurement provides evidence of
behavior change, but measurement
alone cannot reveal why the change
occurred.

•Applied behavior analysts conduct

experiments to determine the factors
responsible for behavior change.

•This is the analysis in ABA.

Components of ABA
Experiments:

1. Subject/participant

2. Behavior

3. Setting

4. Measurement system and ongoing

visual analysis

5. Intervention/treatment

6. Manipulation of the independent

variable

7. Experimental question
1. Subject/Participant

•at least 1 participant - but “single-subject”

design is a
misnomer

•each subject serves as his or her own control

2. Behavior - Dependent variable

•The target behavior in an ABA experiment or

•a measurable dimensional quantity

of that behavior (e.g. rate, duration) is
called the dependent variable

•Behavior - the dependent variable in

an ABA study must be . .

- a socially significant behavior

- a genuine behavior (could a dead

man do it?)

- the behavior of interest, not a related

behavior that serves as a proxy for
the “real thing”

•Behavior - some studies measure

multiple
dependent variables to . .

- provide data patterns that serve

as controls for replicating the effects
•Behavior - some studies measure
multiple dependent variables to . .

- … provide data patterns that serve

as controls
for replicating the effects of an
intervention

- … assess treatment effects on

multiple
behaviors
•Behavior - some studies measure
multiple dependent variables to . .

-.. provide data patterns that serve as

controls for replicating the effects of an
intervention

- .. assess treatment effects on multiple

Behaviors

-..which can lead to secondary

dependent variables

•Behavior - some studies measure

multiple dependent variables to . .

- .. determine whether changes in the

behavior of a person other than the
subject occur during the experiment
can explain changes in the subjectʼs
3. Setting – the environment is always

•present – no such thing as out-of-

context Behavior

•relevant – environmental events

select and maintain behavior

•“Control the environment and you

will see
order in behavior.” – B. F. Skinner
(1967)

4. Measurement System and

Ongoing Visual
Analysis

•repeated measurement reveals

behavior
change overtime
Making Sense of the
Data:

Percent correct responses:

80, 82, 78, 85, 80, 90, 85, 85, 90, 92

Responses per
minute on successive
school days:
65, 72, 63, 60, 55, 68,
71, 65, 65, 62, 70, 75,
79, 63, 60
Percent correct responses:
80, 82, 78, 85, 80, 90, 85, 85, 90, 92

Responses per
minute on successive
school days:
65, 72, 63, 60, 55, 68,
71, 65, 65, 62, 70, 75,
79, 63, 60
•What do these numbers tell you?

•How long did it take you you reach those

conclusions?

•How sure are you?

•Would others reach same conclusions?

•What if these data sets included many more

measures?

•How could you effectively communicate these

data to
•others?
Graphs:

•visual formats for organizing and

displaying data clearly and concisely.

•provide a simple format for

communicating student
progress to students, parents, and
administrators.

•enable visual analysis of behavior change

over time as a
function of treatment.
5. Intervention/Treatment – The
independent variable in an ABA study

•the particular aspect of the

environment the experimenter
manipulates to find out whether it
affects the subjectʼs behavior

•also called the intervention or

treatment

•as it is manipulated by the

researcher

•must meet the applied dimension

of ABA

Underline the Dependent

Variable and
Circle the Independent Variable:

•The effects of study cards on the

percentage of
homework completion.

•Effects of self-monitoring and self-

graphing on studentsʼ task engagement.

•Number of words spelled correctly by

students immediately following a
spelling quiz review.

•Use of an electronic signal device during

Experimental Control:
The Path to and Goal of
Behavior Analysis:
Experimental control (defined):

A predictable change in behavior (dependent

variable) can be reliably produced by the
systematic manipulation of some aspect of the
person’s environment (independent variable)

Experimental analysis (defined):

•Experimentally determining the effects of

environmental manipulation on behaviour and
demonstrating that those effects can be reliably
produced -

- Can be achieved when a reliable functional

relation between behaviour and some
specified aspect of the environment has
been demonstrated convincingly.

Internal validity:

•The extent to which an experiment shows

convincingly that changes in behaviour are a
function of the independent variable and not the
result of uncontrolled or unknown variables.

•Studies without a high degree of internal validity

yield no meaningful statements about functional
relations lack generality.
Internal Validity: To what extent can the
intervention account for the changes or
results.

External Validity: To what extent can the

result be generalized or extended to
people settings, times, measures or
outcomes.

Threats To Internal Validity:

•History: Any event which occurs

simultaneously which
can affect the results

•Maturation: Changes over time in subjects

•Instrumentation: Changes in measurement

due to Observers

•Testing: Changes due to repeated testing

Threats To External Validity:

•Generality across subjects: extent to which

results can
be extended to subjects, clients.

•Generality across settings:

•Generality across time:

•Generality across change agents:

When planning an experiment investigators
must look
out for threats to internal validity.

Uncontrolled variables which influence

the Dependent variable are also called
Confounding variables.

Example: effect of guided

notes on students quiz
results.
Experimental Control:
The Path to and Goal of
Behaviour Analysis:

“ the goal of experimental design is

to eliminate as many uncontrolled
variables as possible and to hold
constant the influence of all other
variables except the independent
variable, which is purposefully
manipulated to determine its
effects”
Single Subject Experimental
Design & Baseline Logic

Components of Experiments in ABA:

• At least one

– Subject or participant
– Behaviour (dependent variable)
– Setting
–Treatment or intervention condition
(independent variable)

Research Question:

•We conduct an experiment to find out

something we don’t know.
(Murray Sidman (1960, p.214)

• Research question
– “a brief but specific statement of what
the researcher wants to learn from
conducting the experiment”
(Johnston & Pennypacker, 1993b,
p.366)
Most research questions are one the four types:

• Demonstration – will the intervention work?

•Parametric – does more or less of Intervention

work better

•Component – how effective if various

components are added or subtracted?

•Compound – does one intervention work

better than another

Components of Experiments in ABA

(continued)

• A system for measuring the behaviour and

ongoing analysis of the data

•Manipulations of the independent

variable so that its effects on the
dependent variable, if any, can be
detected
– Experimental design

• Participant
–In ABA experiments are referred to as
singlesubject or single-case designs.

– It does not mean experiments are conducted

on
only one participant.
– The experimental logic is:

•To detect and analyse functional relations

between environmental variables and
behavior change that employs each
participant as her own control.

• Participant

-For this repeated measures are conducted

- And the participant is exposed to the

presence
or absence of the IV.

– Most studies have more than one

participant, each subjects data are
graphed and analysed separately

• Behaviour(s)

Are the measureable dimensional

quantity (eg. Rate, duration) and are
the Dependent variable(s)

• In some studies multiple behaviors

are
measured.
Reasons for multiple dependent measures:

–Provide data patterns that can serve as

controls for evaluating & replicating the
effects of an independent Variable

–Assess the presence and the extent of the

independent variable’s effects on behaviours
other than the response class to which it was
directly applied

–Determine whether changes in the behaviour of a

person other than the subject occur during the
course of an experiment & if such changes can
explain changes in the subject’s behaviour

• Setting -

“Control the environment and you will see order

in
behaviour.”
(Skinner, 1967, p. 399)

• Control two sets of environmental variables to

demonstrate experimental control -

– Independent variable

• Presenting, withdrawing, or varying its value

– Extraneous variables

• Prevent unplanned environmental variation

Measurement System and
Ongoing Visual Analysis:

• Observation & recording

procedures must be
conducted in a standardized
manner

• Standardization involves every

aspect of the
measurement system

– Definition of the target

behavior to scheduling of
observations

•Behavior analysts must develop skills in

the detection of changes in the level,
trend, and degree of variability in
behavioral data

Intervention or Treatment:
Independent
Variable:

• Independent variable (defined)

–The particular aspect of the environment

that the experimenter manipulates to find
out whether it affects the subject’s
behavior

–The researcher controls or manipulates

Experimental Design Defined:

• The particular arrangement of conditions in

a study
so that meaningful comparisons of the
effects of the presence and absence of
the independent variable can be made

Experimental design:

• Nonparametric study
– Independent variable is either present or
absent during a time period or phase of
the study

• Parametric study
– The value of the independent variable is
Manipulated

– Seeks to discover the differential effects

of a
range of values

Fundamental Rule:

• Change only one variable at a time

– Experimenter can attribute any

measured
changes to a specific independent variable

–If investigating the effects of a

“treatment package”
Some Additional Rules:

• Do not get locked into textbook “designs”

–Often require a priori assumptions

about the nature of the functional
relations one seeks to Investigate

–May be insensitive to unanticipated

changes in Behaviour

• Select & combine experimental tactics

that
best fit the research questions

Steady State Strategy & Baseline

Logic:

•“A pattern of responding that exhibits

relatively little variation in its
measured dimensional quantities over
a period of time” (Johnston &
Pennypacker, 1993a, p. 199)

• Provides the basis for baseline logic

• Steady state strategy

– Repeated exposure of a given subject to

a given condition while trying to
eliminate or control extraneous
influences on behaviour & obtaining a
 Steady State Strategy & Baseline Logic:

• Baseline logic entails three elements

–Prediction
–Verification
–Replication
Nature & Function of Baseline Data:

• Serves as a control condition

• Does not imply the absence of intervention

– Absence of a specific independent variable

• Why?

– To establish a baseline level of

responding to use the subject’s
performance, in the absence of the
independent variable, as an objective
basis for detecting change

• Applied benefits of establishing a

baseline
level of responding

–To obtain descriptions of antecedent-behaviour

consequent correlations for the planning of an
effective treatment

–Valuable guidance in setting initial

criteria for reinforcement

–Baseline data may reveal the behavior

Prediction
:
“the anticipated outcome of a
presently known or future
measurement. It is the most
elegant use of quantification upon
which validation of all scientific
and technological activity rests.”
(Johnston & Pennypacker, 1980)
Affirmation of the Consequent:

• The predictive power of steady state responding

enables
the behavior analyst to employ a form of
inductive logic known as affirmation of the
consequent (Johnston & Pennypacker,
1980)

• Affirmation of the consequent

–Inductive logic

•“if the independent variable were not applied,

the behaviour, as indicated by the baseline data
path, would not change

• If-A-then-B statement:

1.If A is true B is true

2. B is found to be true
3. Therefore A is true
Verification:

•Verification of a previously predicted level of

baseline responding by termination or
withdrawal of the treatment variable.

Replication:

“Replication is the essence of believability” (Baer,

Wolf, & Risley, 1968)
•Replication of the experimental
effect accomplished by
reintroducing the treatment
variable.
Component and Parametric
Analysis:

Fundamental rule:

Sometimes multiple Ivs are part of an

experiment

• Change only one variable at a time

• Experimenter can attribute any measured

changes to a
specific independent variable

• If investigating the effects of a “treatment

package”

• Ensure that the entire package is presented or

withdrawn each time a manipulation occurs

• Using a “ package of Ivs”

• Entire package added or withdrawn – so it is

the
“package “ that’s being evaluated

• Component analysis will be needed to assess

relative
contribution of different combinations of a
package
Experimental Design:

• Presence, absence or different values of IV –

effects on DV
• IV can be
• Introduced
• Withdrawn
• Increased or decreased
• Combined across bx/ settings/ subjects

Child cries at home:

• Mom gives a hug and switches on TV

• Dad comes over and gives some medicine for
acidity
•Grandpa gives some water and cookies as it is
known to be effective in dealing with ‘heartburn’

Experimental design:

• Nonparametric study
• Independent variable is either presented or
absent during
a time period or
phase of the study
• Parametric study
• The value of the independent variable is
manipulated
• Seeks to discover the differential effects of a
range of
values
Lerman etal., ( 2002) Why parametric study?

•If Reinforcement is too small or too large or too

delayed or too intermittent – response may occur
intermittently – Mazur ( 1998)

•SR+ parameters affect extent to which

alternative bx persists during extinction
( resistance to extinction) – Lerman and Iwata
( 1998)

• Especially important in the context of DRA

Lerman etal ( 2002)

•Experiment I – Two reinforcement magnitudes

on resistance to extinction examined by
reinforcing and extinguishing communication
responses after these were acquired as part of
treatment

•Experiemnt 2 – Evaluated the effects of SR+

magnitudes on characteristics of responding –
response rate and post reinforcement pause
during maintenance of communication response.
Experiment I – Reinforcement
magnitude and resistance to extinction

• Timmy and Gary – Demand condition +

communication card

• Continuous instructions + graded 3 step

prompting
procedure

• 20 sec rbeak contingent on bx occurrence

•Rachel – Tangible condition + comm card – 20 sec

access contingent on occurrence of problem bx

•Initial comm training – 10 min sessions, problem

bx on extinction, graduated prompt procedure to
teach comm response if Participant did not
initiate a comm response within 5 secs or
engage in problem bx

• Prompt delay gradually reduced from 5 sec to 1

sec

• Particpants:

Timmy 4 YO boy with MR

Rachel 20 Yo profound MR – SIB and aggression
Gary 10Yo with MR and autism
• Response
• Touching a communication card for all
participants
•Disruption – ejection of task materials/ Aggression -
Experiment 1:

• Reinforcement schedules thinned systematically

• Initial training phase concluded when terminal

reinforcement schedule was reached

• Response maintenance phase Initiated

• Alternative bx on terminal SR+ sched and Problem

bx on
extinction

•Two different maintenance conditions conducted

with each participant :: Small magnitude = 20 sec
access, large magnitude = 60 sec access

• Each session stopped after SR+ contacted 5 times

• Extinction condition:

• Neither problem bx nor alternative bx contacted

SR+

• No programmed consequence for either

• Continued until alternative response dropped and

was
stable at less than 20%
General Discussion:

Relation between reinforcement magnitude and

responding during DRA- will only minimally
influence resistance to extinction.

Authors go on to discuss further about the number

of
possible confounds.

Further research needed on conditions that

produce a reliable relation between
reinforcement magnitude and responding.

Experiment 2 – Magnitude and responding

during
maintenance:

• Across conditions Alternative bx reinforced on

terminal
VI schedule, problem bx on extinction

• Session ended when alternative bx was

reinforced 5 times

•3 different maintenance conditions for each

participant- small magnitude 20sec, medium
magnitude 60 s and large magnitude 300 secs
( remember response rates were similar under 20
and 60 secs in experiment 1)

• Reversal design used

Response Blocking – Punishment or
Extinction at work?

• Lerman and Iwata (1996) – Hand mouthing

of Paul

• When all responses are blocked extinction

or punishment
could be in effect ?

•Comparison made under conditions of different

proportions of response blocking ( Bl, 100%, Bl,
50%, 25%, 50%, 67%, 75%, 100%)

• In FR1 , no extinction burst was observed

• In FR1 , rapid decrease in bx observed

•If extinction were in operation, blocking some

but not all responses should maintain or increase
bx ( intermittent reinforcement) – did not happen

• As punishment sched became denser, bx reduced

faster

• Conclusion – Punishment at work and not

extinction
Context:

•Component analyses are different than

experiments that compare two or more
distinct treatments ( comparative analysis).

• Component = variables that comprise a

treatment
package

• Sometimes , components are themselves

independent
treatments

•E.g., Relaxation training, diff

reinforcement , extinction…

• CA also different from Parametric

analyses (
Ward-Horner, J., & Sturmey, P.
(2010). Component Analysis (CA)
Review
• Literature review of 30 studies
that had component
analysis
• Majority had identified a
necessary component
• Most did not evaluate sufficiency
of the necessary
component
•CA- Can improve social validity by removing in
effective or effortful or un necessary or
punishing components
• CA is part of BACB task analysis suggesting that
it is an
important repertoire for Bas to have

Notations – Types of component and SSD

Combinations:

• [BSL]_[XYZ]_[FU]

• [BSL]_[XYZ]_[ FU]

•MBL –P referes to participant , L indicates leg of a

design – L1, L2…

• P1_L1_[X]_[Y]_[YZ], P1_L2_[X]_[Y]_[YZ],
P1_L3_[X]_[Y]_[YZ] = MBL
across behaviors for one participant

• P1_L1_[X]_[Y]_[YZ], P2_L2_[X]_[Y]_ [YZ],

Ward-Horner, J., & Sturmey, P. (2010).
Purpose of Study:

(a) to review the definition of component

analysis,

(b) to introduce a notation system for

evaluating the
experimental designs of component analyses,

(c)to discuss experimental procedures for

performing component analyses, and

(d)to review previous studies that have

conducted component analyses

Studying effects researchers should

consider:

(a) The additive effects of components,

(b)the multiplicative or interactive relations

among components,

(c) the necessity and sufficiency of each

component,

(d) behavioral effects related to combining

different
components, and

(e) sequence effects.

Additive effects:

• Additive effects refer to the possibility that the

effects of
individual
components of the treatment package are
independent of
each other.

• Y Alone is enough?

Additive?
Examine
necessary and
sufficient
Examine necessity and
succiciency of components:

Both
sufficient ,
neither
necessary

What effect are we seeing

here:

Also Additive?
Both necessary
but neither
sufficient
 Necessary? Sufficient? – sequence
effects

•Difficult to interpret; Y after

10 could be successful
because of prior YZ – e.g.,
conditioned reinforcer

•Also possible- Y good after Z

but not before Z?

•Sequence effects –
mitigate by
counterbalancing or
reversing treatmenet
order
Multiplicative effect:

•Multiplicative effects refer to the possibility that

the effects of one component might depend on the
presence of another; for instance,

Both necessary
but neither
sufficient
Ways to do Component Analysis:

• Drop out – systematically remove components

•Tgt bx improves immediately, components
needed to maintain improvement
is known
• Combination could mask individual
components
effectiveness
•Systematically assess individually and by adding –
good for finding
sufficiency
• Reversal and AT designs more powerful but
• MBL is better when target response is not
reversible

Example – Drop out analysis:

• Majority of studies using drop out found

atleast one necessary component

• L.J. cooper etal., (1995) :

•One child Treatment Package = Choice of

food(S) , Contingent attention upon acceptance
(T), Continued presentation on refusal- escape
extinction (V),Warm up (z)
• Used Baseleine, STVZ increased food
• Alternated
acceptanceTVZ with not much difference
• TV
Is S necessary => Z
not needed
• Alternated STVZ withgreater
TVZ - intake with V =>
no difference
• Comapred
S is notSo VTV
without with
was T than
necessary
component
necessary
 Case
Assignment:

Add –in component Analysis – Example

• Fischer etal., 1993 – 5 individuals

• Components- Extinction (X), Punishment (Y),
FCT (Z)
• For one participant – used reversal design within
MBL (
Vow!) across
disruption, aggression and self-injury.
• 6 phases – Baseline (A), Extinction (B),
Punishment ( C)
, FCT ( D), FCT
+ Punishment ( E), FCT + Extinction ( F)
• Disruptive bx targeted in first leg using BCBEDE
design
• Aggression and Self injury in next two legs –
ACAEDE design
• Can we independently assess effect of X and Y
components?
Fischer etal., 1993 – Continued

• Reversal design in a different environment for

disruptive behavior
• Sequence FEFE
• MBL in first setting showed Problem bx
reduced in
Punishment ( c) , FCT ( D) and FCT +
Punishment ( E)
•Reversal design in second setting Problem
behavior did not reduce in first F phase but did
in remaining 3 phases of FEFR
• All Components evaluated independently,
however
several limitations
can be discussed ( page 14 of 20 under
add –in component analyses)
Articles/
Journals :
THAN
K
YOU!