Teaching of Psychology

http://top.sagepub.com/

Negative Reinforcement and Positive Punishment

James V. McConnell
Teaching of Psychology 1990 17: 247
DOI: 10.1207/s15328023top1704_10

The online version of this article can be found at:

http://top.sagepub.com/content/17/4/247

Published by:

http://www.sagepublications.com

On behalf of:

Society for the Teaching of Psychology

Additional services and information for Teaching of Psychology can be found at:

Email Alerts: http://top.sagepub.com/cgi/alerts

Subscriptions: http://top.sagepub.com/subscriptions

Reprints: http://www.sagepub.com/journalsReprints.nav

Permissions: http://www.sagepub.com/journalsPermissions.nav

>> Version of Record - Dec 1, 1990

What is This?

Downloaded from top.sagepub.com at University of British Columbia Library on December 10, 2014
It is easy to introduce yourself to others and strike up a References
conversation in this setting. Attending the business meeting
also gives one some ideas about issues facing the division. Division on the Teaching of Psychology. (1983). By-laws of the
This information can be used by members to identify divi- Division on the Teaching of Psychology of the American Psy-
sion activities for which they are well suited and to locate the chological Association. Teaching of Psychology, 10, 246-247.
names of people who coordinate these activities. Members Matthews, J. R., Rogers, A. M., & Scheirer, C. J. (1986). Se-
can then volunteer for specific projects or make suggestions lected resources for college teachers of psychology. Teaching of
Psychology, 13, 3-7.
about how current issues might be addressed. Discussion of
such projects and issues can carry over to the social hour,
which follows the business meeting. Because most members
Notes
wear their convention name badges, the social hour is an
excellent place to meet people in a friendly and relaxed 1. An earlier version of this article was presented at the annual
atmosphere. convention of the American Psychological Association, New
., 1989.
Orleans.. Aueust
2. Requests for reprints or additional information should he sent to
Janet R. Matthews, Department of Psychology, Loyola Univer-
Division T w o Governance sity, New Orleans, LA 701 18.

Although involvement in Division Two governance is

limited by the number of positions available in any given
year, it has definite advantages. Faculty from small depart- Negative Reinforcement and Positive
ments who do not have colleagues with similar interests will
find that such involvement provides an additional source of
Punishment
stimulation and friendship. Working on general educational
issues also gives one a broader perspective on issues facing James V. McConnell
the discipline.
Unizlersity of Michigan
Division governance positions include elected officers as
well as appointed committees and task forces. Appointment The concept of negative reinforcement is notoriously difficult to
may result through contacts made in the networking de- teach to introductory students. Reasons for this difjiculty include:
scribed earlier. Annually, TOP prints a list of the current surplus meanings associated with the terms negative and punish-
committee and task force chairs. Work as a reviewer for the ment, the fact that Skinner ( 1 938, 1953) described two types of
Program Committee is used here to illustrate some of the reinforcement but only one type ofpunishment, the tendency that
benefits of involvement in governance activities. Serving as most students have to view reward and punishment in subjective
a reviewer for the Program Committee has a continuing terms, and Skinner's (1938, 1953) conflicting definitions of re-
education function similar to that of reviewing for TOP. inforcement and punishment. Pointing out to students that
Material submitted for program consideration is brief. Con- there are two types of punishment-positive and negative-as
tinuing education comes through awareness of the range of well two types of reinforcement-positive and negative-helps
current work in the discipline rather than content. Serving them learn that negative reinforcement is not a synonym for
on other committees or task forces may provide either a pnishment.
continuing education or networking function. Division Two
committees are described in the bylaws (1983). Task forces There vrobablv is no concern more difficult to teach to
are created to meet specific needs that arise in the division. introductory students than that of negative reinforcement
Current task force topics include: ethical issues, minority (Tauber, 1988). There are at least four reasons why this is so:
issues, and student affairs. (a) the surplus meaning carried by such terms as negative
and pnishment; (b) the fact that Skinner (1938, 1953)
described two types of reinforcement (positive and nega-
tive), but only one type of punishment; (c) the tendency
Summary that most students have to oerceive reinforcement and
punishment in terms of their effects on the organism's emo-
Getting involved in any of the Division Two activities tions rather than on overt behaviors; and (d) Skinner's
described in this article requires some assertiveness. A first (1938, 1953) habit of using the words reinforcement and
request may not lead to the level of involvement desired; bunishment in at least two different wavs. Finding a more
persistence is important. When a task force of committee 'effective way to teach the concept of negative reinforcement
needs a new member, your name is more likely to be raised requires resolving all four problems.
for the vacancy if you have already volunteered to partici-
pate. The need for additional help on projects being dis-
cussed may also be raised during convention sessions. Vol- Surplus Meaning of Negative and Punishment
unteering at that time will also increase the chance of
participation. Activities described in this article involve Webster's New Collegiate Dictionary defines reinforcement
varying degrees of time and work. The benefits for profes- as "the act of strengthening," clearly a behavioral defini-
sional growth and development are there for those who seek tion. However, the dictionary also defines negative as "dis-
them out. agreeable" and punishment as "suffering, pain, or loss," words

Vol. 17, No.4, December 1990

Downloaded from top.sagepub.com at University of British Columbia Library on December 10, 2014
that appeal to inner feelings or emotions, rather than to one of his own experiments, Skinner (1953) said: "The
measurable behaviors. Little wonder, then, that most stu- effect of punishment was a temporary suppression of the
dents (and some psychologists) belleve that negative rein- behavior, not a reduction in the total number of responses"
forcement refers to the act of punishment-that is, to the (p. 184).
onset of pain or dissatisfaction-rather than an action that At first glance, it would appear that Skinner successfully
strengthens a response by reducing or terminating an aver- defined both punishment and reinforcement in terms of thelr
slve stimulus. behavioral consequences. However, an analysis of Skinner's
writings suggests that he also used the terms reinfi)rcement
and punishment in quite a different way-to describe the
Two Types of Reinforcement but Only One Type of onset and offset of certain classes of stimuli under the experr-
Punishment? menter's control. This conflict in Skinner's definitions
shows up most clearly in his discussion of negative reinforce-
Although Skinner (1953) mentioned two types of rein- ment. According to Skinner, "You can distinguish between
forcement-positive and negative-he described but one punishment, which is making an aversive event contingent
type of punishment. Positive reinforcement is the technical upon a response, and negative reinforcement, in which the
term h e used in place of reward; however, he did not supply a elimination or removal of an aversive stimulus, conditioned
technical term to replace punishment. or unconditioned, is reinforcing" (cited in Evans, 1968, p.
Given that reward is a synonym for positive reinforcement 183). Here, surely, Skinner defined both puiushrnent and
and that negative is defined as "something disagreeable," negative reinforcement as the onset or offset of a noxious stim-
many students assume that negative reinforcement should ulus, not as consequential behaviors.
be the technical term that is synonymous with punishment O n e reason, perhaps, why it is difficult to explain nega-
(Tauber, 1988). Had Skinner provided us with a technical tive reinforcement to students is that most psychologists
term for punishment, this confusion might not exist. tend to use the term in Skinner's second sense--ah "the
elimination or removal of an aversive stimulusw--not as an
increase in the organism's response rate. It is also true, of
course, that the words noxious and uuersrutr appeal as much to
Feelings Rather Than Behaviors
emotional mediators as they do to behavioral reactions.
To avoid discussing what goes on inside the organism
when reinforcement or punishment occurs, Skinner defined
these concepts in terms of their behavioral consequences. Positive and Negative Punishment
He (1953) noted, for example, that organisms tend to ap-
proach positive reinforcers and escape from punishers. How- Fortunately, there is a fairly simple way of overcoming the
ever, it is difficult to convince most students that the orga- difficulties associated with teaching our students what nega-
nism's emotions do not mediate the behavioral changes that tive reinforcement actually is. More than 2 decades ago,
Skinner spoke of. Indeed, one question students frequently Catania (1968) noted that there are acti~allytwo types o i
ask is, "But why does the organism approach or avoid the punishment: "Like reinforcement, punishment can be
stimulus?" They know full well that they like rewards and positive or negative" (p. 241). In a subsequent publication,
dislike punishers and presume that their consequential he- he (1979) defined bositive gunishment as that associated w ~ t h
haviors are motivated by their feelings. Most students also the presentation of a n averslve 5t~mulus,whereas negativt.
assume that other organisms feel and respond as they do. punishment results from the withdrawal of a po\lt~velyrein-
forcing st~mulus.Rachlin (1970) took a slmllar posltlon,
although he d ~ dnot adopt the terms posrttve and negutwe
Skinner's Definitions pun~shmentuntll several years later (Rachl~n,1976). Redd,
Porterfield, and Anderson (1979) ,also spoke of two types of
Most of the time, Skinner ( 1938, 1974) defined reinforce- ~ u n l s h m e n tIn them text on behawor modrficatlon
ment as a behavioral consequence that increases the proba- Despite the loglc of Catanla's assertron that there must he
bility that a response will be repeated in the future. He both posltive and negative punrshment, the terms seldol~i
.
(197 1), uut it this wav: "When a bit of behavior is followed appear even In operant-or~entedlournals. A search of P n -
by a certain kind of donsequence, it is more likely to occur cholonrcal
- Abstracts slnce 1966-the earliest date that the
again, and a consequence having this effect is called a rein- Abstracts are available in computerized form-turned up but
forcer" (p. 27). 18 references for each term. However, in 14 of these articles,
Most of the time, too, Skinner s ~ o k eof ~unishmentin - con-
the authors actuallv referred to the positive or nenatlve
terms of its behavioral consequences. Suppose an organism sequences of using punishment as a means of controlling
is punished for making a certain response. The effects of this behavior. In only 4 of the articles did the authors use the
~ u n i s h m e n tare threefold: he said: First. there is a time- terms as Catania defined them (Jackson & Molloy, 1983;
iimited reduction in the raie at which the organism
- emits the Scott & Wood, 1987; Whitehurst &Miller, 1973; Zirpoli &a
undesirable response; second, there is an increase in escape Lloyd, 1987). The fact that punishment can be either
or avoidance behaviors: and third. if the ~ u n i s h m e n tis positive or negative is discussed in two recent books on
strong enough, the undesirable response actually increases learning and memory ((;ordon, 1989; Hall, 1989) and in
in strength once punishment is terminated. In describing one introductory text (Wade & Tavris, 1987).

Teaching of Psychology
Downloaded from top.sagepub.com at University of British Columbia Library on December 10, 2014
 Table 1. A Comparison of Positive and Negative References
Reinforcement and Punishment
- - - - - - - - ---- pp

Catania, A. C. (Ed.). ( 1968). Contemporary research in operant

Positive Negative behavior. Glenview, IL: Scott, Foresman.
Catania, A. C. (1979). Learning. Englewood Cliffs, NJ: Prentice-
Reinforcement ON + OFF -
Hall.
Punishment ON - OFF +
Evans, R. I. (Ed.). (1968). B. F. Skinner: The man and his ideas.
Note: ON or OFF denotes onset or offset of a stimulus; the symbols New York: Dutton.
+ and - denote the subjective quality associated with the stimulus, Gordon, W. C. (1989). Learning and memory. Pacific Grove, CA:
either satisfaction (+) or dissatisfaction ( -). Brooks/Cole.
Hall, J. F. (1989). Learning and memory (2nd ed. ). Boston: Allyn &
Bacon.
Jackson, H. J . , & Molloy, G. N. (1983). Tangible self-consequa-
tion and arithmetic problem,solving: An exploratory com-
Presenting T h e s e Concepts in Class parison of four strategies. Perceptual and Motor Skills, 57, 471-
477.
When I discuss reinforcement and punishment in my in- Rachlin, H. (1970). Introduction to modern behaviorism. San Fran-
troductory classes, 1 make two small changes in Catania's cisco: Freeman.
terminology. First, 1 substitute the terms stimulus onset and Rachlin, H. ( 1976). Behavior and learning. San Francisco: Freeman.
offset for stimulus presentation and withdrawal, because the Redd, W. H., Porterfield, A. L., & Anderson, B. L. (1979).
Behavior modification. New York: Random House.
latter phrases imply that reinforcemer~tand punishment are
Scott, D. W., & Wood, R. L. (1987). The role of food substitutes
limited to situations in which an experimenter does some- in a token economy system. British Journal of Psychiatry, 150,
thing to a subject. Second, I use the Thorndikean terms 864-867.
satisfying and dissatisfying rather than their operant equiv- Skinner, B. F. (1938). The behavior of organisms: A n experimental
alents, positiwely reinforcing and awersit~e. analysis. New York: Appleton-Century-Crofts.
Once students have mastered infcrmation in Table 1, Skinner, B. F. (1953). Science and human behavior. New York: Free
they appear more receptive to the Skinnerian belief that Press.
feelings can be defined as external behaviors. Use of the table Skinner, B. F. (1971). Beyond freedom and dignity. New York:
also helps students appreciate the fact that a given stimulus Knopf.
input may be both positively and m:gatively reinforcing. Skinner, B. F. (1974). About behaviorism. New York: Knopf.
Tauher, R. T. (1988). Overcoming misunderstanding about the
Food, for example, is a positive reinforcer because it tastes
concept of negative reinforcement. Teaching of Psychology, 15,
+
good (ON ; see Table l ) , but it is also a negative reinforcer 152-153.
because it reduces hunger pangs (OFF -; see Table 1). Re- Thomdike, E. L. (1935). The psychology of wants, interests and
fore I began presenting the table in class, most students attitudes. New York: Appleton-Century-Crofts.
assumed that food was only a positive reinforcer. Wade, C., & Tavris, C. (1987). Psychology. New York: Harper &
Row.
Whitehurst, C., & Miller, E. (1973). Behavior modification of
aggressive behavior on a nursery school bus: A case study. Journal
Conclusions of School Psychology, I 1, 123-128.
Zirpoli, T. J., & Lloyd, J. W. (1987). Understanding and manag-
ing self-injurious hehavior. Remedial and Special Education, 8,
I draw two conclusions from my attempts to clarify the 46-57.
concept of negative reinforcement for my students. First, it
would be simpler if we spoke of onset and offset reinforce-
ment and punishment rather than plxitive and negative Note
reinforcement and ~unishment.However, the older terms
are so ingrained in the literature that it probably would be I thank Charles L. Brewer and three anonymous reviewers for
easier to add the adjectives positive and negative to punish- their many helpful suggestions. I am especially indebted to the
ment than to remove them from reinforcement. reviewer who called the work of Charles Catania and others to my
Second, perhaps it is time to abandon Skinner's fuzzily attention.
defined concepts of reinforcement and punishment and re-
turn to Thorndike's (1935) terminolcgy, namely satisfiers
and dissatisfiers. Thorndike defined a sa1:isfier as a stimulus or
situation that the organism either approached or did nothing A Classical Conditioning Laboratory for
to avoid; he defined a dissatisfier as a 5timulus or situation the Psychology of Learning Course
that the organism either avoided or did nothing to approach.
These definitions are not only behavioral in viewpoint, they G a r y B. Nallan
are simple to comprehend because Thorndike presumed that D. M a r k Bentley
the organism's emotions mediate many of its behaviors. Fur- University of North Carolitza at Asheville
thermore, because Thorndike's terms are not widely used,
adding the adjectives onset and offset to satisfier and dissatis- Laboratory exercises in the psychology of learning course can help
fier should be a less formidable task than trying to do so with students understand the principles of conditioning, learning, and
punishment and reward. memory. This article describes a classical conditioning lab project

Vol. 17, No. 4, December 1990

Downloaded from top.sagepub.com at University of British Columbia Library on December 10, 2014