Download Functional Analysis of Depression: Behavioral Perspective by C.B. Ferster and more Study notes Functional Analysis in PDF only on Docsity!
A Functional Analysis of Depression
C. B. FERSTER American University
Depression, like infantile autism (Ferster, 1961), appears to be an especially appropriate field for the behavioral psychologist because of the missing items of behavior that are so prominent. The be- havioral style of analysis emphasizes the frequency of behavior as the primary datum, while the par- ticular categories of behavior whose frequency is to be accounted for are sought from the clinical literature or from common experience. A behav- ioral approach is useful for communicating, clarify- ing, and making objective knowledge of human behavior that has been discovered clinically or ex- perientially. Thus, the behavior analysis may be used to complement rather than substitute for clinical knowledge (Ferster, 1972). This approach is derived from Skinner's functional analysis of be- havior, particularly in his book Science and Hu- man Behavior (Skinner, 1957). Especially in the chapters analyzing self-control, education, religion, government and law, and social behavior, Skinner defines the major kinds of activities that charac- terize various performances and seeks the variables that account for and influence their frequency. Ferster has provided more details of a functional analysis of self-control (Ferster, 1962) and child- hood psychosis (Ferster, 1961).
The first task in a behavioral analysis is to de- fine behavior objectively, emphasizing functional (generic) classes of performances consistent with prevailing clinical facts, the component behaviors of which can be observed, classified, and counted. Then, the basic behavioral processes can be applied to discover the kinds of circumstances that can in- crease and decrease the frequency of particular
(^1) The writing of this article was in part supported by Grant 10123.4829 from the National Institute of Mental Health. The manuscript profited from thoughtful comments by Margaret Rioch, Stuart Culbertson, and James Gray. Requests for reprints should be sent to C. B. Ferster, Department of Psychology, American University, Washing- ton, D.C. 20016.
ways of acting. Finally, an objective account of the depression phenomenon can provide a frame- work for experiments that measure complex, valid clinical phenomena. An objective account of the functional relation between the patient's behavior and its consequences in the physical and social en- vironment can identify the effective parts of a ther- apeutic procedure so that they can be applied more frequently and selectively.
DEPRESSION AS REDUCED FREQUENCY OF ADJUSTIVE BEHAVIOR
The clinical definition of depression (American Psy- chiatric Association, 1968) as an "emotional state with retardation of psychomotor and thought proc- esses, a depressive emotional reaction, feelings of guilt or criticism and delusions of unworthiness [p. 36]" is a good starting point to uncover the actual forms of conduct that describe the way a depressed person interacts with his environment. To observe the actual behaviors alluded to in the clinical definition of depression, we need look to the frequency of various classes of the depressed per- son's activity as compared with those of a person who is not depressed. The most obvious charac- teristic of a depressed person is a loss of certain kinds of activity coupled with an increase in avoid- ance and escape activity such as complaints, crying, and irritability. A depressed person may sit si- lently for long periods, or perhaps even stay in bed all day. The latency of a reply to a question may be longer than usual, and speaking, walking, or carrying out routine tasks will also occur at a slower pace. While he may at a particular time answer questions, ask for something, or even speak freely, the overall frequency is low. Certain kinds of verbal behavior such as telling an amusing story, writing a report or a letter, or speaking freely without solicitation may seldom occur. Beck's (1967) clinical allusion to depression as representing a reduction in gratification suggests
AMERICAN PSYCHOLOGIST • OCTOBER 1973 • 857
the close connection between the frequency of a
person's acts and the reinforcers supporting them.
A reduced frequency of some activities in a de-
pressed person's repertoire may even lessen the ef-
fectiveness of reinforcers closely connected to physi-
ological processes, such as eating or sex. Because
these activities in their complete function also in-
volve a complex collateral social repertoire, their
reduced frequency does not necessarily mean that
sex or eating is no longer reinforcing. Thus, if
there is a depression of social activity, the by-
product could be a reduced frequency of eating
and sex. The depressed person's commonly re-
ported lack of interest in hobbies and sports and
a lack of concern for emotional attachments could
be an example of a lack of collateral social be-
havior, analogous to the connection between the
physiological and social components of sex and
eating. Playing bridge, for example, not only re-
quires bidding, playing a hand, and scoring, but
also arranging that four people meet, dealing with
the interpersonal effects of winning and losing, and
conversing before and after play and between
hands. Emotional attachments, whatever their
quality of affect, are expressed by substantial
amounts of activity either reinforced or influenced
in some way by the other person. Even fantasy
surrounding an emotional attachment requires a
substantial amount of "silent" if not overt talking.
Thus, almost any significant activity occurs for
multiple sources, and a depression may represent a
weakening of one or more of these sources of be-
havior.
AVOIDANCE AND ESCAPE
The depressed person engages in a high frequency
of avoidance and escape from aversive stimuli, usu-
ally in the form of complaints or requests for help,
along with the reduced frequency of positively re-
inforced behavior. He repeatedly tells how badly
he feels, cries, talks about suicide, and complains
of fatigue and illness. Suicide, except of course
when it is manipulative, is the ultimate expression
of the aversiveness of life's experiences, even
though, as a dependent variable, it does not lend
itself to measurement as an operant whose major
dimension is its frequency. A complaint (e.g., I
can't sleep, or I feel miserable) is a class of activity
that has removed or ameliorated aversive condi-
tions in the past, even though there is nothing the
listener could do or is likely to do in the present
instance. "Please turn down the volume on the
radio," for example, is a minor complaint which is
commonly reinforced when the other person reduces
the radio's volume. The frequent reinforcements
of complaints in ordinary life account for the ex-
tensions of similar but ineffective performances in
the face of other aversive situations. Skinner
(1957, pp. 41-48) made a functional analysis of
this kind of behavior, which he designated super-
stitious or extended mands, akin to the person on
a rainy day who says, "I wish it would stop rain-
ing," or the proverbial "My kingdom for a horse."
These performances are called "extended" be-
cause they are accounted for by past instances
where similar behaviors have been effective. Thus,
the person who says "I wish it would stop rain-
ing" has in the past effectively terminated aversive
situations by statements like, "Would you stop
splashing the water, you are getting me wet."
Three primary conditions combine to produce an
extended mand: (a) Similar performances have
terminated aversive stimuli in the past, (b) an
aversive stimulus is currently present and of suffi-
cient influence to produce a performance that would
eliminate or escape it, and (c) an effective perform-
ance is unavailable in the current repertoire. It
is the aversiveness of the rain or the absence of
a horse coupled with the impossibility of stopping
the rain or continuing the battle without a horse
that induces a person to extend performances from
their past experience.
Complaints and other negatively reinforced com-
ponents of the depressed person's repertoire are
sometimes accompanied by high frequencies of agi-
tated activities such as hand wringing, pacing, or
compulsive talking. These activities serve a func-
tion similar to complaints because they mask (by
prepotency and hence escape and avoidance) other
aversive conditions such as silence, inactivity, or
anxiety-producing activities. Were more effective
methods of avoiding aversive situations available
to the depressed person, they would be prepotent
over the less effective, simpler, and more primitive
ones.
AVOIDANCE AND OTHER AVERSIVELY CAUSED ACTIVITIES PREEMPT POSITIVELY REINFORCED BEHAVIOR
It may not be possible to say whether the reason
a depressed person is agitated, disturbed, and com-
858 • OCTOBER 1973 • AMERICAN PSYCHOLOGIST
to himself. He is more likely to be the listener than the speaker on social occasions. Thus, much of the passive character of the depressed person's repertoire appears to be connected with the pre- ponderance of aversive control applied by others and the absence of positive control derived from the person's own repertoire. Two kinds of actions may occur when a person faces an aversive situation: direct action, which can alter it, or indirect activity, such as complaints, which simply acknowledges it. For example, if an employer requires more work than can be realisti- cally accomplished, a person might take direct action by making plans for finding another job, collect data about how long it took to do different parts of his job, or confront the employer in a dis- cussion about what realistically could be expected; or the reaction could be indirect if he complained to others about how difficult it was to endure the job conditions and how he wished he had a better job. The two kinds of reactions can also be dis- tinguished by the kinds of reinforcers maintaining the person's activity. In the case of direct action, the performances are negatively reinforced oper- ants, which reduce or terminate aversive stimuli. Such a repertoire is active rather than passive because it alters the environment which is control- ling the person's actions. The indirect actions— magical, superstitious, or extended avoidance per- formances—are passive because there is little chance that they can influence the aversive envi- ronment very much even though they are con- trolled by it. The active person acts on the en- vironment reciprocally; the passive person reacts to the environment as in a reflex. Another aspect of the depressed person's passiv- ity is the reaction to the other person's initiative in interactive social situations. Person 1, in a normal interaction, for example, says, "I just read an interesting book" and proceeds to tell about it in reply to Person 2 saying, "Tell me about it." Person 2 interrupts or continues later saying, "I found that useful because.. ." or "It sounds so very different from the author's other books, I wonder... ." In such an interaction there is likely to be positive reinforcement for the behavior of both participants. Two-person interactions may occur, however, in which one person has an active role, the other person has a passive role, and the reinforcer is negative because the passive person emits little behavior or is reinforced by avoiding
some aversive stimulus. Although the behavior of both persons may be reinforced when a depressed person converses with someone else, it is profitable to note the different kinds of reinforcement that govern each person. The speaker is controlling the listener because he is emitting performances rein- forced by the listener's reaction; and the listener is controlling the speaker by prompting, reinforcing, and complying. One person is acting flexibly and with initiative, while the other is avoiding the aversiveness of silence or isolation. An allusion appears to be made to the passivity of the de- pressed person when it is said, clinically, that a depressed person is unwilling to take responsibility for his actions. A person who sulks in the face of difficult interpersonal problems is passive in the same sense as the previous examples. While the sulking sometimes serves as a punishment for someone with whom the depressed person is angry, it has the same magical or extended quality typified by a complaint rather than direct (reinforced) action. The failure to deal with, avoid, or escape from aversive social consequences characterizes many of the situations that produce depression in a therapy group, such as when someone severely criticized during a previous session fails to attend a particu- lar meeting. The therapist observes that the work is not being done, and the remaining activity has the form of statements or speculations of blame for having injured the missing person. When the thera- pist suggests that there is some connection between the missing member and the members' inability to work, the conversation turns to the task of clarify- ing the events surrounding the absence of the miss- ing member, and the group returns to work after the absence is defined. The passivity here lies in the turn to blame and criticisms in the face of the aversiveness the absence of the missing member causes rather than direct action to clarify the de- tailed circumstances surrounding the event. The therapist's prompt provides enough additional sup- port for a discussion of the actual circumstances of the absence to increase the frequency of per- formances that are incompatible with blame. The passive aspects of the behavior in the pre- ceding example concern the kind of behavioral con- trol exerted on the person by aversive stimuli. An active repertoire consists of performances that re- move, alter, or escape from the aversive situation. In a passive repertoire, the aversive stimulus pro-
860 • OCTOBER 1973 • AMERICAN PSYCHOLOGIST
duces magical or extended avoidance; or it disrupts the ongoing repertoire by its aversive by-products.
FREQUENCY OF THE DEPRESSED PERSON'S PERFORMANCE IS THE DATUM OF RESEARCH AND THERAPY
Because a behavioral concept of depression defines the behaviors of the depressed person functionally rather than topographically, the main datum is frequency. The focus on the frequency, rather than topography of a performance, is probably the most important characteristic of a behavioral analysis. Although a description of a depressed person's repertoire stresses activities he does not engage in, these absent performances are usually parts of his present or potential repertoire, but they occur with a low frequency. On many occasions in the past he has dressed, traveled to work, completed his job, and engaged in many performances reinforced by their interpersonal effects. The problem is that the current conditions do not support the activities of which he is potentially capable. A topographic description of the depressed person's repertoire does not distinguish it from the normal one. Al- most any item of conduct observed in a depressed person can be seen at one time or another in a nondepressed person. The depressed person is dis- tinguished from one who is not depressed by the relative frequency of these performances in the total repertoire. Most persons, at one time or an- other, while looking quietly out of a window, say "That was a dumb thing for me to do." They can, at times, be sad, unhappy, or dejected, or lose inter- est in an activity. In any one of these instances it may not be possible to distinguish them from a pathologically depressed person.
The Basic Behavioral Process Which Contributes to or Reduces the Frequency oj a Person's Conduct Since the common denominator among depressed persons is the decreased frequency of many differ- ent kinds of positively reinforced activity, we can- not expect that there will be one cause of depres- sion or a single underlying psychological process, because behavior is a product of so many psycho- logical processes. Using the analogue of genetics, the dependent variable of depression—frequency— is a phenotype that can be caused by a variety of
environmental conditions (genotypes). When we understand all of the processes that can reduce the amount of positively reinforced behavior, we can begin to identify how the person's physical and social environment provides the conditions respon- sible for them. Although there is considerable argument about how much depression is caused by endogenous factors (such as hormones or neuro- chemicals) and how much by the environment (in- ternal conflicts and sudden losses) (Beck, 1967, p. 65), the purposes of a behavioral analysis are best served by avoiding the distinction except insofar as functional analyses of the behavior are different. Whatever the physiological substrate of the de- pressed person's repertoire, we still need to know the functional relation between the behavior and the environment that prompts, shapes, and main- tains it. In most general terms, the processes of behavior fall into three categories: (a) the reinforcement of behavior that explains its origin and cause; (b) its continued maintenance despite infrequent and often uncertain (intermittent) reinforcement; and (c) its selective control by those parts of the physi- cal and social environment that signal the occa- sions when it can be reinforced.
A LIMITED REPERTOIRE OF OBSERVATION LEADS TO A LOW FREQUENCY OF POSITIVE REINFORCEMENT
Depressed repertoires are commonly clinically cate- gorized as having a distorted, incomplete, and mis- leading view of the environment. The indications range from hallucinations and delusions, distortions of body image and physical appearance, distortions of the depressed person's competence, exaggeration of errors, complete inability to evaluate the way other people see him, a tendency to take the blame for events for which there really is no responsi- bility, and a limited and hopeless view of the world. Behaviorally, such a description also alludes to a low frequency of positive reinforcement because so much of the depressed person's activity occurs so inappropriately that it cannot be reinforced. If a person's every act occurred under the circum- stances where it could be effective, reinforcement would occur maximally. Conversely, if a person cannot observe the environment around him accu- rately (the environment does not control the per- formances that will activate it), much behavior will be unsuccessful and will go unreinforced, thereby
AMERICAN PSYCHOLOGIST • OCTOBER 1973 • 861
she acts in return, further distinguishes the occa- sions on which the child acts on the mother and enlarges the child's perceptual capability. Development of the child's perceptual repertoire may be interrupted if there is a serious interfer- ence with or interruption of the reinforcers main- taining the child's activities. A child, for example, may experience difficulties in feeding which prevent the give-and-take which normally makes food in- take a natural result of a continuous interaction with the mother. The mother may not be aware of the flow of milk from the bottle so that the milk passively pours down the child's throat, or the flow might be so slow that it requires such magnitudes of sucking that the movements are not reinforced. A mother who does not react to the tension and relaxation of the child's muscular pos- ture will fail to reinforce the child's movements as he adjusts his posture to produce greater body con- tact with the mother or to escape from discomfort when the mother produces physical strain because she is holding the child like a "sack of potatoes." Not only is there a loss of repertoire that would normally emerge from the successful reinforcement of these interactions, but there is a corresponding lack of perceptual development. The child who does not interact in close correspondence with his mother as she holds him also does not learn to observe the nuances that prompt and cue the inter- actions. When an important performance is not rein- forced, an important by-product may be a large- scale emotional reaction. Not only are such reac- tions momentarily disruptive, but they also influ- ence the parent so much that the parental reaction reinforces the child's atavistic actions. The child who does not receive food from his mother satisfactorily enough to satiate the under- lying deprivation, or one who experiences collateral aversive effects such as choking or extreme physical constraint, may react emotionally. Such activities in turn generate a reaction in the parent, who may either remove the aversive situation by providing the food, or react emotionally (in return) in direct response to the child's elicitation. The result in either case is an increase in frequency of the rage and frustration because of their influence on the parent. Not only does a primitive, atavistic mode of dealing with the parent become a prominent part of the child's repertoire, but it indirectly blocks the enlargement of the child's perception of his world
because the diffuse emotional responses are pre- potent over the smaller magnitude component ac- tivities of a normal interaction. The behaviors involved in such disruptions ap- pear to be the same ones described psychodynami- cally along the dimensions of primary to secondary process. The shift from primary to secondary proc- ess appears to describe the adjustment between the child's current behavior and his progressive adjust- ment to the complex features of his social environ- ment. The child whose interactions with his mother are primarily associated with his own deprivations reinforced because his actions are aversive to her is ultimately blocked from developing an adequate perception of other people, and hence adequate ways of interacting with them interpersonally (sec- ondary process). The child who fails to come under the control of the nuances of the mother's behavior is progressively left behind in his develop- ment of interpersonal behaviors, and whole sectors of interpersonal reactivity are not available to him as a means of commerce with the external world, much along the lines of the classical conno- tations of the fixation of a personality at a particu- lar stage of development. Such failures in the perceptual area may at once suggest some causes of some kinds of depression and at the same time a means of ameliorating them. Behaviorally, the most general way of in- creasing the perceptual repertoire is to begin with simple activities whose reinforcement is reliable but not so invariant that there are not some cir- cumstances where the performance is appropriate and others where it is not. The reinforcement of the performance on one occasion and its nonrein- forcement on another teach the person to observe the appropriate features. The most important ele- ment, however, is an increase in the person's tend- ency to act positively on the environment rather than to react passively and emotionally. A useful schedule of reinforcement, applicable to such a problem, is the differential reinforcement of other behavior. The increase in frequency of re- inforcement behaviors other than primitive or ata- vistic activities eventually decreases their frequency by prepotency and nonreinforcement (Ferster & Perrott, 1968). Ideally, a therapeutic interaction with a psychotherapist simulates just such a dif- ferential reinforcement of other behavior when the therapist observes and functionally analyzes the current verbal and emotional activity. By his re-
AMEEICAN PSYCHOLOGIST • OCTOBER 1973 • 863
actions and questions he reinforces selected parts of the patient's current interaction. Many of these behaviors constitute the patient's talking about or otherwise observing his activity. Although the ultimate goal of therapy is the patient's activity rather than his talking about it, the verbal action serves an important function. First, it is an in- crease in general verbal activity which of itself could be of practical use. Second, it becomes a means for the patient to observe his own activity because speech is differentially reinforced (by the therapist) in relation to his own activity. Third, the patient's descriptions of his own primitive re- actions to aversive or thwarting situations may prompt more effective ways to escape or produce positive reinforcers, when he observes the incom- patibility between what he is doing and what he can say about it rationally. Such talking about one's own behavior needs to be quite durable and of a high frequency before it can be incompatible with and preempt more primitive, less effective forms of conduct.
SCHEDULES OF REINFORCEMENT
A schedule of reinforcement is as important a fac- tor influencing how frequently a reinforced act will occur as any other behavioral process. Perform- ances that occur stably when reinforced frequently will weaken under intermittent reinforcement. Not only does intermittent reinforcement usually reduce the amount of behavior generated by reinforcement, but the kind of intermittency will influence how frequently the act occurs. Schedules of reinforcement requiring large amounts of behavior to produce the relevant change in the environment (e.g., fixed-ratio schedules of reinforcement; Ferster & Skinner, 1957; Skinner,
- are those which are most susceptible to loss. Most work activities are of this sort. A rela- tively fixed number of shovels of dirt are needed to fill a hole. The critical factor is a fixed and large amount of activity required for each rein- forcement. Calling on a large number of persons before consummating a sale, studying all semester for a final examination, working on a term paper, writing a novel, persuading someone, carrying out an experiment which requires long and arduous procedures without indication of success before completion, dealing with a difficult therapeutic en- counter where much thought and stress go into
small indicators of progress, and routine house- work which may require a fixed and large amount of repetitive work, all exemplify a schedule of re- inforcement which can potentially weaken the be- havior severely. The result is frequently seen as an abulia in which the novelist, for example, is unable to work for considerable periods of time after completing the previous work. The effects of such schedules of reinforcement are hard to ob- serve at times even though the predominant result is long periods of inactivity. The parallel to the manic side of depression comes immediately to mind. The enormous influence of such schedules of reinforcement apart from the reinforcer or the associated deprivation is conveyed by animal ex- periments in which a pigeon, for example, pecking for food on a schedule requiring a fixed number of pecks per reinforcement, will starve to death because the bird does not peck often enough to produce the amount of food needed to sustain him. Yet the same bird, when exposed to a variable schedule of reinforcement, sustains its activity easily even when the amount of food received does not meet the bird's metabolic requirement. It is tempting to speculate that this particular schedule of reinforcement exemplifies the middle period of life when most individuals settle down to a routine in which there is a constant steady work requirement as opposed to the variability in quality and amount of work that occurs as one prepares for a career or to climb a career ladder. Perhaps relevant here is the classical phenomenon of the professional, highly successful in his work, who on reaching the pinnacle of success undergoes a profound depression. The upwardly striving per- son is one whose schedules of reinforcement are variable, sometimes requiring large amounts of ac- tivity for reinforcement and at other times requir- ing less. Such variable schedules of reinforcement are much less likely to produce low frequencies (strain) than the schedules associated with a stable work situation in which day in and day out there is a constant amount of activity associated with the required accomplishment.
CHANGES IN THE ENVIRONMENT
Where is an organism's behavior when he is not engaging in it? Where is the patellar reflex when it is not being elicited? Where are the reminis- cences that occur with a close friend when the
864 • OCTOBER 1973 • AMERICAN PSYCHOLOGIST
cupation may generate a repertoire too different to
make contact with the possibilities of a retirement
environment. A successful transition depends on
whether the retired person has a sufficient reper-
toire to make contact with the environment of
retirement. The problem is doubly complex be-
cause the transition may be so sudden.
ANGER AS A FACTOR IN DEPRESSION
Anger, a frequent reference in clinical descriptions,
is commonly said to contribute to depression as a
by-product of its frequent and severe punishment
by parents and others. To understand behavior-
ally how the punishment of anger can be an im-
portant factor in depression, we first need to de-
scribe anger as an operant performance which has
a large impact on the other person and, second,
to understand the processes by which punishment
reduces the frequency of the punished act.
Angry or aggressive acts tend to be performances
reinforced by the injury or loss they produce to
another person. It is tempting to overlook the
operant characteristics of anger because it is so
often ineffective (and repressed). Angry acts which
do not appear to act on others like operant per-
formances are probably extended from past ex-
perience of related behaviors which did effectively
injure. The injury may be physical, as in a direct
attack on the person, or it may be a loss of rein-
forcers by an attack on property; a withdrawal of
attention as in ostracism in the extreme case (or
sulking); criticism which is functionally equivalent
to a withdrawal of reinforcement; or insult and
blame, which are functionally equivalent to a loss
of social reinforcement and approval.
Angry or aggressive acts are defined as a class of
performances, identified, not by their topography,
but by the way they influence another person.
Colloquially, the term aggressive may refer either
to vigorous and pronounced effect of a performance
or the injury it produces. Clinical examples tend
to refer to acts that have both properties at once.
Aggressive acts are frequently disguised or "soft-
ened" because a "naked" aggressive act will pro-
duce such a large aversive reaction. Sarcasm ex-
emplifies one kind of softening; aggressive humor
is another. The form of an aggressive remark
may be as unobtrusive as a tendency to comment
on the unfavorable aspects of another person's con-
duct, either by criticism or by especially noting un-
fortunate occasions.
Since aggressive acts are aversive to others by
design, their punishment is the rule rather than
the exception. Experimental psychology contrib-
utes to our understanding of this problem by
knowledge of the characteristics of the punishment
process. There is some dispute in the psychological
literature about whether punishment can directly
reduce the frequency of the punished act, as in an
algebraic subtraction (Azrin & Holz, 1966) or
whether the reduced frequency is always a tempo-
rary suppression. However this dispute is resolved,
it is clear that punished acts are frequently sup-
pressed rather than eliminated from a person's rep-
ertoire. The process is illustrated by the child,
facing a piece of bric-a-brac, who has been pun-
ished for playing with it. The tendency to play
with the attractive but fragile toy induces the child
to put-his hand behind his back. Or the child in
school, provoked to laughter by his classmates or
a classroom event, bites his lip to the point of
pain because the teacher would punish laughter.
Both of these performances prevent the punished
act because they are incompatible with it. Thus,
the smiling and the reaching for the piece of bric-
a-brac may remain intact in the repertoire, but
with a reduced frequency because any incipient
tendency creates an aversive situation whose re-
moval reinforces and increases the frequency of
the incompatible performances, such as putting the
hands behind the back or biting the lip.
The consequences of such suppression might not
be serious if it were limited to one specific act, but
the more likely situation is the suppression of the
generic class of behaviors, of which the particular
act is an example. Someone who becomes angry
with another person, besides being inclined to
injure, is also confronting him about the situation
causing the anger. The punishment of the single
act of anger is not nearly as serious as the resulting
disinclination to confront others over the inter-
personal causes of anger. The phenomenon of
psychodynamic suppressions appears to be a close
analogue to the examples of everyday children's
behavior, except that the performances tend to be
verbal. As a word, a phrase, a thought, or an
association comes into consciousness (behaviorally
we would say, as its probability of emission in-
creases), there is an automatic reinforcement of
incompatible behavior. The repression, according
866 • OCTOBER 1973 • AMERICAN PSYCHOLOGIST
to this process, is an actual activity or performance with a certain persistence and frequency which will occur in a dynamic balance with the punished or anxiety-provoking performance. Because the be- haviors that actually suppress the punished acts are a prominent and frequent activity that do not serve any useful function in a person's commerce with the external environment, it subtracts from the finite amount of activity of which a person is capa- ble. The metaphor of a fixed amount of energy which may be apportioned to the repression activity or the external world seems to convey the sense of the behavioral analysis. The repression of pun- ished behavior appears to be a potentially serious contributor to depressions because it commits such a large part of a person's repertoire to activities that do not produce positive reinforcement.
SUPPRESSED ANGER IMPLIES A LOSS OF IMPORTANT SOCIAL REINFORCERS
Because angry or aggressive acts are injurious to the person they are directed at, an important by- product is a loss of reinforcers. There is an obvi- ous incompatibility between acting to injure some- one and expecting him to provide positive events in a social interaction. Depression also function-. ally insulates the person from the vigorous counter- reaction it is likely to evoke. Repressed and fanta- sized forms of anger are much less likely to be punished than overt forms. The usual experience is likely to be an aggressive counterreaction in addi- tion to the loss of reinforcers. It is easy to con- jecture vignettes from a child's developmental his- tory in which a parent significantly withdrew from a child, perhaps, totally in the face of his anger. The effect on the child is exacerbated because the withdrawal of reinforcers may increase the child's anger and emotionality, leading to further instances of loss of parental attention, affection, and ordi- nary items of daily support. Anger automatically has some of the properties of a "time out," a technical behavior term which describes a specific occasion during which rein- forcement is discontinued. The aversiveness of anger because of the loss of positive reinforcement is equally serious even when the angry acts do not influence another person adversely because they are suppressed. Common experience undoubtedly provides many instances that demonstrate and re- inforce the incompatibility of being angry at some-
one and expecting positive interactions to continue normally. The emotional impact of anger, hence its con- tribution to depression, may also be increased be- cause it comes to function as a "preaversive stim- ulus"—in this case preceding the loss of positive reinforcement. Such preaversive stimuli, in classi- cal animal experiments, markedly reduce the fre- quency of the ongoing operant behavior in the sense of an emotional change—a state of the or- ganism which has a global effect. There is every reason to expect that the same process operates under the comparable conditions in human be- havior.
Implications of a Behavioral Analysis for Therapy and Research
HOW A VERBAL INTERACTION WITH A THERAPIST CAN INCREASE THE FREQUENCY OF THE PATIENT'S BEHAVIOR ELSEWHERE
Even though conventional therapy primarily in- volves speaking and listening among several people, a functional analysis of the interaction suggests how an augmented verbal repertoire developed with the therapist can increase the frequency of positive reinforcement elsewhere. The augmentation can occur because a general increase in verbal activity is a means of achieving an accurate view of how the environment works. An accurate view of how the environment works will husband behavior for those occasions when it can be effective and pre- vent acts that will cause trouble because they occur in inappropriate occasions. The process is clearer behaviorally because of the emphasis on a perfor- mance whose reinforcement occurs on one occasion and not another. Probably the most important way that we learn to observe the environment is to comment on it and describe it verbally. In fact, many kinds of distinctions critically important for human func- tions can only occur verbally (Skinner, 1957, p. 109). The low frequency of verbal activity, other than complaints, is a serious impediment to an im- provement of the depressed person's limited and often distorted view of the world. A person who does not talk to other people very much will not be exposed to the differential reaction that occurs when audiences react differently. Much conversa- tion may go unreinforced when it is inappropriate
AMERICAN PSYCHOLOGIST • OCTOBER 1973 • 867
behavioral not because it differs from a clinical
interpretation but because the component events of
the episode are described as would be the case for
any other natural occurrence. Clinical descriptions,
using the personal reactivity of both the therapist
and patient, have the advantage that they can
occur instantly, like a barometer, to register im-
portant human events without the necessity of an
intellectual functional analysis. As such, they pro-
vide the data and objectives for an experimental
analysis of behavior. Much research in the area
of depression deals with topographic aspects of
the depressed person's repertoire such as muscular
tension, reaction time, formal characteristics of
handwriting, pursuit rotor performances, and paper-
and-pencil tests.
Perhaps the tendency to dichotomize "behavior
as a syndrome of important events elsewhere" and
valid clinical phenomena is one of the reasons why
these very indirect and somewhat remote indicators
of real events are used rather than direct measure-
ment of relevant phenomena.
The behavioral analysis of a human event into
a performance and its effect on the environment
does not preclude data like the patient's self-evalu-
ation (Beck, 1967, p. 177). The way a depressed
patient talks about himself is an important class
of factual information, and it becomes nonbehav-
ioral and unobjective only when it is taken as
symptomatic of events elsewhere rather than as an
activity of importance in itself. A behavioral anal-
ysis of the patient's self-evaluation stresses its
functional relation to the person the patient is
talking with or to whom he is complaining about
himself (Ferster, 1972).
It is possible to measure valid clinical phenomena
in the natural environment with the objectivity of
paper-and-pencil tests and tests of motor behavior
if the events are defined functionally and described
as objective aspects of performance and the en-
vironment. What prevents objective description
of the natural environment is a systematic basis
of observations rather than its inherent complexity.
A motion picture camera could record a complex
humorous event completely, for example. What is
lacking is a concept of human nature, like that of
a functional analysis of behavior, which can make
the description of interrelations between the events
as objective as the record of the events themselves,
and a way of taking historical factors into account.
The aspects of mental life believed heretofore to
be inaccessible to objective descriptions can now
be uncovered by recording the functional relation
between the objectively observed components of
the patient's activity, including the interpersonal
effects of the patient's performance. A functional
analysis of the patient's behavior into generic
classes of activity, defined by their function (re-
inforcement, past or present), has the potential of
reducing the mass of data presented by a clinical
episode to manageable records of frequency.
Some starts in this enterprise have already been
made (Lewinsohn, 1969; Lieberman, 1970), and
the results appear to be valid and successful. Per-
haps the near future will see the continued appli-
cation of objective measurement validated by the
intense concentration on the fine-grain analysis of
the individual patient that has exemplified psycho-
analytic research.
Conclusion
Behavioral and clinical concepts are combined here
as a method of uncovering the actual events of
psychopathology and the procedures of therapy,
rather than an attempt to reconcile or bridge the
gap between the conflicting chains of the two
camps. The effectiveness of the psychodynamic
procedures is probably considerably more than the
behavioral psychologists claim and considerably
less than would be expected from the huge amount
of psychotherapy that occurs in the United States.
REFERENCES AMERICAN PSYCHIATRIC ASSOCIATION. Diagnostic and sta- tistical manual of mental disorders. (2nd ed.) Wash- ington, D.C.: Author, 1968. AZRIN, N. H., & HOLZ, W. C. Punishment in operant be- havior: Areas of research and application. New York: Appleton-Century-Crofts, 1966. BECK, A. T. Depression. Philadelphia: University of Pennsylvania Press, 1967. CHODOFP, P. Depression. Paper presented at the Depres- sion Conference, Airlie House, 1973, in press. FERSTER, C. B. Positive reinforcement and behavioral deficits of autistic children. Child Development, 1961, 32, 437-456. FERSTER, C. B. (with J. I. Nurnberger & E. B. Levitt) The control of eating. Journal of Mathetics, 1962, 1, 87-109. FERSTER, C. B. The experimental analysis of clinical phe- nomena. Psychological Record, 1972, 22, 1-16. FERSTER, C. B., & PERROTT, M. C. Behavior principles. New York: Appleton-Century-Crofts, 1962. FERSTEH, C. B., & SKINNER, B. F. Schedules of reinforce- ment. New York: Appleton-Century-Crofts, 1957. HERON, W. T. The pathology of boredom. Scientific American, 1968, 1, 52-56. LEWINSOHN, P. M., WEINSTEIN, M. S., & SHAW, D. A.
AMERICAN PSYCHOLOGIST • OCTOBER 1973 • 869
Depression: A clinical-research approach. In R. D. Rubin & C. M. Franks (Eds.), Advances in behavior therapy. New York: Academic Press, 1969. LIEBERMAN, R. A behavioral approach to group dynamics. Behavior Therapy, 1970, 1, 141-175. SKINNER, B. F. The behavior of organisms. New York: Appleton-Century-Crofts, 1938.
SKINNER, B. F. Walden two. New York: Macmillan,
SKINNER, B. F. Verbal behavior. New York: Appleton- Century-Crofts, 19S7. SKINNER, B. F. What is psychotic behavior? In, Cumu- lative record. New York: Appleton-Century-Crofts,
Proposed Bylaws Amendments
At the meeting of August 27 and 30, 1973, the
Council of Representatives approved the following
recommendations from the Membership Committee
and the Board of Directors:
Amendment to Article II, Section 8, of the APA
Bylaws to clarify the conditions under which an
associate of the American Psychological Association
is advanced to full membership by deletion of the
word "automatically" which causes confusion as it
seems to contradict the fact that an evaluation of
eligibility is involved.
Amendment, to be submitted to the membership,
of Article XVIII, Section 7, of the APA Bylaws to
permit the exemption from dues of a member who
has been totally and permanently disabled. The
Council further voted that, if this amendment is
approved, a Rule of Council would specify that the
determination of disability would be by an employer
or by the Social Security Administration.^1
ARTICLE II, SECTION 8
Associates who meet the standards for Member status will, upon application, [automatically] be advanced to Member on the January first next following the deadline date of application for such advancement....
ARTICLE XVIII, SECTION 7
Any Fellow, Member, or Associate who has reached the age of 65 and has been a member of the Association for at least 25 years, or, regardless of age or length of membership, who has been adjudged totally and permanently disabled, shall be exempt from dues. (^1) Bracketed material is deleted; italicized material is added.
870 • OCTOBER 1973 • AMERICAN PSYCHOLOGIST
