Professional Documents
Culture Documents
Sosial Learning Ya Ini
Sosial Learning Ya Ini
ALBERT BANDURA
Stanford University
guide their actions; they can solve problems symbol¬ they develop thoughts or hypotheses about the
ically without having to enact the various alterna¬ types of behavior most likely to succeed. These
tives; and they can foresee the probable consequences hypotheses then serve as guides for future actions
of different actions and alter their behavior accord¬ [Dulany & O'Connell 1963]. Accurate hypotheses
ingly. These higher mental processes permit both give rise to successful performances, whereas erro¬
insightful and foresightful behavior. neous ones lead to ineffective courses of action. The
A third distinguishing feature of man is that he is cognitive events are thus selectively strengthened or
capable of creating self-regulative influences. By disconfirmed by the differential consequences ac¬
managing the stimulus determinants of given activi¬ companying the more distally occurring overt be¬
ties and producing consequences for their own havior. In this analysis of learning by experience,
actions, people are able to control their own behavior reinforcing consequences partly serve as an unarticu-
to some degree. As illustrated later, cognitive and lated way of informing performers what they must
self-regulative influences often serve important func¬ do in order to gain beneficial outcomes or to avoid
tions in causal sequences. The remainder of this punishing ones,
paper is devoted to a detailed social learning analysis
of how patterns of behavior are acquired and how
Motivational Function of Reinforcement
their expression is continuously regulated by the
interplay of self-generated and other sources of Because of man’s anticipatory capacity, conditions
influences. of reinforcement also have strong incentive-motiva¬
tional effects. Most human behavior is not controlled
by immediate external reinforcement. As a result of
Learning by Direct Experience
prior experiences, people come to expect that certain
In the social learning system, new patterns of be¬ actions will gain them outcomes they value, others
havior can be acquired through direct experience or will have no appreciable effects, and still others will
by observing the behavior of others. The more rudi¬ produce undesired results. Actions are therefore
mentary form of learning, rooted in direct experi¬ regulated to a large extent by anticipated consequen¬
ence, is largely governed by the rewarding and ces. Homeowners, for instance, do not wait until they
punishing consequences that follow any given action. experience the misery of a burning house to buy fire
People are repeatedy confronted with situations with insurance; people who venture outdoors do not
which they must deal in one way or another. Some ordinarily wait until discomforted by a torrential
of the responses that they tiy prove unsuccessful, rain or a biting snowstorm to decide what to wear;
while others produce more favorable effects. Through nor do motorists usually wait until inconvenienced
this process of differential reinforcement successful by a stalled automobile to replenish gasoline.
modes of behavior are eventually selected from Through the capacity to represent actual outcomes
exploratory activities, while ineffectual ones are symbolically, future consequences can be converted
discarded. into current motivators that influence behavior in
It is commonly believed that responses are auto¬ much the same way as actual consequences. Man’s
matically and unconsciously strengthened by their cognitive skills thus provide him with the capabiity
immediate consequences. Simple performances can for both insightful and foresightful behavior.
be altered to some degree through reinforcement
without awareness of the relationship between one’s
Cognitive Mediation of Reinforcement Effects
actions and their outcomes. However, man s cogni¬
tive skills enable him to profit more extensively from A great deal of research has been conducted on
experience than if he were an unthinking organism. whether behavior is learned through the automatic
Within the framework of social learning theory, action of consequences or whether the effects of
reinforcement primarily serves informative and in¬ reinforcement are cognitively mediated. Most of
centive functions, although it also has response- these studies have employed verbal conditioning
strengthening capabilities. situations in which subjects converse freely or con¬
struct sentences and the experimenter rewards cer¬
tain classes of words but ignores all others. Changes
Informative Function of Reinforcement
in the incidence of reinforced verbalizations are then
During the course of learning, people not only examined as a function of whether the participants
perform responses, but they also observe the differ¬ were aware that their verbal utterances were selec¬
ential consequences accompanying their various tively reinforced and whether they recognized the
actions. On the basis of this informative feedback, types of words that produced reinforcement.
Albert Bandura
4
Social Learning Theory
5
Albert Bandura
6
Social Learning Theory
to command greater attention than others. The func¬ right and left turns (for example, RLRRL) than b'
tional value of the behaviors displayed by different reliance upon visual imagery of the itinerary. Ob
models is highly influential in determining which
servational learning and retention are facilitated b'
models will be closely observed and which will be such codes because they carry a great deal of in
ignored. Attention to models is also channeled by formation in an easily stored form.
their interpersonal attraction. Models who possess After modeled activities have been transformer
interesting and winsome qualities are sought out, into images and readily utilizable verbal symbols
whereas those who lack pleasing characteristics tend these memory codes serve as guides for subsequen
to be ignored or rejected, even though they may reproduction of matching responses. That symbolic
excel in other ways.
coding can enhance observational learning is showi
Some forms of modeling are so intrinsically re¬ by studies conducted both with children [Bandura
warding that they can hold the attention of people Grusec, & Menlove 1966, Coates k Hartup 1969] anc
of all ages for extended periods. This is nowhere with adults [Bandura & Jeffery 1971, Gen>tN1971]
better illustrated than in televised modeling. Indeed,
Observers who code modeled activities into eithei
models presented in televised form are so effective words, concise labels, or vivid imagery leam anc
in capturing attention that viewers leam the depicted retain the behavior better than those who simpl)
behavior regardless of whether or not they are given observe or are mentally preoccupied with othei
extra incentives to do so [Bandura, Grusec, k Men- matters while watching the performance of others
love 1966],
In addition to symbolic coding, rehearsal serves a<
Retention processes. A person cannot be much an important memory aid. People who mentally re¬
influenced by observation of a model's behavior if hearse or actually perform modeled patterns of be¬
he has no memory of it. A second major function havior are less likely to forget them than are those
involved in observational learning concerns long¬ who neither think about nor practice what the)
term retention of activities that have been modeled have seen. Some of the behaviors that are learned
at one time or another. If one is to reproduce a observationally cannot be easily strengthened by
model’s behavior when the latter is no longer present overt enactment either because they are socially pro¬
to serve as a guide, the response patterns must be hibited or because the necessary appartus is lacking.
represented in memory in symbolic form. By this It is therefore of considerable interest that mental
means past influences can achieve some degree of rehearsal of modeled activities can increase their
permanence.
retention [Bandura k Jeffery 1971, Michael k Mac-
Observational learning involves two representa¬ coby 1961].
tional systems—an imaginal and a verbal one. During Some researchers [Gewirtz k Stingle 1968] have
exposure, modeling stimuli produce, through a pro¬ been especially concerned about the conditions that
cess of sensory conditioning, relatively enduring, produce the first imitative responses on the assump¬
retrievable images of modeled sequences of be¬ tion that they help explain observational learning at
havior. Indeed, under conditions where stimulus later stages of development. There Is some reason to
events are highly correlated, as when a name is question whether initial and later imitations have
consistently associated with a given person, it is vir¬ equivalent determinants. In early years imitative
tually impossible to hear the name without experi¬ responses are evoked directly and immediately by
encing imagery of the person's physical characteristics. models actions. In later periods, imitative responses
Similarly, reference to activities (golfing, siding), are usually performed in the absence of the models
places (San Francisco, Paris), and things (one's au¬ long after their behavior has been observed. Imme¬
tomobile, Washington Monument) that one has pre¬ diate imitation does not require much in the way of
viously observed immediately elicits vivid imaginal cognitive functioning because the behavioral repro¬
representations of the absent physical stimuli. duction is externally guided by the model's actions.
The second representational system, which prob¬ By contrast, in delayed imitation, the absent modeled
ably accounts for the notable speed of observational events must be internally represented, and factors
learning and long-term retention of modeled con¬ such as symbolic transformation and cognitive organ¬
tents by humans, involves verbal coding of obseiwed ization of modeling stimuli and covert rehearsal,
events. Most of the cognitive processes that regulate which facilitate retention of acquired contents, serve
behavior are primarily verbal rather than visual. The as determinants of observational learning. The differ¬
route traversed by a model can be acquired, retained, ence between physically prompted and delayed imi¬
and later reproduced more accurately by verbal tation is analogous to drawing a picture of one's
coding of the visual information into a sequence of automobile when it is at hand, and from memory.
7
Albert Bandura
8
Social Learning Theory
The representation serves as a guide for behavioral ing stimuli alone does not ensure that they will be
reproduction. Observational learning without per¬ retained for any length of time.
formance is abundantly documented in modeling Anticipation of reinforcement is one of several fac¬
studies using a nonresponse acquisition procedure tors that can influence what is observed and what
[Bandura 1969, Flanders 1968]. After watching goes unnoticed. Knowing that a given model's be¬
models perform novel modes of response, observers havior is effective in producing valued rewards or
can later describe the entire pattern of behavior with averting negative consequences can enhance observa¬
considerable accuracy, and given the appropriate tional learning by increasing observers’ attentiveness
conditions, they often achieve errorless behavioral to .the model’s actions. Moreover, anticipated rein¬
reproductions on the first test trial. forcement can strengthen retention of what has been
It is commonly believed that controversies about learned observationally by motivating people to code
the locus of learning cannot be satisfactorily resolved and to rehearse modeled responses that have high
because learning must be inferred from performance. value. Theories of modeling primarily differ in the
This may very well be the case in experimentation manner in which reinforcement influences observa¬
with animals. To determine whether a rat has mas¬ tional learning rather than in whether reinforcement
tered a maze one must run him through it. With may play a role in the acquisition process. As shown
humans, there exists a reasonably accurate index of in the schematization below, the issue in dispute is
learning that is independent of motor performance. whether reinforcement acts backward to strengthen
To measure whether a human has learned a maze by preceding imitative responses and their association to
observing the successful performances of a model, stimuli or whether it facilitates learning through its
one need only ask him to describe the correct pat¬ effects on attentional, organizational, and rehearsal
tern of right and left turns. Such an experiment processes. It would follow from social learning theory
would undoubtedly reveal that people can learn that a higher level of observational learning would
through modeling before they perform. be achieved by informing observers in advance about
the payoff value for adopting modeled patterns of
Role of Reinforcement in Observational Learning behavior than by waiting until observers happen to
imitate a model and then rewarding them for it.
Another issue in contention concerns the role of In social learning theory reinforcement is con¬
reinforcement in observational learning. As pre¬ sidered a facilitate rather than a necessary condi¬
viously noted, reinforcement-oriented theories [Baer tion because there are factors other than response
& Sherman 1964, Miller & Dollard 1941, Gewirtz & consequences that can influence what people will
Stingle 1968] assume that imitative responses must attend to. One does not have to be reinforced, for
be reinforced in order to be learned. Social learning example, to hear compelling sounds ‘or to look at
theory, on the other hand, distinguishes between prominent visual displays. Hence, when people’s
learning and performance of matching behavior. attention to modeled activities can be gained through
Observational learning, in this view, can occur physical means, the addition of positive incentives
through observation of modeled behavior and ac¬ does not increase observational learning [Bandura,
companying cognitive activities without extrinsic Grusec, & Menlove 1966]. Children who intently
reinforcement. This is not to say that mere exposure watched modeled actions on a television screen in a
to modeled activities is, in itself, sufficient to produce room darkened to eliminate distractions later dis¬
observational learning. Not all stimulation that im- played the same amount of imitative learning re¬
pinges on individuals is necessarily observed by gardless of whether they were informed in advance
them, and even if attended to, the influence of model- that correct imitations would be rewarded or given
Reinforcement Theories
l I f
^ modeling->R-^ S ^
stimuli
9
Albert Bandura
10
Social Learning Theory
syntactic style in accord with the rules guiding the applications of modeling principles [Bandura 1971a].
modeled verbal constructions [Bandura & Harris In these studies, people who dread and avoid certain
1966, Liebert, Odom, Hill, & Huff 1969, Rosenthal activities are able to perform them in varying degrees
& Whitebook 1970]. In addition, modeling influ¬ after observing others perform the feared behaviors
ences have been successful in modifying moral judg¬ repeatedly without any harmful effects.
mental orientations [Bandura & McDonald 1963, The actions of others can also serve as social cues
Cowan, Langer, Heavenrich, & Nathanson 1969], that influence how others will behave at any given
delay of gratification patterns [Bandura & Mischel time. Response facilitation by modeling can be dis¬
1965, Stumphauzer 1969], and styles of information tinguished from observational learning and disinhibi-
seeking [Rosenthal, Zimmerman, & Duming 1970]. tion by the fact that the models actions neither teach
Researchers have also begun to study how modeling new behaviors nor reduce inhibitions, because the
influences alter cognitive functioning of the type behavior in question, which already exists, is socially
described by Piaget and his followers [Rosenthal & sanctioned and therefore is unencumbered by re¬
Zimmerman 1970, Sullivan 1967]. straints. Inhibitory and disinhibitory effects of model¬
The broader effects of modeling are further re¬ ing are examined later in the context of vicarious
vealed in studies employing several models who reinforcement, while social facilitation effects are
display different patterns of behavior. Observers may given detailed consideration in the discussion of
select one or more of the models as the primary stimulus control of behavior.
source of behavior, but they rarely restrict their Modeling influences can have additional effects,
imitation to a single source, nor do they adopt all of though these may be of lesser importance. The be¬
the characteristics of the preferred model. Rather, havior of models directs observers’ attention to the
observers generally exhibit relatively novel responses particular objects used by the performer. As a re¬
representing amalgams of elements from different sult, observers may subsequently use the same
models [Bandura, Ross, & Ross 1963a]. Paradoxical objects to a greater extent, though not necessarily
as it may seem, innovative patterns can emerge solely in an imitative way. In one experiment, for example,
through modeling. Thus, within a given family same- children who had observed a model pummel a large
sex siblings may develop distinct personality charac¬ doll with a mallet not only imitated this specific
teristics as a result of adopting different combinations aggressive action but spent more time pounding
of parental and sibling attributes. A succession of other things with a mallet than did those who did
modeling influences, in which observers later be¬ not see a person handle this particular instrument
come sources of behavior for new members, would [Bandura 1962]. Research findings, considered to¬
most likely produce a gradual imitative evolution of gether, disclose that modeling influences can serve
novel patterns that might bear little resemblance to as teachers, as inhibitors, as disirihibitors, as response
those exhibited by the original models. In homoge¬ elicitors, as stimulus enhancers, and as emotion
neous cultures, where all models display similar arousers.
modes of response, imitative behavior may undergo
little or no change across successive models. It is
diversity in modeling that fosters behavioral inno¬ Regulatory Processes
vation.
The discussion thus far has been concerned solely A comprehensive theory of behavior must explain
with the process of learning through modeling. A not only how response patterns are required but how
second major function of modeling influences is to their expression is regulated and maintained. In so¬
strengthen or to weaken inhibitions of responses that cial learning theory, human functioning relies on
observers have previously learned [Bandura 1971b]. three regulatory processes. They include stimulus,
The effects that models have on behavioral restraints cognitive, and reinforcement control. For explanatory
are largely determined by observation of rewarding purposes, these control functions are discussed sepa¬
and punishing consequences accompanying models' rately, although in reality they are closely inter¬
responses. As a result of seeing a model’s actions related.
punished, observers tend to inhibit behaving in a
similar way. Conversely, observing models engage
Stimulus Control
in threatening or prohibited activities without ex¬
periencing any adverse consequences can reduce To function effectively a person must be able to
inhibitions in observers. Such disinhibitory effects anticipate the probable consequences of different
are most strikingly revealed in recent therapeutic events and courses of action and regulate his be¬
ll
Albert Bandura
havior accordingly. Without a capacity for anticipa¬ stimuli with allergens that evoked respiratory dys¬
tory or foresightful behavior, man would be forced functions. A careful analysis of the situations ir.
to act blindly in ways that might eventually prove which other patients regularly experienced asthmatic
to be highly unproductive, if not perilous. Informa¬ attacks revealed that a varied array of stimuli had
tion about probable consequences is conveyed by acquired controlling value; these included, among
environmental stimuli, such as traffic signals, verbal other things, radio speeches by influential politicians,
communications, pictorial messages, distinctive children’s choirs, the national anthem, elevators,
places, persons, or things, or the actions of others. goldfish, caged birds, the smell of perfume, water¬
In the earliest years of development, environ¬ falls,, bicycle races, police vans, and horses. Once the
mental stimuli, except those that are inherently conditioned stimuli had been identified in a particu¬
painful, exert little or no influence on infants and lar case, Dekker and Groen [1956] were able to
young children. As a result of paired experiences, induce attacks of asthma by presenting the evocative
either direct or vicarious, formerly neutral stimuli stimuli in actual or pictorial form.
begin to acquire motivating and response-directive A great deal of human behavior is under aversive
properties. Stimulus control over emotions and ac¬ stimulus control. In this form of emotional learning,
tions is established in both instances through asso¬ persons, places, and events become endowed with
ciation, but they differ in what gets associated. anxiety-arousing value through association with pain¬
ful experiences. A prime function of most anticipa¬
Stimulus Control of Physiological and tory behavior is to provide protection against poten¬
Emotional Responsiveness tial threats. Once established, defensive behavior is
exceedingly difficult to eliminate because it derives
Physiological responses can be most readily brought self-reinforcing power from its capacity to reduce
under the control of environmental stimuli by having distress. Any protective activities that remove or
two stimulus events occur contiguously. If a formerly avoid discomfort aroused by conditioned threats
neutral stimulus is closely associated with one that are thereby strengthened even though the fears may
is capable of eliciting a given physiological response, no longer be realistically justified. This process is
the former stimulus alone gradually acquires the graphically illustrated by the apocryphal case of a
power to evoke the physiological response or a frac¬ compulsive who, when asked by a psychiatrist why
tional component of it. Although some types of auto¬ he incessantly snapped his fingers, replied that it
matic responses are more difficult to condition than kept ferocious lions away. When informed that
others, almost every form of somatic reaction that obviously there were no lions in the vicinity, the
an organism is capable of making, including changes compulsive client replied, "See, it works.” Inhibi¬
in respiration, heart rate, muscular tension, gastro¬ tions and avoidance responses commonly regarded
intestinal secretions, vasomotor reactions, and other as neurotic (for example, phobias and obsessive-
indices of emotional responsiveness, has been classic¬ compulsive rituals) are similarly strengthened by
ally conditioned to innocuous stimuli [Kimble 1961]. their capacity to lessen subjectively distressing but
Environmental events can likewise acquire the objectively nonexistent threats. To the extent that
capacity to control cortical activity through associa¬ formerly aversive situations are successfully avoided,
tion with either external evocative stimuli or direct individuals prevent themselves from discovering that
brain stimulation [John 1967], Conditioning is by current conditions of reinforcement may differ sub¬
no means confined to events in the external envi¬ stantially from those in the past, when their anxiety
ronment. In recent years researchers [Razran 1961, reactions were appropriate.
Slucki, Adam, & Porter 1965] have been able to con¬ In an early-study Miller [1948] showed how sub¬
dition physiological reactions as well as defensive jective threats can motivate, and fear reduction re¬
behaviors to variations in the intensity of visceral inforce, a variety of defensive behaviors in a cur¬
stimulation from internal organs.
rently benign environment. Animals were shocked in
The conditioning process described above has im¬ a white compartment of a shuttle box where they
portant implications for the understanding of, among quickly learned to escape the painful stimulation by
other things, psychosomatic disorders, defensive be¬ running through an open door into a black compart¬
havior, and evaluative reactions having an affective ment. The formerly neutral white cues thus acquired
component. In the psychosomatic field Dekker, aversive properties so that the animals continued to
Pelser, and Groen [1957] established asthmatic run when placed in the white box even after the
attacks in two patients by pairing formerly ineffective shocks had been completely discontinued. To examine
12
Social Learning Theory
13
Albert Bandura
others are also likely to suffer in one way or another. of internal and external stimuli that may be either
This speculation receives some support in a study closely related to, or temporally remote from, physical
by Church [1959], who found that expression of pain experiences. The fact that stimulus events can be en¬
by an animal evoked strong emotional arousal in dowed with emotion-arousing potential on a vicarious
animals that had suffered pain together; it had much basis further adds to the complexity of conditioning
less emotional effect on animals that had undergone processes. Additionally, it will be shown later that for¬
equally painful experiences hut unassociated with merly neutral stimuli can acquire evocation properties
suffering of another member of their species, and it through association with thought-produced arousal.
left unmoved animals that were never subjected to Once conditioned stimuli have acquired eliciting
any distress. power, this capacity transfers to other sets of stimuli
In vicarious conditioning, events take on evocative that possess similar physical characteristics, to seman¬
properties through association with emotions aroused tically related cues, and even to highly dissimilar
in observers by affective experiences of others. In stimuli that happen to be associated in peoples
laboratory studies of this phenomenon, an observer experiences.
hears a tone and shortly thereafter he sees another By associating the term behaviorimi with odious
person exhibit pain reactions (actually feigned) as images of salivating dogs and animals driven by
though he were severely shocked [Berger 1962]. carrots and sticks, critics of behavioral approaches
Observers who repeatedly witness this sequence of skillfully employ Pavlovian conditioning procedures
events begin to show emotional responses to the on their receptive audiences to endow this point of
tone alone even though no pain is ever inflicted on view with degrading properties. The fact that valua¬
them. In everyday life, of course, pain may be wit¬ tion of places, persons, and things is affected by
nessed from a variety of sources. Observation of one's emotional experiences, whether they be fear¬
failure experiences and the sight of terrified people ful, humiliating, mournful, or pleasurable, does not
threatened by menacing animals have, for instance, mean that such conditioning outcomes reflect a base
served as arousers for emotional learning [Bandura, animal process. To expect people to remain unaffect¬
Blanchard, & Ritter 1969, Craig & Weinstein 1965]. ed by paired experiences is to require that they be
Despite the importance of vicarious learning, there less than human. Moreover, to be sensitive to the
has been surprisingly little study of the factors de¬ consequences of one's actions indicates intelligence
termining how strongly people can be affectively rather than subhuman functioning.
conditioned through the experiences of others. The
nature of the relationship between the observer and
Cognitive Control of Conditioning Phenomena
the sufferer is undoubtedly an influential factor.
People are generally less affected emotionally by the Both popular accounts and psychological descrip¬
adversities of strangers than by the suffering and joy tions of conditioning phenomena convey the impres¬
of those close to them and on whom they depend. sion that emotional responsiveness is conditioned
Observers’ sensitivity to expressions of suffering, automatically through paired stimulation occurring
derived from their past social experiences, may be in a certain temporal relationship. This view is
another contributor. Bandura and Rosenthal [1966], typically reinforced by diagrammatic portrayal of the
for example, found that the degree to which observ¬ process as a nonmediated one in which conditioned
ers were emotionally aroused affected their level of stimuli are directly connected to responses evoked by
vicarious conditioning. Those who were under mod¬ unconditioned stimuli. Results of several lines of
erate emotional arousal displayed the highest rate research indicate that, in humans, conditioning
and most enduring conditioned autonomic responses, phenomena, cannot be fully understood without en¬
whereas those who were either quite calm or highly compassing the influence of cognitive control.
aroused showed the weakest vicarious conditioning. Studies of the relationship between awareness of
Apparently, anguished reactions proved so upsetting paired events and degree of learning reveal that con¬
to observers who themselves were beset by high ditioning involves more than simply linking stimuli
arousal that they diverted their attention from the to new responses through contiguous association.
suffering person and sought refuge in distracting Fuhrer and Baer [1965] report an experiment in
thoughts of a calming nature. which one tone was always followed by shock but a
It is evident from the preceding discussion that different tone was presented alone. People who
emotional learning is much more complex than is recognized that one of the tones signified shock re¬
commonly assumed. Emotional responses can be sponded emotionally whenever it appeared, whereas
brought under the control of intricate combinations those who remained unaware of the stimulus con-
14
Social Learning Theory
tingeneies did not. Other researchers have likewise ditioning. In aware individuals, events that forebode
shown that repetitive association of neutral and distress activate fear-arousing thoughts that produce
aversive stimuli does not produce conditioned emo¬ emotional responses. On the other hand, those who,
tional responses in people who fail to recognize that for one reason or another, fail to notice that the con¬
the two sets of events are related [Dawson & Grings ditioned stimulus foreshadows pain do not conjure
1968, Dawson & Satterfield 1969]. up frightening ideas. As a result, the conditioned
That awareness is a determinant of conditioning stimulus rarely evokes emotional responses even
rather than vice versa is convincingly demonstrated though it is often paired with unpleasant experiences.
by Chatterjee and Eriksen [1962]. People who were Sudden disappearance of conditioned emotional re¬
informed that shock would follow a particular word sponses following awareness that the threat has been
in a chain of associations quickly developed condi¬ removed is also explainable in terms of self-arousal
tioned heart-rate responses. In contrast, subjects who processes. Given such knowledge, conditioned stim¬
were led to believe that the occurrence of shock was uli no longer activate frightening thoughts, and a
not related in any consistent way to their verbaliza¬ major source of emotional responses is thus removed.
tions evidenced no autonomic conditioning even It follows from self-arousal theory that emotional
though they repeatedly experienced the same paired conditioning can be achieved on a cognitive basis in
stimulation as their aware counterparts. the absence of physically painful experiences. Grings
The most striking evidence of cognitive control of and others [Bridger & Mandel 1964, Dawson &
conditioned responses is provided by studies com¬ Grings 1968] provide some evidence on this point.
paring extinction of emotional reactions to condi¬ In their experiments, subjects are told that a given
tioned stimuli in subjects who are informed that the stimulus will sometimes be followed by shock, but
stimuli are no longer followed by painful events and except for a sample experience, this in fact never
in those who are never told that the threat has ceased occurs. Formerly neutral stimuli acquire arousal
to exist. Induced awareness promptly abolishes con¬ capabilities through association with thought-pro¬
ditioned autonomic responses [Grings & Lockhart duced emotional responses. That covert self-stimula¬
1963, Notterman, Schoenfeld, & Bersh 1952, Wickens, tion plays an influential role in this type of
Allen, & Hill 1963] and avoidance behavior [Lindley cognitively based conditioning is shown by Dawson
& Moyer 1961, Moyer & Lindley 1962], whereas un¬ [1966]. He reports that the level of emotional con¬
informed subjects lose their fear only gradually. ditioning was positively related both to the degree
Self-arousal interpretation of conditioning. Ac¬ to which subjects believed that they would be
cording to social learning theory, conditioned emo¬ shocked and to the severity of pain they anticipated.
tional responses are typically mediated through Although the development and extinction of con¬
thought-produced arousal rather than being directly ditioned emotional responses are subject to cognitive
evoked by conditioned stimuli. The power to arouse control, this does not mean that all conditioned re¬
emotional responses is by no means confined to sponses are necessarily consciously mediated. The
external events. People can easily make themselves degree of cognitive control partly depends on the
nauseated by imagining revolting experiences. They conditions under which the emotional responses are
can become sexually aroused by generating erotic originally acquired. Bridger and Mandel [1964]
fantasies. They can frighten themselves by fear- found that autonomic conditioning was similar re¬
provoking thoughts. And they can work themselves gardless of whether stimuli were associated with
up into a state of anger by ruminating about mis¬ threat of shock alone or with threat combined with
treatment from offensive provocateurs. Indeed, Barber shock experiences. Emotional responses established
and Hahn [1964] found that imagined painful stimu¬ on the basis of actual painful experiences, however,
lation produced subjective discomfort and physio¬ were less susceptible to cognitive control. Thought-
logical responses similar to those induced by the induced conditioned responses promptly disappeared
actual painful stimulation. The incomparable Satchel with the knowledge that shocks would no longer be
Paige, whose extended baseball career provided forthcoming. By contrast, conditioned emotional
many opportunities for anxious self-arousal, color¬ responses originating in painful experiences persisted
fully described the power that thoughts can exert for some time despite awareness that the physical
over viscera] functioning when he advised, “If your threat was completely removed.
stomach disputes you lie down and pacify it with These findings may be explained in several ways.
cool thoughts.” It may mean that conditioned responses contain dual
The findings reported in the preceding section are components, as Bridger and Mandel suggest. One of
consistent with the self-arousal interpretation of con¬ the components—created by self-arousal influences—
15
Albert Bandura
is readily modifiable by having a person alter his conditioning does not occur unless the relationship
thoughts. The second component may be a non- between the events is recognized.
mediated one that is directly evoked by external Developmental theories of psychological function¬
stimuli and hence requires disconfirming experiences ing often draw sharp distinctions between associa¬
for its extinction. Snake phobics, for example, will tive and cognitive processes as though they repre¬
instantaneously respond with fear at the sight of a sented independent functions. As noted above,
coiled snake before they have time to cogitate about cognitive influences can markedly affect perform¬
the potential dangers of reptiles. ances regarded as associative in nature. Extent of
An alternative interpretation is that in instances cognitive control can, in turn, be influenced by asso¬
where people either have undergone distressing ex¬ ciative factors. Mandel and Bridger [1967], for in¬
periences or there is a remote chance that they might stance, demonstrated that awareness hastened fear
get hurt, external stimuli become such powerful elici- extinction, but the speed with which aware subjects
tois of fear-provoking thoughts that they are not lost their apprehensions depended on how closely
easily subject to voluntary control. To tell phobics and in what order neutral and aversive stimuli were
who dread heights that they can gaze down safely associated when the fear was first learned. Most
from the rooftop of a tall building because of protec¬ changes in behavior undoubtedly result from the
tive railings does not mean that they will be able to interactive effects of associative and cognitive influ¬
turn off thoughts about the horrendous things that ences.
could conceivably happen. Here fearfulness is still
cognitively mediated but the individual is unable to
Disorders Arising from Inappropriate
control his thoughts, however safe the eliciting situa¬
Stimulus Control
tions may be.
Nonmediational theories of conditioning assume The development of stimulus functions has con¬
that associated stimulus events must be registered in siderable adaptive value, but unfortunatelv, as
the nervous system of the organism. Studies exam¬ alluded to in some of the illustrative material, it can
ining the influence of awareness on conditioning also create needless distress and constricting defen¬
should therefore obtain evidence that the input from siveness. Dysfunctions of this sort can arise in several
the stimulus to be conditioned has in fact been re¬ different ways. Events that happen to occur in the
ceived. It is not inconceivable that in experiments context of traumatic experiences but are in no way
where, to reduce awareness, subjects’ attention is causally related to them sometimes take on aversive
diverted to irrelevant features of the situation, the properties and produce inappropriate generalization
conditioned stimuli may not be registered in a of anxiety reactions. The following letter taken from
sufficiently consistent manner to produce stable con¬ the advice column of a newspaper illustrates such
ditioned responses. Neural responses to afferent input inappropriate generalization.
can be substantially reduced by focusing attention
Dear Abby:
on irrelevant events. In neurophysiological studies
by Hemandez-Peon, Scherrer, and Jouvet [1956], for My friend fixed me up with a blind date and I should
instance, auditory neural responses to a loud sound have known the minute he showed up in a bow tie
were virtually eliminated in cats when they gazed that he couldn't be trusted. I fell for him like a rock.
He got me to love him on purpose and then lied to
at mice, attentively sniffed fish odors, or received
me and cheated on me. Every time I go with a man
shocks that disrupted their attentiveness. Horn who wears a bow tie, the same thine happens. I think
[1960] noted a similar weakening of neural responses girls should be warned about men who wear them.
to a light flash during active attention to other sights
Against Bow Ties
and sounds.
People who direct their attention to extraneous In this above example, the letter writer had gene¬
features may neither experience nor recognize the ralized a whole pattern of social behavior to bow
conditioned stimulus. Absence of conditioning, under ties, a stimulus one would not expect to be routinely
such circumstances, may be erroneously attributed to correlated with deceitfulness. To the extent that her
lack of conscious recognition when, in fact, it derives anticipatory hostile actions evoke negative counter¬
from deficient sensory registration of the stimulus. reactions in bow-tied men, her defensive behaviors
Proof that awareness is necessary for learning would create adverse experiences and are thereby self-
require evidence that, despite adequate neural regis¬ perpetuated. Here the inappropriate behavior is
tration of the paired environmental events, classical maintained by a self-produced reality rather than
16
Social Learning Theory
one that had been in effect in the past but no longer extinction method disabling anxieties are removed
exists under changed conditions of life. by gradually exposing clients to progressively more
Irrational defensive behaviors often occur on the frightening situations or through massive contact
basis of overgeneralization from events associated with what they fear. Some understandably loathe
with traumatic experiences to others that are similar facing what they dread, thus requiring positive in-
to them either physically or semantically. In the
often quoted study by Watson and Rayner [1920], desensitize anxious people by presenting to them the
for example, several pairings of a rat and a loud things- they fear together with anxiety neutralizers
sound produced in a young boy, not only marked (for example, muscular relaxation, tranquilizing im¬
fear of the rat, but the fear generalized widely to agery, or the reassurance and security of a supportive
other furry objects including rabbits, dogs, fur coats, person) capable of eliciting highly positive reactions.
cotton, wool, and even human hair. By reducing the level of anxiety arousal, positive
Clinical evidence exists that suggests that relatively activities allow persons to tolerate threats they might
innocuous stimuli may be invested with powerful ordinarily avoid without experiencing them as un¬
aversive properties through semantic generalization. duly aversive. A third major approach relies upon
Walton and Mather [1963] report the case of a wo¬ modeling procedures wherein anxiety disorders are
man who suffered from obsessions about being dirty eliminated by having anxious clients observe others
and spent much of her life performing incapacitating engaging in threatening activities without experi¬
ygienic rituals. encing any untoward consequences. After undergo¬
ing this form of treatment they lose their fears, they
The main feature was compulsive handwashing which
appeared to arise from doubts about contamination of can engage in activities they formerly inhibited, and
herself by dirt. After every daily activity or whenever they develop more favorable attitudes toward the
she believed she may have touched something which things they abhorred.
had been handled by another person before ner, she
would wash her hands four or five times (taking fifteen
minutes). Door knobs and taps were particularly Stimulus Control of Action
anxiety-provoking as they were handled most fre¬
quently. Going to the toilet would always be followed The same actions can produce markedly different
by handwashing and additional scrubbing of the nails. consequences, depending upon the time, the place,
Taking a bath and washing her hair would occupy sev¬
and the persons toward whom they are expressed.
eral hours due to continual rewashing of herself, the
bath, and the washbasin. To avoid contamination, she Driving through a busy intersection on a red signal,
would ensure that she never brushed against walls or for example, will have painfully different effects from
other people’s clothing and she always kept her own crossing on a green light. People therefore come to
clothes in a special place untouched by others. Any
attend closely to stimuli that predict reinforcement
street or thoroughfare would be examined for patches
and to ignore those that do not, and they utilize cues
of dirt. Many such activities, such as turning on taps
and opening doors were delegated to her mother. She that signify probable consequences in regulating
would in fact never venture outside the house alone their behavior. Stimuli indicating that given actions
lest she were required to handle something and there will be punished or nonrewarded tend to inhibit their
were no facilities for washing. Sitting on public seats
performance, whereas those signifying that the ac¬
was also difficult for her [p. 169].
tions are permissible or rewardable facilitate their
The obsessive-compulsive behavior began with her occurrence. The capacity to regulate responsiveness
severe guilt and feelings of “dirtiness” because of on the basis of information conveyed by antecedent
sexual relations in a love affair with a married man. stimuli about likely response consequences provides
Eventually, a wide range of stimuli related to uro¬ the mechanism for foresightful behavior.
genital activities and all forms of dirt became dis¬ Stimuli acquire controlling power by being corre¬
turbing to her.
lated with differential response consequences. Tra¬
Maladaptive emotional responses can not only be ditional explanations of how stimulus control is de¬
created by unpleasant paired experiences, but they veloped focus primarily on direct training in which
can be eliminated by therapeutic methods utilizing responses are reinforced only in the presence of cer¬
similar principles [Bandura 1969a, Franks 1969]. tain cues but never in a different stimulus context.
This is achieved by repeated exposure to threatening Stimulus control is undoubtedly established and
events without the occurrence of any adverse con¬ maintained in many instances through correlation
sequences. Several variant procedures have been de¬ with experienced response consequences. However,
vised and applied with relatively high success. In the man's symbolic capacity enables him to gain such
17
Albert Bandura
18
Social Learning Theory
intent to get others to follow his example. The be¬ petent observers would have little incentive to learn.
havior of a few models who were attributed high The attributes of models exert greatest influence
power largely determined the conduct of the camp when it is unclear what consequences their behavior
members. The influence of prestigeful modeling is is likely to have. Observers must therefore rely on
demonstrated even more convincingly by Lefkowitz, status cues that may have been correlated with rein¬
Blake, and Mouton [1955]. Pedestrians were more forcement in their past experiences. A prestigeful or
likely to cross a street on a red light when they saw attractive model may induce a person to try a given
a presumably high-status person in executive attire course of action, but if the behavior should prove
do it than when the same transgression was per¬ unsatisfactory, it will be discarded and the model’s
formed by the same model dressed in soiled patched future influence diminished. For this reason studies
trousers, scuffed shoes, and a blue denim shirt. in which response consequences are not evident may
It is not difficult to explain why model status exaggerate the role played by model characteristics
facilitates matching behavior. The actions of models in the long-term control of social behavior.
who have gained some status are more likely to be In everyday life cues that predict the likely con¬
successful and hence have greater functional value sequences of different actions appear as part of a
for observers than the behavior of models who pos¬ bewildering variety of irrelevant events. To compli¬
sess relatively low vocational, intellectual, and cate matters further, the rules governing reinforce¬
social competencies. In situations where people are ment often involve particular combinations of sev¬
uncertain about the wisdom of modeled courses of eral environmental factors. To borrow an example
action, they rely on model characteristics and status- from the study of conceptual behavior, a person is
conferring symbols (for example, speech, dress, de¬ asked to sort a set of pictures that differ in a variety
portment, possession of material goods), which of ways into correct and incorrect piles without being
serve as tangible indicants of past successes. The told the relevant feature. Let us arbitrarily designate
effects of a model's prestige tend to generalize from all pictures containing an adult drinking an alcoholic
one area of behavior to another, as when prominent beverage as positive instances of the concept, and
athletes advise on breakfast cereals as though they those that do not as negative instances. As the per¬
were nutrition experts. Unfamiliar persons also gain son sorts the pictures on the basis of provisional
influence by their similarity to models whose behav¬ guesses, he observes which of his placements get re¬
ior proved successful in the past. warded and he eventually abstracts the common fea¬
Some efforts have been made to identify the types ture from the array of irrelevant stimuli. Response
of people who are most responsive to modeling in¬ consequences are generally determined by the joint
fluences. Observers who achieve better outcomes by presence of several factors. Continuing with the
following the examples of others than through their above example, only pictures depicting drinking by
own independent behavior are readily susceptible to an adult at home or in a bar, in the evening, and in
modeling influences. Therefore, those who have been the company of others might be considered appro¬
frequently rewarded for imitative behavior and those priate, whereas solitary or daytime drinking and
who lack self-esteem, who feel incompetent, and who imbibing in work settings would not. Here the rein¬
are highly dependent are especially prone to pattern forcement rule combines temporal, social, and situ¬
their behavior after successful models [Bandura ational cues. The important role played by cognitive
1969b, Campbell 196]]. processes in rule-regulated behavior is discussed later.
The preceding generalizations must be accepted Defective stimulus control. Effective social func¬
with reservations because the functional value of tioning requires highly discriminative responsiveness,
modeled behavior overrides the influence of either often to subtle variations in stimulus events. Some
model or observer characteristics. It is exceedingly behavior disorders primarily reflect defective stimu¬
unlikely that dull, anxious, and dependent students lus control, due either to faulty reinforcement prac¬
would gain more from observing skilled ski instruc¬ tices or to a loss of such functions under stressful
tors, brain surgeons, or airline pilots than understud¬ experiences. Rosenbaum [1953, 1956], for example,
ies who are bright, attentive, and self-assured. When found that people were less able to distinguish criti¬
modeled activities are highly valued, the more ven¬ cal from irrelevant features of the environment under
turesome and talented observers benefit most from strong than mild threat, and those prone to emotional
exposure to exemplary models. Laboratory studies arousal were most adversely affected in this regard.
reporting prosaic correlates of imitativeness typically Social behavior is extensively regulated by verbal
employ inconsequential modeled behaviors that com¬ cues. We influence people's actions in innumerable
19
Albert Bandura
20
Social Learning Theory
original reinforcement practices are reinstated to de¬ their treatment. Voluntary control of one's own phys¬
termine whether the deviant behavior was in fact iological functioning is achieved through instantane¬
maintained by its social consequences. In this third ous feedback techniques. In this procedure, people
stage, for example, the teachers behaved in their are told to maintain a given level of physiological re¬
customary way, being inattentive to his sociability sponsiveness during which they are signaled with a
but responding with comforting ministrations when tone or light whenever they achieve a desired state
he was alone. The effect of this traditional “mental of body function. Dworkin and Miller [Miller,
hygiene approach was to drive the child back into Di Cara, Solomon, Weiss, & Dworkin 1970] found
seclusiveness. Clearly, social practices should be that a subject could increase and decrease his dia¬
evaluated in terms of the effects they have on recipi¬ stolic blood pressure to some degree by informative
ents rather than in terms of the humanitarian intent feedback of changes in selected directions. In a fas¬
of the practitioners.
cinating project, Kamiya [1968] succeeded in teach¬
In the final phase, the therapeutic contingencies
ing people to control their alpha brain states (exper¬
are reintroduced, the deviant patterns are eliminated,
ienced as feelings of serene well-being) by monitor¬
and the adaptive ones are generously rewarded until
ing their brain waves and sounding a tone when they
they then are adequately supported by their natural
reduced their brain waves to a rhythm between eight
consequences. In the above case, the teachers grad¬ to twelve cycles per second. During alpha periods
ually reduced their rewarding attentiveness as the subjects felt relaxed, they exerted no mental effort,
boy derived increasing enjoyment from play activi¬ and they experienced no visual imagery.
ties with his peers. In follow-up observations he con¬
Budzynski Stoyva, and Adler [1970] successfully
tinued to enjoy his social relationships, which con¬ applied the biofeedback method in treating tension
trasted conspicuously with his previous seclusiveness. headaches resulting from sustained contraction of
Similar treatments conducted with both children scalp and neck muscles. Patients heard a tone with a
and adults reveal that a wide variety of grossly devi¬
frequency proportional to the electromyographic
ant behaviors—including self-injurious actions, hypo¬ (EMG) activity in the monitored forehead muscles.
chondriacal and delusional preoccupations, infantile They were instructed to keep the tone low by relax¬
and regressive patterns, extreme withdrawal, chronic ing their facial muscles. As the patients became more
anorexia, psychogenic seizures, asthmatic attacks, adept at muscular relaxation, the criterion was in¬
psychotic tendencies, and countless other disorders creased in graded steps requiring progressively more
of long standing—have been successfully eliminated, relaxation to achieve low-pitched tones. Through
reinstated, and removed for a second time by alter¬ this method patients who had experienced daily
ing the amount of social reinforcement they elicit headaches over a period of several years quit tensing
from others.
their facial muscles and eventually eliminated their
The regulatory influence of reinforcement has been headaches. The clinical applicability of biofeedback
demonstrated not only with deviant conditions but techniques will depend on the degree of voluntary
with all other forms of behavioral functioning as control people can exert on their physiological re¬
well. Until recently it was widely believed that inter¬
sponses. It would appear from preliminary findings
nal physiological states could be externally aroused that they possess greater self-regulatory capacities
but were not subject to reinforcement control. In an than was previously believed.
ingenious series of studies the eminent psychologist Reinforcement control of behavior is further shown
Neal Miller [1969] succeeded in altering visceral by evidence that people behave quite differently de*
responsiveness in animals, including changes in heart pending on the pattern and frequency with which
rate, blood pressure, blood flow, intestinal contrac¬ their actions are reinforced. Those who have been
tions, and even rate of urine formation by reinforcing rewarded each time they respond are likely to be¬
the animals whenever physiological responses of a come easily discouraged and to give up quickly when
selected rate or magnitude occurred. The preciseness their efforts fail. On the other hand, individuals
of reinforcement control is most impressively re¬ whose behavior has been reinforced intermittently
vealed in a study in which animals were rewarded
tend to persist for a considerable time despite set¬
for relatively greater volume flow of blood to one ear backs and only occasional success.
than to the other. The differential reinforcement pro¬
Intermittent reinforcement can take a variety of
duced corresponding changes in the amount of blood
forms. Sometimes behavior is reinforced only after a
flow to the two ears.
specified period of time has elapsed (fixed-interval
Preliminary studies indicate that this line of re¬ schedule), as in pay periods, eating schedules, and
search holds promise for advancing understanding of recreational cycles. When rewards occur on a fixed
the development of psychosomatic disorders and temporal basis, the payoff is the same regardless of
21
Albert Bandura
22
Social Learning Theory
ties should be performed for their own sake. It is for the wishes and rights of others, and demanded
feared that reinforcing practices may not only inter¬ immediate gratification of their desires. Uncondi¬
fere with the development of self-determining char¬ tioned love, were it possible, would make children
acteristics but may render people so dependent upon directionless, irresponsible, asocial, and quite unlov¬
extrinsic supports that they remain unresponsive able. Most readers are probably acquainted with
without payoffs. Some of the more intemperate families where parents who attempted to approxi¬
critics, whose own activities are generously rein¬ mate this condition succeeded in producing "self-
forced by salaries, consulting fees, book royalties, actualized” tyrants. Guideless interest is clearly not
and the applause of sympathetic audiences, consider enough. Fortunately, the vast majority of parents are
the use of reinforcement to be manipulative and not indiscriminate dispensers of affection. Being hu¬
debasing.
man, they tend to be displeased with reprehensible
The fact that behavior is controlled by its conse¬ actions. It also comes as no surprise that some of the
quences is not a phenomenon created by behavioral strongest advocates of unconditional regard are quite
scientists, any more than physicists are responsible selective in their own social responsiveness, approv¬
for the laws of gravity. The process of natural selec¬ ing things they like and disapproving those they do
tion has favored organisms adept enough to regulate not [Murray 1956, Truax 1966]. And recipients of
their behavior on the basis of the effects it produces. this differential treatment accommodate to their
Continuing with the above example, a social com¬ preferences.
mentator might express moral indignation over gravi¬ There are a number of symbolic reinforcers, other
tational control of behavior and denounce it as de¬ than social responses, that take on reinforcing func¬
humanizing and degrading. A poetic view of man tions. Money, having exchange value for countless
might be more flattering but it would in no way things that people want, is a dependable and dur¬
reduce the likelihood that people will continue to fall able generalized reinforcer of behavior. When ex¬
should they fling themselves off heights. The major emplary achievements are recurrently rewarded,
purpose of psychological science is not to romanticize qualitative differences in performance acquire rein¬
human behavior but rather to understand it. forcing value. After signs of progress and merited
Some of the criticisms that have been levied against attainment become a source of personal satisfaction,
reinforcement practices fail to recognize the com¬ knowledge that one has done well can function as a
plexity and the developmental changes in the conse¬ reward. However, the reinforcement ensuing from
quences that influence behavior. At the earliest de¬ successful results is most likely mediated through
velopmental levels, infants and young children are self-reinforcement. For this reason, correctness feed¬
responsive only to immediate physical events, such back on tasks that are personally devalued or very
as food, painful stimulation, and physical contact. It simple is unlikely to have much reinforcing value.
would be foolhardy for parents to rely on children's On the other hand, attainments that exceed personal
self-actualizing tendencies to keep them out of fires, standards of what constitutes a worthy performance
electric outlets, or busy thoroughfares. In the course activate positive self-evaluations, whereas inadequate
of development, as physically rewarding experiences accomplishments produce self-dissatisfaction.
are repeatedly associated with expressions of interest Many of the activities that enhance competency
and approval, and punishments with disapproval and are initially tiresome and uninteresting and it is not
withdrawal of interest, the social reactions them¬ until one acquires proficiency in them that they as¬
selves eventually acquire reinforcing properties. At sume reinforcing functions. During initial stages of
this higher level of psychological functioning behav¬ skill development children are often coerced to per¬
ior is extensively governed by symbolic social rein¬ form required activities by demands and threats that,
forcements. It would be a rare, unfeeling person who more often than not, instill antipathies rather than
could remain totally indifferent to the sentiments of competencies. This prevalent problem can be largely
others.
avoided by rewarding children's efforts, both socially
Some child-rearing authorities popularized the and tangibly, until the behavior is developed to the
view that healthy personality development can be stage at which it produces more natural reinforcing
achieved only through "unconditioned love." If this effects. Thus, for example, children may temporarily
principle were, in fact, applied, parents would re- require extrinsic encouragement to teach them how
J spond warmly and affectionately regardless of how to read, but after they become skilled at it they read
their children behaved—whether or not they treated on their own for the enjoyment it provides. Many
others maliciously, stole whatever they wanted, other forms of behavior, such as verbal facility and
amassed a record of failing grades, showed contempt manual skills, which enable people to deal more
23
Albert Bandura
25
Albert Bandura
26
Social Learning Theory
In everyday situations people not only see the con¬ strengthening connections between stimuli and re¬
sequences of another’s behavior but how he responds sponses. Such a mechanism of operation would require
to his treatment. There is some evidence, presented observers, not only to perform covert matching re¬
by Ditrichs, Simon, and Greene [1967], that the sponses concurrently with the model, but to experi¬
performers responsiveness to reinforcement signifi¬ ence indirectly the reinforcements. A reinforcement
cantly affects how observers later react when they process of this type is not implausible, though it
themselves are rewarded for displaying similar be¬ seems highly improbable.
havior. Children who observed models give progres¬ Although the preceding discussion is concerned
sively more hostile responses for social approval later with possible mechanisms through which vicarious
increased their own output of hostile responses under reinforcement affects observers, the alternative ex¬
positive reinforcement, whereas when models gave planations apply equally to interpretation of how
progressively fewer rewarded hostile responses or direct reinforcement influences performers. Rein¬
reacted in random fashion, observers did not increase forcements convey information to performers about
their expression of hostility even though they were the types of responses that are appropriate- selective
positively reinforced whenever they did so. Suscep¬ reinforcement directs performers’ attention to corre¬
tibility to direct reinforcement influences is thus lated environmental stimuli that signify probable re¬
increased by observed positive responsiveness and sponse consequences; previous reinforcements create
reduced by observed resistance. expectations that motivate actions designed to secure
In addition to the aforementioned effects of vicar¬ desired rewards and to avoid injurious outcomes;
ious reinforcement, social status can be conferred on punishing experiences can endow persons, places,
performers by the manner in which their behavior and things with fear-arousing properties that inhibit
is reinforced. Punishment tends to devalue the model responsiveness; a given history of positive or nega¬
and his behavior, whereas the same model assumes tive reinforcement can alter people’s self-evaluations
emulative qualities when his actions are praised and in ways that affect their willingness to exhibit behav¬
otherwise rewarded [Bandura, Ross, & Ross 1963b, iors that are discrepant with their self-attitudes and
Hastorf 1965]. Modification of model status, in turn, the determination with which they perform them;
influences the degree to which observers pattern their and finally, the treatment one receives alters liking
own actions after behavior exemplified by different and respect for the reinforcing agent.
models. There are conditions, of course, where ob¬
served punitive treatment enhances rather than low¬
Self-Reinforcement
ers the recipient's social status. People who risk pun¬
ishment for upholding their basic rights and beliefs The discussion thus far has illustrated how people
gain the admiration not only of their peers but of regulate their behavior on the basis of response con¬
others as well, especially when the protest is directed sequences that they either observe or experience
at social practices that violate the professed values of firsthand. If actions were determined solely by exter¬
society. It is for this reason that authoritative agen¬ nal rewards and punishments, people would behave
cies are usually careful not to discipline deviators like weathervanes, constantly shifting in radically
in ways that might martyr them. different directions to conform to the whims of
Observed reinforcements can alter the valuation others. They would act like segregationists with a
of reinforcing agents a5 well as recipients. When racial bigot, like John Birchers with a zealous Bircher,
societal agents misuse their power to reward and like Communists with a devoted Communist, like
punish, they undermine the legitimacy of their au¬ Republicans with a staunch conservative, and like
thority and generate strong resentment. Under these scoundrels with a villainous character. Close scru¬
conditions seeing inequitable punishment, rather tiny of social interactions would most likely reveal,
than prompting compliance, may free incensed ob¬ barring powerful coercive pressures, steadfast adher¬
servers from self-censure of their own actions and ence to ideological positions rather than compliant
thus increase transgressive behavior. This is most behavior reversals. Anyone who attempted to change
likely to occur when retaliative counterreactions are a Bircher into a Communist, or a Catholic into an
self-justified as rectifying past grievances or prevent¬ atheist, would quickly come to appreciate the exist¬
ing further maltreatment. Otherwise considerate peo¬ ence of potent internal sources of behavior control.
ple can thus be readily provoked by observed injus- The notion that behavior is controlled by its con¬
) tice to behave cruelly without remorse. sequences is unfortunately interpreted by most peo¬
None of the foregoing explanations assumes that ple to mean that actions are at the mercy of situa¬
vicarious reinforcement produces its effects by tional influences. In fact, behavior can, and is, exten-
27
Albert Bandura
28
Social Learning Theory
self freely available rewards and reacts in a self- frequent self-dissatisfaction [Bandura, Grusec, 6
derogatory manner. Observers later perform the task Menlove 1967].
alone, receiving a predetermined set of scores. The Children observed a highly capable adult who re
performances for which they reward and punish warded himself only when he obtained superior
themselves are recorded. scores that children rarely achieved when they late)
Results of such experiments show that people tend performed the same task. Prior to this experience
to adopt standards of self-reinforcement displayed the adult model treated half the children in a warm
by exemplary models, they evaluate their own per¬ rewarding manner and the others in a neutral, busi
formances relative to that standard, and then' they nesslike fashion. The quality of the social relation
serve as their own reinforcing agents. In a study by ship was varied on the assumption that the reward
Bandura and Kupers [1964] children who observed ingness of a model, which tends to increase interper
a model set a high standard of self-reinforcement sonal attraction, would facilitate emulation of the
later rewarded themselves sparingly and only when model’s exacting norms.
they achieved superior performances, whereas chil¬ Adherence to high performance standards is gen
dren exposed to models who considered low achieve¬ erally publicly acclaimed. Without societal valuatior
ments deserving of self-reward tended to reinforce most people would forego lofty aspirations if onl)
themselves for mediocre performances. A control because their attainment requires arduous work anc
group of children, who had no exposure to models, much self-denial of readily available gratifications
did not reward themselves selectively for differential To measure the effects of observed social rewards
levels of achievement. Subjects in the experimental with half the children in the experiment the aduf
conditions not only adopted the modeled standards model was praised for adhering to stringent stand
of self-reinforcement but matched variations in the ards of self-reinforcement, but with the remaining
generosity with which the models rewarded their children the model received no social recognition fo:
own performances. his high-standard-setting behavior.
Social groups contain members of widely differing Ordinarily, individuals are exposed to a variety o:
abilities, so that a given individual must select the modeling influences, many of which operate in op
modeled standards against which to evaluate his posing directions. Speculations about the influenc(
own accomplishments. His level of self-satisfaction of multiple modeling on social learning general!)
and of self-disappointment will be determined to a assign importance to conflicting identification witl
large extent by the models with whom he compares adult and peer models. To study how children re
himself. A study by Bandura and Whalen [1966] solve the problem of simultaneous exposure to antag
found that children readily adopted self-reinforce¬ onistic modeling influences, half the children in eacl
ment patterns displayed by either low-achieving subgroup observed both the stringent adult and i
models who were satisfied with mediocre perform¬ peer model who displayed a low standard of self
ances or moderately competent models subscribing reward. When faced with a conflict between adul
to self-reward standards within their reach. Accord¬ and peer standards, children would be predisposec
ing to social comparison theory, people tend to choose toward peer modeling. Not only would a peer b<
reference models similar in ability and to disregard viewed as a more appropriate comparison model bu
those who are too divergent from themselves. Con¬ emulation of high aspirations results in frequent self
sistent with this view, children rejected lofty stand¬ criticism of one’s performances, which discourage:
ards of superior models and adopted lower self-reward adoption of such standards. It was assumed, how
requirements within the range of their achievements. ever, that the tendency for peer modeling to reduce
Although there is an understandable reluctance to the impact of adult modeling might be counteractec
emulate exacting norms of distinguished models, it by opposing influences arising from positive ties tc
is nevertheless not uncommon for people to adopt the adult model and from social recognition of high
stringent standards of self-reinforcement. Indeed, standard-setting behavior.
universities are heavily populated with students who Children exposed to conflicting modeling influ
are self-satisfied only with superior performances in ences were more inclined to reward themselves fo:
whatever academic work they undertake. We con¬ low achievements than children who had observet
ducted an experiment in our laboratory to explore only the adult model consistently adhering to a higl
some of the social conditions that might lead people standard of self-reinforcement. Children were alsc
to emulate austere standards of self-reinforcement more likely to impose severe criteria of self-rewarc
even though adherence to such demands results in on themselves when the adult model received socia
29
Albert Bandura
30
Social Learning Theory
ceptualized within a social learning framework From great deal of personal distress stemming from exces¬
this perspective, self-esteem is the result of discrep¬ sively high standards of self-evaluation often sup¬
ancies between a person’s behavior and the standards ported by unfavorable comparisons with models
that he has selected as indices of personal merit. noted for their extraordinary achievements. As an
When behavior falls short of one's evaluative stand¬ unidentified pundit once remarked, “If you compare
ards, the person judges himself negatively or holds yourself with others, you may become vain or bitter;
himself in low self-esteem. On the other hand, when for always there will be greater and lesser persons
performances coincide with, or exceed, a person's than oneself ” Yet social comparisons are inevitable,
standards he evaluates himself favorably, which is especially in societies that place a high premium on
considered indicative of high self-esteem. competitiveness and individual achievement. Iron¬
The self-concept also reflects the phenomenon of ically, talented individuals who have high aspirations
self-reinforcement. Self-concept usually signifies a that are possible but difficult to realize are especially
person’s tendency to regard different aspects of his vulnerable to self-dissatisfaction despite their notable
behavior positively or negatively. In measuring this achievements. As Boyd [1969] graphically describes
personality characteristic individuals are presented this phenomenon, “Each violinist in any second chair
with a set of evaluative statements in the form of started out as a prodigy in velvet knickers who ex¬
adjective check lists, Q-sorts, or inventories and are pected one day to solo exquisitely amid flowers flung
asked to rate which statements apply to them. The by dazzled devotees. The 45-year-old violinist with
individual responses are then summed to provide a spectacles on his nose and a bald spot in the middle
global index of self-evaluation, which represents his of his hair is the most disappointed man on earth.”
self-concept. Within a social learning approach, a Linus, the security-blanketed member of the Peanuts
negative self-concept is defined in terms of frequent clan, also alluded to this phenomenon when he
negative self-reinforcement of one’s behavior; con¬ wisely observed, “There is no heavier burden than a
versely, a favorable self-concept is reflected in a dis¬ great potential."
position to engage in high positive self-reinforcement. In its more extreme forms, a harsh system of self¬
Marston [1965b] has conceptualized the phenomena reinforcement gives rise to depressive reactions,
subsumed under the term self-concept in much the chronic discouragement, feelings of worthlessness,
same way. and lack of purposefulness. Excessive self-disparage¬
Personality theories often attribute variations in ment, in fact, is one of the defining characteristics of
behavior to differences in values. Behavioral ap¬ psychotic depression. As Loeb, Beck, Diggory, and
proaches treat values largely in terms of incentive Tuthill [1967] have shown, depressed adults evalu¬
preferences. As we have already observed, the kinds ate their performances as significantly poorer than
of responses a person makes are affected by the posi¬ do nondepressed subjects, even though their actual
tive and negative outcomes they are likely to pro¬ achievements are the same. People also suffei from
duce. Traditional value theory; however, is more considerable self-devaluation when they experience
concerned with valuation of behavior than of ex¬ a loss in ability due to age or physical injury but
ternal outcomes. Established personality theories continue to adhere to their original standards of
tend to regard values as global personal entities that achievement. In the latter instances, most of their
influence behavior but they fail to explain exactly performances are negatively self-reinforced to the
how values control specific actions. Within the social point where they eventually become apathetic and
learning framework, self-reinforcement is the con¬ abandon activities that previously brought them a
trol mechanism. The behavioral standards represent great deal of personal satisfaction.
the values and the anticipatory self-satisfaction and When a person’s behavior produces self-punishing
self-criticism for actions that correspond to or deviate consequences, any activities that avert or reduce
from the adopted standards serve as the controlling these disturbing effects are thereby strengthened and
influences. maintained. A variety of deviant behaviors can serve
Dysfunctions in self-reinforcement systems often as¬ as means of escaping or avoiding self-generated dis¬
sume major importance in psychopathology through tress. Some people whose accomplishments bring
their capacity to create excessive self-punishment them a sense of failure resort to alcoholic self¬
and aversive conditions that can maintain deleterious anesthetization; others escape into grandiose ideation
forms of deviant behavior. Many of the people who where they achieve in fantasy what they failed in
seek psychotherapy are behaviorally competent and reality; others protect themselves against self-con¬
free of debilitating anxiety, but they experience a demnation for their lack of advancement by devel-
31
Albert Bandura
32
Social Learning Theory
performance requirements that incurred high effort serve a reinforcing function in regulating overt be¬
costs at minimum material recompense. These find¬ havior. Weiner [1965] reports an experiment in which
ings are at variance with what one might expect on
inappropriate motor responses by adults were either
the basis of reward-cost theories, unless such formu¬
punished by withdrawal of monetary points or by
lations include the self-esteem costs of rewarding having the subjects imagine the same loss of mone¬
devalued behavior.
tary points, or their performances had no conse¬
In recent years psychologists have been developing quences. He found that imagined aversive con¬
self-reinforcement procedures that would enable peo¬ sequences and the actual occurrence of the same
ple to control their own behavior more effectively.
negative outcomes both reduced responding com¬
These methods are applied most extensively to the pared to the condition involving no feedback. Covert
modification of behaviors that are personally disturb¬
self-punishment, however, produced somewhat weak¬
ing or that create chronic difficulties for others. Given er reductive effects.
clearly defined objectives and some practice on how
Although experimental demonstrations are lacking,
valued rewards can be self-administered contingently, there is every reason to expect that people can in¬
children are able to manage their aggressive and crease desired behavior by covert self-reward. The
scholastic behavior as well as or better than their
operation of covert self-reinforcement complicates
teachers through self-reinforcement practices [Good- interpretation of behavior changes accompanying ex¬
let & Goodlet 1969, Glynn 1970, Lovitt & Curtis ternal reinforcements. It may not be the lights, the
1969]. Self-administered aversive consequences have
scores, and the evaluative statements used as re¬
been used with some degree of success to reduce
wards in psychological experiments that are reinforc¬
stuttering, obsessional ruminations, craving for ad¬
ing; rather, such extrinsic events serve as cues that
dictive drugs, and deviant sexual behavior [Bandura
elicit covert self-satisfaction or self-criticism in sub¬
1969a].
jects. It would follow from this line of reasoning that
Recent investigations of techniques of self-control correctness feedback on tasks that are personally de¬
also assign a principal role to self-managed reinforce¬ valued or regarded as trifling is unlikely to activate
ment [Ferster, Numberger, & Levitt 1962, Harris
self-reward and hence will not operate as a positive
1969, Stuart 1967]. In such treatment programs
reinforcement. On the other hand, behavioral im¬
changes in highly refractory behavior are induced by
provements are possible in the absence of extrinsic
having people regulate the stimuli that ordinarily
rewards because people can easily supply them to
control undesired and competing response patterns.
themselves. Preliminary findings of studies of self¬
However, unless positive consequences for self-con¬
reinforcement processes suggest that learning ap¬
trolling behavior are also arranged, the well-inten¬
proaches hold considerable promise for increasing
tioned practices are usually short-lived.
people’s capacity to regulate their own feelings,
Self-controlling behavior is difficult to maintain thoughts, and actions.
because it tends to be associated, at least initially,
with relatively unfavorable conditions of reinforce¬
ment. Prepotent activities such as heavy drinking by Conditions Supporting
alcoholics and excessive eating by obese people are Self-Reinforcement Systems
immediately rewarding, whereas their detrimental
Since self-reinforcement systems are developed
consequences are not experienced for some time. through social influence, there is no reason to assume
Conversely, self-control measures usually produce that, once established, they become autonomous reg¬
immediate unpleasant effects while the personal
ulators completely impervious to subsequent social
benefits are considerably delayed. Self-reinforcement
influences. An interesting, but inadequately explored,
practices are, therefore, employed to provide im¬
question is what supports self-reinforcing actions.
mediate support for self-controlling behavior until
No elaborate theory is needed to explain why people
the benefits that eventually accrue take over the rein¬
reward themselves. The challenging question requir¬
forcing function. This is achieved by having individ¬ ing explanation is why people deny themselves avail¬
uals select a variety of activities that they find
able rewards over which they have full control, why
rewarding and make them contingent upon the per¬
they adhere to exacting standards that require diffi¬
formance of desired behavior.
cult performances, and why they punish themselves.
The preceding studies primarily involve self-admin¬
Several different interpretations have been proposed.
istration of tangible reinforcers. Of considerable in¬
Conditioned relief. In the view advanced by
terest is evidence that imagined consequences can
Aronfreed [1964], people punish themselves because
33
Albert Bandura
34
Social Learning Theory
35
Albert Bandura
36
Social Learning Theory
an act later considered a rebellion against, and mis¬ attempted what I was enjoined. The man seized me, I
trust of, the Almighty, aroused dreadful hallucina¬ tore myself from him, telling him it was necessary for
tions of Hadean torture, which could be banished my salvation; he left me and went down stairs. I then
tried to perform what I had begun; but now I found,
only by performance of arduous bizarre behavior.
either that I could not so jerk myself round on my
The nonoccurrence of subjectively feared, but ob¬ head, or that my fear of breaking my neck was really
jectively nonexistent, threats undoubtedly serves as too strong for my faith. In that case I then certainly
an important source of reinforcement, maintaining mocked, for my efforts were not sincere.
many other types of psychotic behavior. Given fic¬
tional contingencies with a powerful internal rein¬ Failing in my attempts, I was directed to expectorate
forcing system, a person's behavior is likely to remain violently, in order to get rid of my two formidable
enemies; and then again I was told to drink water, and
under very poor environmental control even in the
that the Almighty was satisfied; but that if I was not
face of severe external punishments and blatant dis-
satisfied (neither could I be sincerely, for I knew I had
confirming experiences. not fulfilled his commands), I was to take up my posi¬
tion again; I did so; my attendant came up with an
When I opened the door, I found a stout man servant
assistant and they forced me into a straight waistcoat.
on the landing, who told me that he was placed there Even then I again tried to resume the position to which
to forbid my going out, by the orders of Dr. P. and
I was again challenged. They then tied my legs to the
my friend; on my remonstrating, he followed me into bed-posts, and so secured me (pp. 34-35],
my room and stood before the door. I insisted on go¬
ing out; he, on preventing me. I warned him of the Grotesque homicidal actions provide other strik¬
danger he incurred in opposing the will of the Holy ing illustrations of how behavior can come under
Spirit, I prayed him to let me pass, or otherwise an
bizarre symbolic control. A study of presidential
evil would befal him, for that I was a prophet of the
Lord. He was not a whit shaken by my address, so,
assassins [Weisz & Taylor 1970], for instance, shows
after again and again adjuring him, by the desire of that, with one exception, the murderous assaults were
the Spirit whose word I heard, I seized one of his partly under delusional control. The assassins were
arms, desiring it to wither; my words were idle, no driven to acts of violence by divine inner voices
effect followed, and I was ashamed and astonished.
commanding them to do so, by delusional beliefs
Then, thought I, I have been made a fool of! But I
did not on that account mistrust the doctrines by that the victim was conspiring to harm them, and by
which I had been exposed to this error. The doctrines, grandiose convictions that it was their heroic respon¬
thought I, are true; but I am mocked at by the Almighty sibility to eliminate maleficent leaders in positions
for my disobedience to them, and at the same time, I of power. The assassins, being unusually seclusive in
have the guilt and the grief, of bringing discredit upon
their behavior, shielded the delusional beliefs from
the truth, by my obedience to a spirit of mockery, or,
by my disobedience to the Holy Spirit; for there were corrective social influences.
not wanting voices to suggest 'to me, that the reason
why the miracle had failed, was, that I had not waited
for the Spirit to guide my action when the word was Representational Guidance of Behavior
spoken, and that I had seized the man's arm with the
wrong hand [p. 331. Symbolic processes play a prominent role in the
acquisition and retention of response patterns as well
as in their expression. The memory trace of monetary
The voices informed me, that my conduct was owing
to a spirit of mockery and blasphemy having possession influences is short-lived, but such experiences often
of me . . . that I must, in the power of the Holy Spirit, have lasting behavioral effects. This condition is
redeem myself, and rid myself of the spirits of blas¬ made possible by the fact that transitory external
phemy and mockery that had taken possession of me. events are coded and stored in symbolic form for
The way in which I was tempted to do this was by
memory representation. Patterns of behavior that
throwing myself on the top of ray bead backwards, and
so resting on the top of my head and on my feet alone, have been observed and other experiences long past
to turn from one side to the other until I had broken can thus be reinstated by visualizing them or by
my neck. I suppose by this time 1 was already in a representing them verbally. These internal models
state of feverish delirium, but my good sense and pru¬ of the outside world can serve as guides to overt
dence still refused to undertake this strange action. I
action on later occasions. It will be recalled from the
was then accused for faithlessness and cowardice, of
fearing man more than God. earlier discussion of learning processes that internal
representations of patterned behavior are constructed
from observed examples and from informative feed¬
I attempted the command, the servant prevented me.
1 lay down contented to have proved myself willing to back to one's trial-and-error performances.
obey in spite of his presence, but now I was accused Representational mediators are especially influen¬
of not daring to wrestle with him unto blows. I again tial in early phases of response acquisition. After ac-
37
Albert Bandura
38
Social Learning Theory
cies where they may not ordinarily exist. We‘might markedly different from those aroused in a disco¬
draw again on the advice column of a newspaper theque. A simple cue from a past experience can re¬
counselor to illustrate this process. instate reveries of bygone events. And perturbing
trains of thought can be turned off by directing one’s
Dear Abby:
attention to absorbing matters that elicit superseding
I have trouble with blondes. Every time I go for a girl cognitive activities. This form of self-control, in
and she is a blonde she turns out to be a gold-digger. which thought-produced arousal is diminished by
I notice on TV whenever they have a gold-digger she
engrossment in absorbing literary material, televised
is blonde. The last blonde 1 went with asked me to
programs, vocational and avocational pursuits, and
buy a record every time I took her out. She kept me
busted buying her records. Should I pass up all blondes other engaging projects, is widely used to restore a
from now on? sense of well-being.
The rules and principles that people use to guide
Blonde Trouble their actions do not arise in a mental vacuum. When
Dear Blonde Trouble:
rules defining appropriate behavior in given situ¬
Plenty of golden heads have golden hearts. ations are not explicitly designated, they are arrived
Abby at largely through information conveyed by observed
or experienced response consequences. Provisional
To the extent that the correspondents distrust of
hypotheses that produce responses resulting in rein¬
blondes leads him to behave in ways that provoke
forcement are retained, partially correct hypotheses
unfriendly counteractions from them, the negative
are successively refined on the basis of differential
valence of blonde hair is repeatedly strengthened, and
response feedback until the right one is hit upon, and
it in turn prompts actions that produce reciprocal
hypotheses that give rise to faulty performances are
negative reinforcement. Both processes thus support
promptly discarded. While it is true that implicit
each other.
rules govern behavior, the rules themselves are partly
The way in which beliefs and conscious recog¬
fashioned oul of reinforcement experiences.
nition of environmental contingencies can enhance,
Considering tire intricate interdependence of stim¬
distort, or even negate the influence of reinforcing
ulus, reinforcement, and cognitive control systems,
consequences has already been amply documented
the sharp distinctions frequently drawn between re¬
and needs no further illustration. Cognitive events,
inforcement and cognitive theories are of question¬
however, do not function as autonomous causes of
able value. It has been customary in psychological
behavior. Their nature, their valence, and their oc¬
- theorizing to construct entire explanatory schemes
currence are under stimulus and reinforcement con¬
around a single form of behavioral control, to the
trol. Analysis of cognitive control of behavior is
relative neglect of other obviously influential de¬
therefore incomplete without specifying what con¬
terminants and processes. Some theories have tended
trols the influential cognitions.
to concentrate upon stimulus control effected princi¬
Cognitively based conditioning, for example, can¬
pally through association of experiences; others have
not occur unless the thoughts that serve as sources of
primarily focused upon external reinforcement con¬
emotional responses have been endowed with arousal
trol; and advocates favoring cognitive interpreta¬
potcnial. The research of Miller [1951] and Grose
tions confine their interest largely to symbolic
[1952] demonstrates that thoughts can acquire
mediators. These resolute allegiances to part proc¬
emotion-provoking properties through generalization
esses may encourage intensive investigation of sub¬
from reinforcing experiences associated with overt
systems, but considered alone, they do not provide a
responses. In these studies it was found that thoughts
complete understanding of human behavior.
corresponding to verbalizations that had been pun¬
ished generated physiological arousal, whereas
thoughts representing nonpunished verbalizations Social Learning as a Reciprocal Influence Process
elicited no emotional reactions. If the painful experi¬
ences are sufficiently intense, however, they can con¬ In the social learning view, psychological function¬
dition such potent aversive properties that the dis¬ ing involves a continuous reciprocal interaction be¬
turbing thoughts themselves are completely inhibited tween behavior and its controlling conditions. Early
[Eriksen & Kuethe 1956, Marks & Gelder 1967]. In attempts to incorporate both individual and environ¬
personality theories, thought inhibition is usually mental determinants in personality theory simply de¬
designated as repression.
picted behavior as caused by these two sets of influ¬
Thoughts remain partly under external stimulus- ences. In the widely cited equation B = f(P} E), ac¬
control. Thus, the thoughts elicited in a hospital are tions were presumably best understood by consider-
39
Albert Bandura
40
Social Learning Theory
41
Albert Bandura
BIBLIOGRAPHY
Albert Bandura and Carol J. Kupers, “The Transmission
ol Patterns of Self-Reinforcement through Modeling."
Justin Aronfreed, ‘The Origin of Self-Criticism." Psycho¬ 69^19 ^'^norma^ aru^ Social Psychology, 1964,
logical Review, 1964, 71:193-218.
Albert Bandura and Frederick J. McDonald, “The Influ¬
Teodoro Aylion and E. Haughton, “Control of the Be-
ence of Social Reinforcement and the Behavior of
havoor of Schizophrenic Patients by Food." Journal of
Models in Shaping Children's Moral Judgments."
the Experimental Analysis of Behavior, 1962, 5:343-
352. Journal of Abnormal and Social Psychology, 1963,
67:274—281. ’ *
Teodoro Aylion and Jack Michael, 'The Psychiatric Nurse
Aibert Bandura and Walter Mischel. "Modification of
as a Behavioral Engineer.” Journal of the Experimental Self-Imposed Delay of Reward tlirough Exposure to
Analysts of Behavior, 1959, 2:323-334.
Live and Symbolic Models." Journal of Personality and
Albert Bandura, “Social-Learning Theory of Identificatory Comparative Test of the Status Envy, Social Power
rocesses. In David A. Coslin, ed., Handbook of So¬ and Secondary Remforcement Theories of Identifica-’
cialisation Theory and Research. Rand McNally, 1969b. tory Learning. Journal of Abnormal and Social Psy¬
chology, 1963b, 67:527-534. V
A1priLMnd"ri' T;7chotheraPy Based uPon Modeling Albert Bandura and Carol K. Whalen, ’The Influence of
’rinciples. In Allen E. Bergin and Sol L. Garfield6
Antecedent Reinforcement and Divergent Modeling
Wfle^ 1971a°fc °f Psych°theTaPy and Behavior Change. Cues on Patterns of Self-Reward.” Journal of Person¬
ality and Social Psychology, 1966, 3:373-382.
Theodore X. Barber and Karl W. Hahn, Jr., “Experi-
Tnd S .Studjes i." 'Hypnotic Behavior: Physiological
AJbert Bandura "Vicarious and Self-Reinforcement Proc- and Subjective Effects of Imagined Pain.” Journal of
esses In Robert Glaser, ed., The Nature of Reinforce- Nervous and Mental Disease, 1964, 139:416-425.
ment. Merrill, 1971c. '
Gregory Bateson, ed., Percevals Narrative-. A Patient’s
A|bert Bandura and Peter G. Barab, "Conditions Covem- Account of His Psychosis, 1830-1832. Stanford Uni
mg Nonreinforced Imitation." Developmental Psychol- versity Press, 1961.
°&y> 1B71, m press. 7 Alan A. Benton, "Effects of the Timing of Negative Re-
sponse Consequences on the Observational Learning
A!mn Ban-rdvUran B' Blanchard, and Brunhilde
of Resistance to Temptation in Children.” Dissertation
Thae Relat've Efficacy of Desensitization and
Modeling Approaches for Inducing Behavioral, Affec- Abstracts, 1967,27:2153-2154.
Seymour M. Berger, "Incidental Learning Through Vi-
; ptltU,d'?a Chanees " loumal of Personality
and Social Psychology, 1969, 13:173-199.
ga47r!491einf0rCeme,lt ” PsychoU>gical Reports, 1961,
Aibert Bandura, Joan E. Crusec, and Frances L. Men-
£ve, Observational Learning as a Function of Sym- Seymour M. Berger, “Conditioning through Vicarious In
stigation. Psychological Revieip, 1962,69:450-466.
^K°7n49“ 506nCentiVe Set ” Ch'ld DeveloPment,
SefSJv perger aniSandra L' Johansson, "Effect of
a Model s Expressed Emotions on an Observer’s Resist-
B-cdUra'cJOan, I; G™sec' and Frances L. Men- ance to Extinction.” Journal of Personality and Social
Jove Some Social Determinants of Self-Monitoring
Psychology, 1968, 10:53-58. V
Remforcement Systems" Journal of Personality and
Social Psychology, 1967, 5:449-455. * AIR? E' Wgin-k'S0me ^P^tions of Psychotherapy
Research for Therapeutic Practice." Journal of Abnor-
Albert Bandura and Mary B. Harris, "Modification of
mal Psychology, 1966, 71:235—246. '
Syntactic Style. Journal of Experimental Child Psy¬
chology, 1966, 4:341-352. V L. NL Boyd, "Most Disappointed Men in the World.” San
Francisco Chronicle, March 15, 1969.
Albert Bandura and Robert Jeffery, "Role of Symbolic
Coding and Rehearsal Processes in Observational WTcsR\Bridr and Tn] Mande)’ "A Comparison
S GS,p l esP°nfes Produced by Threat and Elec-
sity 197S1 UnpUb^hed manuscnpt, Stanford Univer-
2 31 '40Ck' °Umal ° PiVchiatTic Research, 19S4
42
Social Learning Theory
1' MutT^i J' tt0yva; an,d C- Adler' "Feedback-Induced RaS°nd 9i!?chs* Seymore Simon, and Barry Greene
Muscle Relaxation: Application to Tension Headache." Effect of Vicanous Scheduling on the Verbal Condi-
Joumai of Behavwr Therapy and Experimental Psy- tiomng of Hostility in Cliildren." Joumai of Personality
chiatry, 1970, 1:205-211. y and Social Psychology, 1967, 6:71-78.
Donald T Campbell, "Conformity in Psychology's The- Donelson E. Dulany "The Place of Hypotheses and In¬
, 1°! Acyu‘red Behavioral Dispositions." hi Irwin tentions: an Analysis of Verba] Control in Verbal
A. Berg and Bernard M. Bass, eds., Conformity and Conditioning. In Charles W. Eriksen, ed„ Behavior
Deviation. Harper, 1961. and Awareness-A Symposium of Research and Inter-
pretation. Duke University Press, 1962.
Randail M. Chambers, "Effects of Intravenous Glucose
Donelson E. Dulany "Awareness, Rules, and Proposi-
Drive "7 °n h,eTrng' General Actlvit)'' a"d Hunger honal Control: a Confrontation with S-R Behavior
Dnve. Journal of Comparative and Physiological Psy-
chology, 1956, 49:558-564. 6 y T T , n, Tnhe1odore R- D«on and David L. Horton
eds., \ erhal Behavior and General Behavior Tlieoru
Bishwa B. Chatterjec and Charles W. Eriksen, ‘Cognitive Prentice-Hall, 1968. J'
Tactors in Heart Rate Conditioning." Journal of Exver
imental Psychology, 1962, 64:272-279. ' P Donelson E. Dulany and Daniel C. O'Connell, "Does
Riwseii M. Church, "Emotional Reactions of Rats to the Partial Reinforcement Dissociate Verbal Rules and the
Behavior They Might be Presumed to Control?” Jour-
7ourm2/ °f Comparative and Physio-
logical Psychology, 1959, 52:132-134. y 2*361-372^ LeaTninS and Verhal behavior, 1963,
Brian Coates and Willard W. Hartup. "Age and Verhali
Joseph W. Eaton and Robert J. Weil, Culture and Mental
Disorders. Free Press, 1955.
r,HDa" arnd P'rC Nanda- ''Mediated Transfer of Atti- A Telt °f the C,assical Conditioning
Mode of Emotion: the Use of Nonpainful Aversive
1963s; 66 12-Te.0/ Abn0Vmal and S0Cial Oology, Stimuli as Unconditioned Stimuli in a Conditioning
Procedure^ Journal of Personality and Social Psu-
Miand chology, 1968, 10:148—156. ^
ng and R^r°n'i‘;COraParison °f Classica! MA
Relational Learning." Unpublished Condition-
thesis
University of Southern California, 1966. Marvin S. Cerst, "Symbolic Coding Processes in Observa-
tiona! Learn,ng. Journal of Personality and Social Psy¬
Mofha«s1cDaJWCon 7 Wi,liamW- Crings, "Comparison chology, 1971, 19:7-17. y
rC0nd,tlOn,n2 and Relational Learning" Jacob L Cew^2 and Karen C. Stingle, "Learning of
Journal of Experimental Psychology, 1968. 76:227-
Generalized Imitation as the Basis for Identification ”
Psychological Review, 1968, 75:374-397.
MlmanelCSR arS0")v d JameA H' Satterfeld. "Can Hu- E. L. dynn, "Classroom Applications of Self-Determined
man GbR Conditioning Occur without Relational
WS 10 °f Applied BehavioT Analysis,
43
Albert Bandura
Arnold Kaufman, Alan Baron, and Rosemarie E. Kopp Isaac M. Marks and Michael C. Gelder, “Transvestism
Some Effects of Instructions on Human Operant Be- and Fetishism: Clinical and Psychological Changes
during Faradic Aversion.” British Journal of Psychiatru
iacM°n‘ r*ychonomic Monograph Supplements, 1966, 1967, 113:711-729.
i. ZiOU.
Gordon A. Marlatt, "Vicarious and Direct Reinforcement
Thomas D Kennedy, "Verbal Conditioning Without
Control of Verbal Behavior in an Interview Setting.”
wareness; the Use of Programmed Reinfqrcement
Unpublished doctoral dissertation, Indiana University
and Recurring Assessment of Awareness." Journal of 1968. 7’
Experimental Psychology, 1970, 84:484-494.
J. Marmor, "Psychoanalytic Therapy as an Educational
Thomas D. Kennedy, "Reinforcement Frequency, Task
Process: Common Denominators in the Therapeutic
Characteristics, and Interval of Awareness Assessment
Approaches of Different Psychoanalytic ‘Schools'.” In
as Factors in Verbal Conditioning Without Awareness."
Jules H Masserman, ed., Science and Psychoanalysis.
Journal of Experimental Psychology, 1971, 88:103-
vol. 5, Psychoanalytic Education. Grime & Stratton,
44
Social Learning Theory
Albert R. Marston, "Imitation, Self-Reinforcement, and Harold L. Rausch, “Interaction Sequences.” Journal of
Reinforcement of Another Person.” Journal of Per¬ Personality and Social Psychology, 1965, 2:487—499.
sonality and Social Psychology, 1965a, 2:255-261.
Gregory Razran, “The Observable Unconscious and the
Albert R, Marston, "Self Reinforcement: the Relevance Inferable Conscious in Current Soviet Psychophysi¬
of a Concept in Analogue Research in Psychotherapy.” ology.” Psychological Review, 1961, 68:81-147.
Psychotherapy: Theory, Research and Practice, 1965b
William J. Redd and Jay S. Bimbrauer, “Adults as Dis¬
2:1-5.
criminative Stimuli for Different Reinforcement Con¬
Donald N. Michael and Nathan Maccoby, "Factors In¬ tingencies with Retarded Children.” Journal of Ex¬
fluencing the Effects of Student Participation on Verbal perimental Child Psychology, 1969, 7:440-447.
Learning from Films: Motivating versus Practice Ef¬ David Riesman, The Lonely Crowd. Yale University
fects. Feedback,' and Overt versus Covert Respond¬ Press, 1950. 7
ing. In Arthur A. Lumsdaine, cd., Student Response
Mary A. Rosekrans and Willard W. Hartup, “Imitative
in Programmed Instruction. National Academy of Sci-
Influences of Consistent and Inconsistent Response
ences—National Research Council, 1961.
Consequences to a Model on Aggressive Behavior in
Neal E. Miller, "Studies of Fear as an Acquirable Drive: Children." Journal of Personality and Social Psy¬
I. Fear as Motivation and Fear-Reduction as Rein¬ chology, 1967, 7:429—434
forcement in the Learning of New Responses.” Journal
Gerald Rosenbaum, “Stimulus Generalization as a Func¬
of Experimental Psychology, 1948, 38:89-101.
tion of Level of Experimentally Induced Anxiety ”
Neal E. Miller, "Leamable Drives and Rewards.” In S. S. Journal of Experimental Psychology, 1953, 45:35-43
Stevens, ed., Handbook of Experimental Psychology Gerald Rosenbaum, "Stimulus Generalization as a Func¬
Wiley, 1951. *
tion of Clinical Anxiety,” Journal of Abnormal and So¬
Neal E. Miller, "Learning of Visceral and Glandular Re¬ cial Psychology, 1956, 53:281-285.
sponses.” Science, 1969, 163:434-445.
Milton E. Rosenbaum, “The Effect of Stimulus and Back¬
Neal E. Miller, Leo V. Di Cara, Henry Solomon, Jay M. ground Factors on the Volunteering Response." Journal
Weiss, and Barry Dworkin, "Learned Modifications of of Abnormal and Social Psychology, 1956, 53:118—121
Autonomic Functions: a Review and Some New Data.” Milton E. Rosenbaum and James L. Bruning, "Direct and
- Circulation Research, Supplement I, 26 & 27 1970’ Vicarious effects of Variations in Percentage of Rein¬
3-11.
forcement on Performance.” Child Development 1966
Neal E. Miller and John Dollard, Social Learning and 37:959-966.
Imitation. Yale University Press, 1941. Milton E. Rosenbaum and Oliver J. Hewitt, “The Effect
Walter Mischel, Personality and Assessment. Wiley, 1968 of Electric Shock on Learning by Performers and Ob¬
Walter Mischel and Robert M Liebert, "Effects of Dis¬ servers." Psychonomic Science, 1966, 5:81-82.
crepancies between Observed and Imposed Reward David Rosenhan, F. Frederick, and Anne Burrowes,
Criteria on Their Acquisition and Transmission." "Preaching and Practicing: Effects of Channel Dis¬
Journal of Personality and Social Psychology, 1966, 3: crepancy on Norm Internalization.” Child Develop¬
45—53. ment, 1968, 39:291-301.
Don Moser, "Screams, Slaps, and Love.” Life Mav 7 Ted L. Rosenthal and Joan S. Whitebook, "Incentives
1965, 90A-101. 7 ’ versus Instructions in Transmitting Grammatical Para¬
meters with Experimenter as Model.” Behaviour Re¬
Kenneth E. Moyer and Richard H. Lindley. "Supple¬
search and Therapy, 1970, 8:189-196.
mentary Report: Effects of Instructions on Extinction
and Recovery of a Conditioned Avoidance Response.” Ted L. Rosenthal and Barry J. Zimmerman, "Modeling by
Journal of Experimental Psychology, 1962, 64 : 95-96. Exemplification and Instruction in Training Conserva¬
tion.” Unpublished manuscript, University of Arizona
Edward James Murray, "A Content-Analysis Method for
1970.
Studying Psychotherapy.” Psychological Monographs
1956, 70(13), svhole no. 420. Ted L. Rosenthal, Barry J Zimmerman, and Kathleei.
Durning. “Observationally-lnduced Changes in Chil¬
Joseph M Notterman, William N. Schoenfeld, and Philip
dren’s Interrogative Classes." Journal of Personality
f. Bersh A Comparison of Three Extinction Proce¬
and Social Psychology, 1970, 16:681-688.
dures Following Heart Rate Conditioning.” Journal of
Abnormal and Social Psychology, 1952. 47:674-677. Jack Sandler and John Quagliano, "Punishment is a Sig¬
nal Avoidance Situation.” Paper read at Southeastern
Edwin P. Parker, "Information Utilities and Mass Com¬
Psychological Association meeting, Gatlinburs, Tenn
munication. In Harold Sackman and Norman Nie,
1964. 6
eds., Information Utility and Social Choice AFIPS
Press, 1970. Robert M. Sasmor, "Operant Conditioning of a Small-
Scale Muscle Response." Journal of the Experimental
Catherine R. Porro, "Effects of the Observation of a Analysis of Behavior, 1966, 9:69-85.
Models Affective Responses to Her Own Transgres¬
M. Sidman, "Avoidance Behavior.” In Werner K. Honig,
sion on Resistance to Temptation in Children” Dis¬
sertation Abstracts, 1968, 28:3064. ed., Operant Behavior. Appleton-Century-Crofts, 1966.
B. F. Skinner, Science and Human Behavior. Macmillan
Leo Postman and Julius Sassenrath, "The Automatic 1953.
Action of Verbal Rewards and Punishments.” Journal
of General Psychology, 1961, 65:109-136. Henry Slucki, Gyorgi Adam, and Robert W. Porter,
‘Operant Discrimination of an Interoceptive Stimulus
Stanley Raclunan, "Sexual Fetishism: an Experimental
in Rhesus Monkeys." Journal of the Experimental Anal¬
Analogue. Psychological Record, 1966, 16:293-296. ysis of Behavior, 1965, 8:405-414.
45
Albert Bandura
46