Psychological Review Vol. 64, No. 6, 1957 MOTIVATIONAL DETERMINANTS OF RISKTAKING BEHAVIOR JOHN W. ATKINSON * University of Michigan preneurship and economic development.2 Earlier studies have searched for a theoretical principle which would explain the relationship of strength of motive, as inferred from thematic apperception, to overt goal-directed performance. The effect of situation cues (e.g., of particular instructions) on this relationship was detected quite early (1), and subsequent experiments have suggested a theoretical formulation similar to that presented, by Tolman (21) and Rotter (20). It has been proposed that n Achievement scores obtained from thematic apperception are indices of individual differences in the strength of achievement motive, conceived as a relatively stable disposition to strive for achievement or success. This motivedisposition is presumed to be latent until aroused by situation cues which indicate that some performance will be instrumental to achievement. The strength of aroused motivation to achieve as manifested in performance has been viewed as a function of both the strength of motive and the expectancy of goal-attainment aroused by situation cues. This conception has provided a fairly adequate explanation of experimental results to date, and several of its implications have been tested (1, 2). The similarity of this conception to the expectancy principle of performance developed by Tolman, which also takes account of the effects of a third variable, incentive, suggested the need for experiments to isolate the effects on motiva- There are two problems of behavior which any theory of motivation must come to grips with. They may finally reduce to one; but it will simplify the exposition which follows to maintain the distinction in this paper. The first problem is to account for an individual’s selection of one path of action among a set of possible alternatives. The second problem is to account for the amplitude or vigor of the action tendency once it is initiated, and for its tendency to persist for a time in a given direction. This paper will deal with these questions in a conceptual framework suggested by research which has used thematic apperception to assess individual differences in strength of achievement motivation (1, 14, 15). The problem of selection arises in experiments which allow the individual to choose a task among alternatives that differ in difficulty (level of aspiration). The problem of accounting for the vigor of response arises in studies which seek to relate individual differences in strength of motivation to the level of performance when response output at a particular task is the dependent variable. In treating these two problems, the discussion will be constantly focused on the relationship of achievement motivation to risk-taking behavior, an important association uncovered by McClelland (14) in the investigation of the role of achievement motivation in entre1 1 wish to acknowledge the stimulation and criticism of colleagues at the Center for Advanced Study in the Behavioral Sciences (1955-56), and also the current support for this research by a grant from the Ford Foundation. 2 McClelland, D. C. Interest in risky occupations among subjects with high achievement motivation. Unpublished paper, Harvard University, June, 19S6. 359 360 JOHN W. ATKINSON tion of variations in strength of expectancy of success and variations in the incentive value of particular accomplishments. The discussion which follows was prompted by the results of several exploratory experiments. It represents an attempt to state explicitly how individual differences in the strength of achievement-related motives influence behavior in competitive achievement situations. A theoretical model will be presented first, then a brief summary of some as yet unpublished experimental evidence will be introduced in order to call the reader’s attention to the kinds of research problems it raises and the scope of its implications. Three variables require definition and, ultimately, independent measurement. The three variables are motive, expectancy, and incentive. Two of these— expectancy and incentive—are similar to variables presented by Tolman (21) and Rotter (20). An expectancy is a cognitive anticipation, usually aroused by cues in a situation, that performance of some act will be followed by a particular consequence. The strength of an expectancy can be represented as the subjective probability of the consequence, given the act. The incentive variable has been relatively ignored, or at best crudely defined, in most research. It represents the relative attractiveness of a specific goal that is offered in a situation, or the relative unattractiveness of an event that might occur as a consequence of some act. Incentives may be manipulated experimentally as, for example, when amount of food (reward) or amount of shock (punishment) is varied in research with animals. The third variable in this triumvirate —motive—is here conceived differently than, for example, in the common conception of motivation as nondirective but energizing drive (3). A motive is conceived as a disposition to strive for a certain kind of satisfaction, as a capacity for satisfaction in the attainment of a certain class of incentives. The names given motives—such as achievement, affiliation, power—are really names of classes of incentives which produce essentially the same kind of experience of satisfaction: pride in accomplishment, or the sense of belonging and being warmly received by others, or the feeling of being in control and influential. McClelland (13, pp. 341352 and 441-458; 15) has presented arguments to support the conception of motives as relatively general and stable characteristics of the personality which have their origins in early childhood experience. The idea that a motive may be considered a capacity for satisfaction is suggested by Winterbottom’s (15, 22) finding that children who are strong in achievement motive are rated by teachers as deriving more pleasure from success than children who are weak in achievement motive. The general aim of one class of motives, usually referred to as appetites or approach tendencies, is to maximize satisfaction of some kind. The achievement motive is considered a disposition to approach success. The aim of another class of motives is to minimize pain. These have been called aversions, or avoidant tendencies. An avoidance motive represents the individual’s capacity to experience pain in connection with certain kinds of negative consequences of acts. The motive to avoid failure is considered a disposition to avoid failure and/or a capacity for experiencing shame and humiliation as a consequence of failure. The principle of motivation. The strength of motivation to perform some act is assumed to be a multiplicative function of the strength of the motive, the expectancy (subjective probability) that the act will have as a consequnece the attainment of an incentive, and the RISK-TAKING BEHAVIOR 361 value of the incentive: Motivation = / only approximated in actual experi(Motive X Expectancy X Incentive). mentation or in the normal course of This formulation corresponds to Tol- everyday life. The discussion will deal man’s (21) analysis of performance ex- only with the effects of the two motives, cept, perhaps, in the conception of a to achieve and to avoid failure, normally motive as a relatively stable disposi- aroused whenever performance is likely tion. When both motivation to ap- to be evaluated against some standard proach and motivation to avoid are of excellence. Behavior directed toward achievement simultaneously aroused, the resultant motivation is the algebraic summation and away from failure. The problem of of approach and avoidance. The act selection is confronted in the level-ofwhich is performed among a set of al- aspiration situation where the individual ternatives is the act for which the re- must choose among tasks which differ sultant motivation is most positive. in degree of difficulty. The problem of The magnitude of response and the per- accounting for the vigor of performance sistence of behavior are functions of the arises in the situation which will be restrength of motivation to perform the ferred to as constrained performance. act relative to the strength of motiva- Here there is no opportunity for the intion to perform competing acts. dividual to choose his own task. He is Recent experiments (2) have helped simply given a task to perform. He to clarify one problem concerning the must, of course, decide to perform the relationship between measures of the task rather than to leave the situation. strength of a particular motive (n There is a problem of selection. In reAchievement) and performance. Per- ferring to this situation as constrained formance is positively related to the performance, it is the writer’s intention strength of a particular motive only to deal only with those instances of bewhen an expectancy of satisfying that havior in which motivation for the almotive through performance has been ternative of leaving the situation is less aroused, and when expectancies of satis- positive or more negative than for perfying other motives through the same formance of the task that is presented. action have not been sufficiently aroused Hence, the individual does perform the to confound the simple relationship. task that is given. The level of perThis is to say no more than that, when formance is the question of interest. Elaboration of the implications of the expectancies of attaining several different kinds of incentives are equally multiplicative combination of motive, salient in a situation, the determination expectancy, and incentive, as proposed of motivation to perform an act is very to account for strength of motivation, complex. Performance is then overde- will be instructive if we can find some termined in the sense that its strength reasonable basis for assigning numbers is now a function of the several different to the different variables. The strength kinds of motivation which have been of expectancy can be represented as a aroused. The ideal situation for show- subjective probability ranging from 0 to ing the relationship between the strength 1.00. But the problem of defining the of a particular motive and behavior is positive incentive value of a particular one in which the only reason for acting accomplishment and the negative incenis to satisfy that motive. tive value of a particular failure is a The theoretical formulation which fol- real stickler. In past discussions of level of aspiralows pertains to such an ideal achievement-related situation, which is at best tion, Escalona and Festinger (see 12) 362 JOHN W. ATKINSON have assumed that, within limits, the attractiveness of success is a positive function of the difficulty of the task, and that the unattractiveness of failure is a negative function of difficulty, when the type of activity is held constant. The author will go a few steps farther with these ideas, and assume that degree of difficulty can be inferred from the subjective probability of success (P8). The task an individual finds difficult is one for which his subjective probability of success (Ps) is very low. The task an individual finds easy is one for which his subjective probability of success (Pa) is very high. Now we are in a position to make simple assumptions about the incentive values of success or failure at a particular task. Let us assume that the incentive value of success (Is) is a positive linear function of difficulty. If so, the value 1 — P, can represent /„, the incentive value of success. When P8 is high (e.g., .90), an easy task, Ia is low (e.g., .10). When Pa is low (e.g., .10), a difficult task, Is is high (e.g., .90). The negative incentive value of failure (//) can be taken as – Ps. When Pa is high (e.g., .90), as in confronting a very easy task, the sense of humiliation accompanying failure is also very great (e.g., — .90). However, when Ps is low (e.g., .10), as in confronting a very difficult task, there is little embarrassment in failing (e.g., — .10). We assume, in other words, that the (negative) incentive value of failure (£/) is a negative linear function of difficulty. It is of some importance to recognize the dependence of incentive values intrinsic to achievement and failure upon the subjective probability of success. One cannot anticipate the thrill of a great accomplishment if, as a matter of fact, one faces what seems a very easy task. Nor does an individual experience only a minor sense of pride after some extraordinary feat against what seemed to him overwhelming odds. The implications of the scheme which follows rest heavily upon the assumption of such a dependence. In Table 1, values of 1 have been arbitrarily assigned to the achievement motive (Ma) and the motive to avoid failure (Mf). Table 1 contains the strength of motivation to approach success (Ma X Pa X 78) and motivation to avoid failure (M, X Pf X //) through performance of nine different tasks labeled A through I. The tasks differ in degree of difficulty as inferred from the subjective probability of success (/%). The incentive values of success and fail- TABLE 1 AROUSED MOTIVATION TO ACHIEVE (APPROACH) AND TO AVOID FAILURE (AVOIDANCE) AS A JOINT FUNCTION OF MOTIVE (M), EXPECTANCY (P), AND INCENTIVE (/), WHERE /.= (!- P,) AND // = (-P.) Motivation to Achieve Task A TaskB TaskC TaskD TaskE TaskF TaskG TaskH Task I M,X P. X /. – Approach Motivation to Avoid Failure Re8ultant Motivation . ,. (Approach—Avoidance) Avoidance . 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 .10 .20 .30 .40 .50 .60 .70 .80 .90 .90 .80 .70 .60 .50 .40 .30 .20 .10 .09 .16 .21 .24 .25 .24 .21 .16 .09 .90 .80 .70 .60 .50 .40 .30 .20 .10 -.10 -.20 -.30 -.40 -.50 -.60 -.70 -.80 -.90 -.09 -.16 -.21 -.24 -.25 -.24 -.21 -.16 -.09 0 0 0 0 0 0 0 0 0 RISK-TAKING BEHAVIOR ure at each of the tasks have been calculated directly from the assumptions that incentive value of success equals 1 — P, and that incentive value of failure equals — P8; and P, and Pj are assumed to add to 1.00. Table 1 may be considered an extension of ideas presented in the resultant valence theory of level of aspiration by Escalona and Festinger (12). The present formulation goes beyond their earlier proposals (a) in making specific assumptions regarding the incentive values of success and failure, and (b) in stating explicitly how individual differences in strength of achievement motive and motive to avoid failure influence motivation.8 When the achievement motive is stronger (M3 > M/). The right-hand column of Table 1 shows the resultant motivation for each of the tasks in this special case where achievement motive and motive to avoid failure are equal in strength. In every case there is an approach-avoidance conflict with resultant motivation equal to 0. This means that if the achievement motive were stronger than the motive to avoid failure—for example, if we assigned Ms a value of 2—the resultant motivation would become positive for each of the tasks and its magnitude would be the same as in the column labeled Approach. Let us therefore consider only the strength of 8 In the resultant valence theory of level of aspiration, the resultant force (/*) for a particular level of difficulty equals probability of success (P,) times valence of success (Vat) minus probability of failure (Pi) times valence of failure (Vat). It is assumed that the valence of a goal IVa(G)] depends partly on the properties of the activity and specific goal (G) and partly on the state of need [f(G)] of the person, IVa(G) = F(G, t(G)] (11, p. 273). In the present conception, the relative rewarding or punishing properties of specific goals (i.e., incentives) and the more general disposition of the person toward a class of incentives (i.e., his motive) are given independent status. 363 approach motivation for each of the tasks, to see the implications of the model for the person in whom the need for achievement is stronger than his disposition to avoid failure. One thing is immediately apparent. Motivation to achieve is strongest when uncertainty regarding the outcome is greatest, i.e., when P, equals .50. If the individual were confronted with all of these tasks and were free to set his own goal, he should choose Task E where P, is .SO, for this is the point of maximum approach motivation. The strength of motivation to approach decreases as P8 increases from .50 to near certainty of success (P, = .90), and it also decreases as P, decreases from .50 to near certainty of failure (P, = .10). If this person were to be confronted with a single task in what is here called the constrained performance situation, we should expect him to manifest strongest motivation in the performance of a task of intermediate difficulty where P, equals .50. If presented either more difficult tasks or easier tasks, the strength of motivation manifested in performance should be lower. The relationship between strength of motivation as expressed in performance level and expectancy of success at the task, in other words, should be described by a bell-shaped curve. When the motive to avoid failure is stronger (Mf>M,). Let us now ignore the strength of approach motivation and tentatively assign it a value of 0, in order to examine the implications of the model for any case in which the motive to avoid failure is the stronger motive. The resultant motivation for each task would then correspond to the values listed in the column labeled Avoidance. What should we expect of the person in whom the disposition to avoid failure is stronger than the motive to achieve? It is apparent at once that 364 JOHN W. ATKINSON the resultant motivation for every task would be negative for him. This person should want to avoid all of the tasks. Competitive achievement situations are unattractive to him. If, however, he is constrained (e.g., by social pressures) and asked to set his level of aspiration, he should avoid tasks of intermediate difficulty (Ps = .50) where the arousal of anxiety about failure is greatest. He should choose either the easiest (Pa = .90) or the most difficult task (Pa = .10). The strength of avoidant motivation is weakest at these two points. In summary, the person in whom the achievement motive is stronger should set his level of aspiration in the intermediate zone where there is moderate risk. To the extent that he has any motive to avoid failure, this means that he will voluntarily choose activities that maximize his own anxiety about failure! On the other hand, the person in whom the motive to avoid failure is stronger should select either the easiest of the alternatives or should be extremely speculative and set his goal where there is virtually no chance for success. These are activities which minimize his anxiety about failure. How does the more fearful person behave when offered only a specific task to perform? He can either perform the task or leave the field. If he chooses to leave the field, there is no problem. But if he is constrained, as he must be to remain in any competitive achievement situation, he will stay at the task and presumably work at it. But how hard will he work at it? He is motivated to avoid failure, and when constrained, there is only one path open to him to avoid failure—success at the task he is presented. So we expect him to manifest the strength of his motivation to avoid failure in performance of the task. He, too, in other words, should try hardest* when P, is .50 and less hard when the chance of winning is either greater or less. The 50-50 alternative is the last he would choose if allowed to set his own goal, but once constrained he must try hard to avoid the failure which threatens him. Not working at all will guarantee failure of the task. Hence, the thought of not working at all should produce even stronger avoidant motivation than that aroused by the task itself. In other words, irrespective of whether the stronger motive is to achieve or to avoid failure, the strength of motivation to perform a task when no alternatives are offered and when the individual is constrained should be greatest when Pa is .50. This is the condition of greatest uncertainty regarding the outcome. But when there are alternatives which differ in difficulty, the choice of level of aspiration by persons more disposed to avoid failure is diametrically opposite to that of persons more disposed to seek success. The person more motivated to achieve should prefer a moderate risk. His level of aspiration will fall at the point where his positive motivation is strongest, at the point where the odds seem to be 50-50. The fearful person, on the other hand, must select a task even though all the alternatives are threatening to him. He prefers the least threatening of the available alternatives: either the task which is so easy he cannot fail, or the task which is so difficult that failure would be no cause for self-blame and embarrassment. The tendency for anxious persons to set either extremely high or very low aspirations has been noted over and over 4 1 do not mean to exclude the possibility that the very anxious person may suffer a performance decrement due to the arousal of some “task-irrelevant” avoidant responses, as proposed in the interpretation of research which has employed the Mandler-Sarason Measure of Test Anxiety (16). 365 RISK-TAKING BEHAVIOR .50 1.00 Probability of Success FIG. I . Strength of motivation to achieve or to avoid failure as a function of the subjective probability of success, i.e., the difficulty of the task. again in the literature on level of aspiration (12). Typically, groups of persons for whom the inference of greater anxiety about failure seems justified on the basis of some personality assessment show a much greater variance in level of aspiration than persons whose motivation is inferred to be more normal or less anxious. When the details of behavior are examined, it turns out that they are setting their aspiration level either defensively high or defensively low. Without further assumptions, the theory of motivation which has been presented when applied to competitiveachievement activity implies that the relationship of constrained performance to expectancy of goal-attainment should take the bell-shaped form shown in Fig. 1, whether the predominant motive is to achieve or to avoid failure. Further, the theory leads to the prediction of exactly opposite patterns for setting the level of aspiration when the predominant motivation is approach and when it is avoidant, as shown in Fig. 2. Both of these hypotheses have been supported in recent experiments. The writer5 offered female college students 6 Atkinson, J. W. Towards experimental analysis of human motivation in terms of a modest monetary prize for good performance at two 20-minute tasks. The probability of success was varied by instructions which informed the subject of the number of persons with whom she was in competition and the number of monetary prizes to be given. The stated probabilities were % 0 > %> Vz> and %. The level of performance was higher at the intermediate probabilities than at the extremes for subjects having high thematic apperceptive n Achievement scores, and also for subjects who had low n Achievement scores, presumably a more fearful group. McClelland6 has shown the diametrically opposite tendencies in choice of level of aspiration in studies of children in kindergarten and in the third grade. One of the original level-of-aspiration experiments, the ring-toss experiment, was repeated with five-year-olds, and a nonverbal index of the strength of achieve- Motive to Achieve +85 -25 Motive to Avoid Failure .50 1.00 Probability of Success FIG. 2. Relative attractiveness of tasks which differ in subjective probability of success (i.e., in difficulty). The avoidance curve has been inverted to show that very difficult and very easy tasks arouse less fear of failure and hence are less unattractive than moderately difficult tasks. motives, expectancies, and incentives. To appear in Motives in fantasy, action, and society. Princeton: Van Nostrand (in preparation). « McClelland, D. C. Risk-taking in children with high and low need for achievement. To appear in Motives in fantasy, action, and society. Princeton: Van Nostrand (in preparation) . 366 JOHN W. ATKINSON ment motive was employed. Children who were high in n Achievement more frequently set their level of aspiration in the intermediate range of difficulty. They took more shots from a modest distance. Children who were low in n Achievement showed a greater preponderance of choices at the extreme levels of difficulty. They more often stood right on top of the peg or stood so far away that success was virtually impossible. The same difference between high and low « Achievement groups was observed on another task with children in the third grade. McClelland views these results as consistent with his theoretical argument concerning the role of achievement motivation in entrepreneurship and economic development (14). He has called attention to the relationship between achievement motivation and an interest in enterprise which requires moderate or calculated risks, rather than very safe or highly speculative undertakings. In an experiment designed for another purpose, Clark, Teevan, and Ricciuti (4) have presented results with college students comparable to those of McClelland. Immediately before a final examination in a college course, students were asked a series of questions pertaining to grade expectations, affective reactions to grades, and the grades they would settle for if excused from taking the exam. A number of indices were derived from responses to these questions, by which the students were classified as: hopeful of success, i.e., if the settle-far grade was near the maximum grade the student thought he could possibly achieve; fearful of failure, i.e., if the settle-far grade was near the minimum grade the student thought he might possibly drop to; and intermediate, i.e., if the settle-for grade fell somewhere between these two extremes. Previously obtained n Achievement scores were significantly higher for the intermediate group than for the two groups who set either extremely high or low levels of aspiration. In terms of the model presented in Table 1, the two extreme patterns of aspirant behavior which are here designated hope of success and fear of failure are to be considered two phenotypically dissimilar alternatives that are genotypically similar. That is, they both function to avoid or reduce anxiety for the person in whom the motive to avoid failure is stronger than the motive to achieve. A question may arise concerning the legitimacy of inferring relatively stronger motive to avoid failure from a low n Achievement score in thematic apperception. The inference seems justified on several counts. First, the kind of learning experience which is thought to contribute to the development of a positive motive to achieve (15, 22) seems incompatible with the kind of experience which would contribute to the development of an avoidant motive. In any specific early learning experience in which successful independent accomplishment is encouraged and rewarded, it seems impossible for incompetence, at the same time, to be punished. Second, even if it is assumed that high and low n Achievement groups may be equal in the disposition to be fearful of failure, the fact that one group does not show evidence of a strong motive to achieve (the group with low n Achievement scores) suggests that fear of failure should be relatively stronger in that group than in the group which does show evidence of strong n Achievement (high n Achievement scores). Finally, Raphelson (19) has presented evidence that n Achievement, as measured in thematic apperception, is negatively related to both scores on the MandlerSarason Scale of Test Anxiety and a psychogalvanic index of manifest anxiety obtained in a test situation. Test 367 RISK-TAKING BEHAVIOR anxiety scores and the psychogalvanic index of manifest anxiety were positively correlated, as they should be if each is an effective measure of fear aroused in a competitive situation. Although a low n Achievement score can hardly be viewed as a direct index of the disposition to avoid failure, there seems good presumptive evidence that fear of failure is relatively stronger than the achievement motive in such a group. And this presumption is all the theory demands to explain the pattern of goal setting which focuses upon the extremes in the range of difficulty among persons low in « Achievement. The details of the exploratory experiments suggest that one further assumption be made. In both experiments, the high n Achievement groups showed evidence of maximum motivation when the observed or stated probability of success was approximately .33. At this point, the high n Achievement group showed the highest level of constrained performance. And this point was most favored by the high n Achievement group in setting level of aspiration in the McClelland experiment. The assumption to be made seems a reasonable one: the relative strength of a motive influences the subjective probability of the consequence consistent with that motive— i.e., biases it upwards. In other words, the stronger the achievement motive relative to the motive to avoid failure, the higher the subjective probability of success, given stated odds. The stronger the motive to avoid failure relative to the achievement motive, the higher the subjective probability of failure, given stated odds or any other objective basis for inferring the strength of expectancy. Some evidence from two earlier studies is pertinent. When subjects stated the score that they expected to make on a test with very ambiguous or conflicting cues from past performance (15, p. 247) or when faced with a novel task at Motive to Achieve Motive to Avoid Failure .50 Objective Probability of Success 1.00 FIG. 3. Strength of motivation to achieve and to avoid failure as a function of the objective probability of success. It is assumed that the subjective probability of the consequence consistent with the stronger motive is biased upwards. which they had no experience (18), the stated level of expectation was positively related to n Achievement. The biasing effect of the motive on subjective probability should diminish with repeated learning experience in the specific situation. When this assumption is made, the point of maximum motivation to achieve now occurs where the stated (objective) odds are somewhat lower than .SO; and the point of maximum motivation to avoid failure occurs at a point somewhat higher than stated odds of .50, as shown in Fig. 3. The implications of this assumption for constrained performance in somewhat novel situations are evident in the figure. When the achievement motive is stronger than the motive to avoid failure, there should be a tendency for stronger motivation to be expressed in performance when the objective odds are long, i.e., below .50. When the motive to avoid failure is stronger than the achievement motive, there should be greater motivation expressed when the objective odds are short, i.e., above .50. The effects of success and failure. Let us return to the model and ask, What are the effects of success and fail- 368 JOHN W. ATKINSON ure on the level of motivation? We may refer back to Table 1 to answer this question. First, let us consider the effects of success or failure on the level of motivation in a person whose motive to achieve is stronger than his motive to avoid failure. In the usual level-ofaspiration situation, he should initially set his goal where Ps equals .SO. In Table 1, this is Task E. If he succeeds at the task, Ps should increase. And, assuming that the effects of success and failure generalize to similar tasks, the P8 at Task D which was initially .40 should increase toward .50. On the next trial, P3 at Task E is now greater than .50, and Ps at Task D now approaches .50. The result of this change in Pa is diminished motivation to achieve at the old task, E, and increased motivation to achieve at Task D, an objectively more difficult task. The observed level of aspiration should increase in a step-like manner following success, because there has been a change in motivation. A further implication of the change in strength of motivation produced by the experience of success is of great consequence: given a single, very difficult task (e.g., P8 = .10), the effect of continued success in repeated trials is first a gradual increase in motivation as Pa increases to .50, followed by a gradual decrease in motivation as P, increases further to the point of certainty (/>, = 1.00). Ultimately, as P, approaches 1.00, satiation or loss of interest should occur. The task no longer arouses any motivation at all. Why? Because the subjective probability of success is so high that the incentive value is virtually zero. Here is the clue to understanding how the achievement motive can remain insatiable while satiation can occur for a particular line of activity. The strength of motive can remain unchanged, but interest in a particular task can diminish com- pletely. Hence, when free to choose, the person who is stronger in achievement motive should always look for new and more difficult tasks as he masters old problems. If constrained, the person with a strong achievement motive should experience a gradual loss of interest in his work. If the task is of intermediate difficulty to start with (Ps = .50), or is definitely easy (Ps > .50), his interest should begin to wane after the initial experience of success. But what of the effect of failure on the person who is more highly motivated to achieve than to avoid failure? Once more we look at the Approach column of Table 1. If he has chosen Task E (Pa = .50) to start with and fails at it, the Ps is reduced. Continued failure will mean that soon Task F (formerly Pa = .60) will have a Ps near .50. He should shift his interest to this task, which was objectively less difficult in the initial ordering of tasks. This constitutes what has been called a lowering of the level of aspiration. He has moved to the easier task as a consequence of failure. What is the effect of continued failure at a single task? If the initial task is one that appeared relatively easy to the subject (e.g., Pa = .80) and he fails, his motivation should increase! The P6 will drop toward .70, but the incentive value or attractiveness of the task will increase. Another failure should increase his motivation even more. This will continue until the Ps has dropped to .50. Further failure should then lead to a gradual weakening of motivation as P3 decreases further. In other words, the tendency of persons who are relatively strong in achievement motive to persist at a task in the face of failure is probably attributable to the relatively high subjective probability of success, initially. Hence, failure has the effect of increasing the strength of their motivation, at least for a time. Ultimately, RISK-TAKING BEHAVIOR however, interest in the task will diminish if there is continued failure. If the initial task is perceived by the person as very difficult to start with (Pa < .50), motivation should begin to diminish with the first failure. Let us turn now to the effect of success and failure on the motivation of the person who is more strongly disposed to be fearful of failure. If the person in whom the motive to avoid failure is stronger has chosen a very difficult task in setting his level of aspiration (e.g., Task A where Ps = .10) and succeeds, Pa increases and his motivation to avoid the task is paradoxically increased! It would almost make sense for him deliberately to fail, in order to keep from being faced with a stronger threat on the second trial. If there are more difficult alternatives, he should raise his level of aspiration to avoid anxiety! Fortunately for this person, his strategy (determined by the nature of his motivation) in choosing a very difficult task to start with protects him from this possibility, because Ps is so small that he will seldom face the paradoxical problem just described. If he fails at the most difficult task, as is likely, Pa decreases further, Pf increases further, and the aroused motivation to avoid failure is reduced. By continued failure he further reduces the amount of anxiety about failure that is aroused by this most difficult task. Hence, he should continue to set his level at this point. If he plays the game long enough and fails continuously, the probability of failure increases for all levels of difficulty. Sooner or later the minimal motivation to avoid failure at the most difficult task may be indistinguishable from the motivation to avoid failure at the next most difficult task. This may ultimately allow him to change his level of aspiration to a somewhat less difficult task without acting in gross contra- 369 diction to the proposed principle of motivation. If our fearful subject has initially chosen the easiest task (Task I where Pa = .90) and if he fails, Pt decreases toward .80, and his motivation to avoid the task also increases. If there is no easier task, the most difficult task should now appear least unattractive to him, and he should jump from the easiest to the most difficult task. In other words, continued failure at a very easy task decreases Ps toward .SO; and, as Table 1 shows, a change of this sort is accompanied by increased arousal of avoidant motivation. A wild and apparently irrational jump in level of aspiration from very easy to very difficult tasks, as a consequence of failure, might be mistakenly interpreted as a possible effort on the part of the subject to gain social approval by seeming to set high goals. The present model predicts this kind of activity without appealing to some extrinsic motive. It is part of the strategy of minimizing expected pain of failure after one has failed at the easiest task. If our fear-disposed subject is successful at the most simple task, his P8 increases, his Pf decreases, and his motivation to avoid this task decreases. The task becomes less and less unpleasant. He should continue playing the game with less anxiety. Table 1, when taken in its entirety, deals with the special case of the person in whom the two motives are exactly equal in strength. The implications are clear. In the constrained-performance situation, he should work hardest when the probability of success is .SO, because motivation to achieve and motivation to avoid failure will summate in the constrained instrumental act which is at the same time the pathway toward success and away from failure. (This summation should also occur in the cases where one motive is stronger.) But in the level-of-aspiration setting 370 JOHN W. ATKINSON where there is an opportunity for choice among alternatives, the avoidance motivation exactly cancels out the approach motivation. Hence, the resultant motivation for each of the alternatives is zero. His choice of level of aspiration cannot be predicted from variables intrinsic to the achievement-related nature of the task. If there is any orderly pattern in this conflicted person’s level of aspiration, the explanation of it must be sought in extrinsic factors, e.g. the desire to gain social approval. Such a desire can also be conceptualized in terms of motive, expectancy, and incentive, and the total motivation for a particular task can then be attributed to both achievement-related motives and other kinds of motives engaged by the particular features of the situation. In recent years there has been something of a rebirth of interest in the problems of level of aspiration, particularly in pathological groups. The tendency for anxious groups to show much greater variability in level of aspiration, setting their goals either very high or very low relative to less anxious persons, was noted in early studies by Sears, Rotter, and others (12). Miller (17), Himmelweit (9), and Eysenck and Himmelweit (8) have produced substantial evidence that persons with affective disorders (neurasthenia or dysthymia) typically set extremely high goals for themselves; hysterics, on the other hand, show a minimal level of aspiration, often setting their future goal even below the level of past performance. In all of these studies, normal control groups have fallen between these two extremes, as might be expected from the present model if normals are relatively more positive in their motivation in achievement-related situations. In the work of Eysenck (7) and his colleagues, both dysthymics and hysterics show greater neuroticism than nor- mal subjects. Eysenck’s interpretation of this factor as autonomic sensitivity is consistent with the implications of the present model, which attributes the setting of extremely high or low levels of aspiration to relatively strong motivation to avoid failure. A second factor, extroversion-introversion, discriminates the affective disorders and hysterics where the present model, dealing only with motives intrinsic to the competitive achievement situation, does not. An appeal to some other motivational difference, e.g., in strength of n Affiliation, might also predict the difference in pattern of level of aspiration. Probability Preferences The present analysis is relevant to another domain of current research interest, that specifically concerned with the measurement of subjective probability and utility. Edwards (5, 6), for example, has reported probability preferences among subjects offered alternative bets having the same expected value. We 7 have repeated the Edwards type experiment (e.g., 6/6 of winning 30^ versus 1/6 of winning $1.80) with subjects having high and low n Achievement scores. The results show that persons high in n Achievement more often prefer intermediate probabilities (4/6, 3/6, 2/6) to extreme probabilities (6/6, 5/6, 1/6) than do persons low in n Achievement. What is more, the same differential preference for intermediate risk was shown by these same subjects when they were allowed to choose the distance from the target for their shots in a shuffleboard game. In other words, the incentive values of winning qua winning, and losing qua losing, presumably developed in achievement activities early in life, generalize to the gambling 7 Atkinson, J. W., Bastian, J. R., Earl, R. W., and Litwin, G. H. The achievement motive, goal-setting, and probability preferences (in preparation). 371 RISK-TAKING BEHAVIOR situation in which winning is really not contingent upon one’s own skill and competence. Social Mobility Aspirations Finally, the present model may illuminate a number of interesting research possibilities having to do with social and occupational mobility. The ranking of occupations according to their prestige in Western societies clearly suggests that occupations accorded greater prestige are also more difficult to attain. A serious effort to measure the perceived probability of being able to attain certain levels on the occupational ladder should produce a high negative correlation with the usual ranking on prestige. If so, then the present model for level of aspiration, as well as its implications for persons who differ in achievementrelated motives, can be applied to many of the sociological problems of mobility aspirations. A recent paper by Hyman (10) has laid the groundwork for such an analysis. SUMMARY A theoretical model is presented to explain how the motive to achieve and the motive to avoid failure influence behavior in any situation where performance is evaluated against some standard of excellence. A conception of motivation in which strength of motivation is a joint multiplicative function of motive, expectancy (subjective probability), and incentive is offered to account for the selection of one task among alternatives which differ in difficulty (level of aspiration), and also to account for performance level when only one task is presented. It is assumed that the incentive value of success is a positive linear function of difficulty as inferred from the subjective probability of success; and negative incentive value of failure is assumed to be a negative linear function of difficulty. The major im- plications of the theory are (a) that performance level should be greatest when there is greatest uncertainty about the outcome, i.e., when subjective probability of success is .SO, whether the motive to achieve or the motive to avoid failure is stronger within an individual; but (b) that persons in whom the achievement motive is stronger should prefer intermediate risk, while persons in whom the motive to avoid failure is stronger should avoid intermediate risk, preferring instead either very easy and safe undertakings or extremely difficult and speculative undertakings. Results of several experiments are cited, and the implications of the theoretical model for research on probability preferences in gambling and studies of social mobility aspirations are briefly discussed. REFERENCES 1. ATKINSON, J. W. Explorations using imaginative thought to assess the strength of human motives. In M. R. Jones (Ed.), Nebraska Symposium on Motivation, I9S4. Lincoln, Nebr.: Univer. of Nebraska Press, 1954, 2. ATKINSON, J. W., & REITMAN, W. R. Performance as a function of motive strength and expectancy of goal-attainment. /. abnorm. soc. Psychol., 1956, S3, 361-366. 3. BROWN, J. S. Problems presented by the concept of acquired drives. In Current theory and research in motivation. Lincoln, Nebr.: Univer. of Nebraska Press, 1953. 4. CLARK, R. A., TEEVAN, R., & Riccnm, H. N. Hope of success and fear of failure as aspects of need for achievement. /. abnorm. soc. Psychol., 1956, S3, 182-186. 5. EDWARDS, W. Probability preferences in gambling. Amer. J. Psychol., 1953, 66, 349-364. 6. EDWARDS, W. The theory of decision making. Psychol. Bull., 1954, 51, 380-417. 7. EYSENCK, H. J. A dynamic theory of anxiety and hysteria. /. ment. Sci., 1955, 101, 28-51. 8. EYSENCK, H. J., & HIMMELWEIT, H. T. An experimental study of the reactions 372 JOHN W. ATKINSON of neurotics to experiences of success and failure. J. gen. Psychol., 1946, 35, 59-75. 9. HIMMELWEIT, H. T. A comparative study of the level of aspiration of normal and neurotic persons. Brit, J. Psychol., 1947, 37, 41-59. 10. HYMAN, H. H. The value systems of different classes: a social psychological contribution to the analysis of stratification. In R. Bendix & S. M. Lipset (Eds.), Class, status, and power. Glencoe, III: Free Press, 1953. 11. LEWIN, K. Field theory in social science. D. Cartwright (Ed.). New York: Harper Bros., 1951. 12. LEWIN, K., DEMBO, T., FESTINOER, L., & SEASS, P. S. Level of aspiration. In J. McV. Hunt (Ed.), Personality and the behavior disorders, Vol. 1, Chap. 10. New York: Ronald Press, 1944. 13. MCCLELLAND, D. C. Personality. New York: William Sloane Associates, 1951. 14. MCCLELLAND, D. C. Some social consequences of achievement motivation. In M. R. Jones (Ed.), Nebraska symposium on motivation, 1955. Lincoln: Univer. of Nebraska Press, 1955. 15. MCCLELLAND, D. C., ATKINSON, J. W., CLARK, R. A., & LOWELL, E. L. The achievement motive. New York: Appleton-Century-Crofts, 1953. 16. MANDLER, G., & SARASON, S. B. A study of anxiety and learning. /. dbnorm. soc. Psychol., 1952, 16, 115-118. 17. MILLER, D. R. Responses of psychiatric patients to threat of failure. J. abnorm. soc. Psychol, 1951, 46, 378-387. 18. POTTHARST, B. C. The achievement motive and level of aspiration after experimentally induced success and failure. Unpublished doctor’s dissertation, Univer. of Michigan, 1956. 19. RAPHELSON, A. Imaginative and direct verbal measures of anxiety related to physiological reactions in the competitive achievement situation. Unpublished doctor’s dissertation, Univer. of Michigan, 1956. 20. ROTTER, J. B. Social learning and clinical psychology. New York: Prentice-Hall, 1954. 21. TOLMAN, E. C. Principles of performance. Psychol. Rev., 1955, 62, 315-326. 22. WINTERBOTTOM, M. R. The relation of childhood training in independence to achievement motivation. Unpublished doctor’s dissertation, Univer. of Michigan, 1952. (Received February 7, 1957) Human Agency in Social Cognitive Theory This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly. Albert Bandura ABSTRACT: The present article examines the nature and function of human agency within the conceptual model of triadic reciprocal causation. In analyzing the operation of human agency in this interactional causal structure, social cognitive theory accords a central role to cognitive, vicarious, self-reflective, and self-regulatory processes. The issues addressed concern the psychological mechanisms through which personal agency is exercised, the hierarchical structure of self-regulatory systems, eschewal of the dichotomous construal of self as agent and self as object, and the properties of a nondualistic but nonreductional conception of human agency. The relation of agent causality to the fundamental issues of freedom and determinism is also analyzed. The recent years have witnessed a resurgence of interest in the self-referent phenomena. One can point to several reasons why self processes have come to pervade many domains of psychology. Self-generated activities lie at the very heart of causal processes. They not only contribute to the meaning and valence of most external influences, but they also function as important proximal determinants of motivation and action. The capacity to exercise control over one’s own thought processes, motivation, and action is a distinctively human characteristic. Because judgments and actions are partly self-determined, people can effect change in themselves and their situations through their own efforts. In this article, I will examine the mechanisms of human agency through which such changes are realized. The Nature and Locus of Human Agency The manner in which human agency operates has been conceptualized in at least three different ways–as either autonomous agency, mechanical agency, or emergent interactive agency. The notion that humans serve as entirely independent agents of their own actions has few, if any, serious advocates. However, environmental determinists sometimes invoke the view of autonomous agency in arguments designed to repudiate any role of self-influence in causal processes. A second approach to the self system is to treat it in terms of mechanical agency. It is an internal instrumentality through which external influences operate mechanistically on action, but it does not itself have any motivative, self-reflective, self-reactive, creative, or selfdirective properties. In this view, internal events are mainly products of external ones devoid of any causal efficacy. Because the agency resides in environmental September 1989 • American Psychologist Copyright 1989 by the American Psychological Association, Inc. 0003-066X/89/$00.75 Vol. 44, No. 9, 1175-1184 Stanford University forces, the self system is merely a repository and conduit for them. In this conception of agency, self-referent processes are epiphenominal by-products of conditioned responses that do not enter into the determination of action. For the material eliminativist, self-influences do not exist. People are not intentional cognizers with a capacity to influence their own motivation and action; rather, they are neurophysiological computational machines. Such views fail to explain the demonstrable explanatory and predictive power of self-referent factors that supposedly are devoid of causal efficacy or do not even exist. Social cognitive theory subscribes to a model of emergent interactive agency (Bandura, 1986). Persons are neither autonomous agents nor simply mechanical conveyers of animating environmental influences. Rather, they make causal contribution to their own motivation and action within a system of triadic reciprocal causation. In this model of reciprocal causation, action, cognitive, affective, and other personal factors, and environmental events all operate as interacting determinants. Any account of the determinants of human action must, therefore, include self-generated influences as a contributing factor. Empirical tests of the model of triadic reciprocal causation are presented elsewhere and will not be reviewed here (Wood & Bandura, in press). The focus of this article is on the mechanisms through which personal agency operates within the interactional causal structure. Exercise of Agency Through Self-Belief of Efficacy Among the mechanisms of personal agency, none is more central or pervasive than people’s beliefs about their capabilities to exercise control over events that affect their lives. Self-efficacy beliefs function as an important set of proximal determinants of human motivation, affect, and action. They operate on action through motivational, cognitive, and affective intervening processes. Some of these processes, such as affective arousal and thinking patterns, are of considerable interest in their own right and not just as intervening influencers of action. Cognitive Processes Self-efficacy beliefs affect thought patterns that may be self-aiding or self-hindering. These cognitive effects take various forms. Much human behavior is regulated by forethought embodying cognized goals, and personal goal setting is influenced by self-appraisal of capabilities. The stronger their perceived self-efficacy, the higher the goals people set for themselves and the firmer their commitment 1175 This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly. to them (Locke, Frederick, Lee, & Bobko, 1984; Taylor, Locke, Lee, & Gist, 1984; Wood & Bandura, in press). As I will show later, challenging goals raise the level of motivation and performance attainments (Locke, Shaw, Saari, & Latham, 1981; Mento, Steel, & Karren, 1987). A major function of thought is to enable people to predict the occurrence of events and to create the means for exercising control over those that affect their daily lives. Many activities involve inferential judgments about conditional relations between events in probabilistic environments. Discernment of predictive rules requires cognitive processing of multidimensional information that contains many ambiguities and uncertainties. In ferreting out predictive rules, people must draw on their state of knowledge to generate hypotheses about predictive factors, to weight and integrate them into composite rules, to test their judgments against outcome information, and to remember which notions they had tested and how well they had worked. It requires a strong sense of efficacy to remain task oriented in the face of judgmental failures. Indeed, people who believe strongly in their problemsolving capabilities remain highly efficient in their analytic thinking in complex decision-making situations, whereas those who are plagued by self-doubts are erratic in their analytic thinking (Bandura & Wood, 1989; Wood & Bandura, 1989). Quality of analytic thinking, in turn, affects performance accomplishments. People’s perceptions of their efficacy influence the types of anticipatory scenarios they construct and reiterate. Those who have a high sense of efficacy visualize success scenarios that provide positive guides for performance. Those who judge themselves as inefficacious are more inclined to visualize failure scenarios that undermine performance by dwelling on how things will go wrong. Cognitive simulations in which individuals visualize themselves executing activities skillfully enhance subsequent performance (Bandura, 1986; Corbin, 1972; Feltz & Landers, 1983; Kazdin, 1978; Markus, Cross, & Wurf, in press). Perceived self-efficacy and cognitive simulation affect each other bidirectionally. A high sense of efficacy fosters cognitive constructions of effective actions, and cognitive reiteration of efficacious courses of action strengthens self-perceptions of efficacy (Bandura & Adams, 1977; Kazdin, 1979), Self-efficacy beliefs usually affect cognitive functioning through the joint influence of motivational and information-processing operations. This dual influence is illustrated in studies of different sources of variation in memory performance. The stronger people’s beliefs in their memory capacities, the more effort they devote to cognitive processing of memory tasks, which, in turn, enhances their memory performances (Berry, 1987). Preparation of this article was facilitatedby Public Health Research Grant No. MH-5162-25 fromthe NationalInstitutefor MentalHealth. Thisarticlewaspresentedas an invitedaddressat the XXIVInternational Congress of Psychology,Sydney,Australia,August 1988. Correspondenceconcerningthisarticleshouldbe addressedto Albert Bandura,Building420, JordanHall,StanfordUniversity,Stanford, CA 94305. 1176 Motivational Processes People’s self-efficacy beliefs determine their level of motivation, as reflected in how much effort they will exert in an endeavor and how long they will persevere in the face of obstacles. The stronger the belief in their capabilities, the greater and more persistent are their efforts (Bandura, 1988a). When faced with difficulties, people who are beset by self-doubts about their capabilities slacken their efforts or abort their attempts prematurely and quickly settle for mediocre solutions, whereas those who have a strong belief in their capabilities exert greater effort to master the challenge (Bandura & Cervone, 1983, 1986; Cervone & Peake, 1986; Jacobs, Prentice-Dunn, & Rogers, 1984; Weinberg, Gould, & Jackson, 1979). Strong perseverance usually pays off in performance accomplishments. There is a growing body of evidence that human attainments and positive well-being require an optimistic sense of personal efficacy (Bandura, 1986). This is because ordinary social realities are strewn with difficulties. They are full of impediments, failures, adversities, setbacks, frustrations, and inequities. People must have a robust sense of personal efficacy to sustain the perseverant effort needed to succeed. Self-doubts can set in quickly after some failures or reverses. The important matter is not that difficulties arouse self-doubt, which is a natural immediate reaction, but the speed of recovery of perceived self-efficacy from difficulties. Some people quickly recover their self-assurance; others lose faith in their capabilities. Because the acquisition of knowledge and competencies usually requires sustained effort in the face of difficulties and setbacks, it is resiliency of self-belief that counts. In his revealing book, titled Rejection, John White (1982) provides vivid testimony that the striking characteristic of people who have achieved eminence in their fields is an inextinguishable sense of efficacy and a firm belief in the worth of what they are doing. This resilient self-belief system enabled them to override repeated early rejections of their work. A robust sense of personal efficacy provides the needed staying power. Many of our literary classics brought their authors repeated rejections. The novelist, Saroyan, accumulated several thousand rejections before he had his first literary piece published. Gertrude Stein continued to submit poems to editors for about 20 years before one was finally accepted. Now that is invincible self-efficacy. Such extraordinary persistence in the face of massive uninterrupted rejection defies explanation in terms of either reinforcement theory or utility theory. James Joyce’s book, the Dubliners, was rejected by 22 publishers. Over a dozen publishers rejected a manuscript by e. e. cummings. When his mother finally published it, the dedication, printed in upper case, read: “With no thanks t o . . . ” followed by the long list of publishers who had rejected his offering. Early rejection is the rule, rather than the exception, in other creative endeavors. The Impressionists had to arrange their own art exhibitions because their works were routinely rejected by the Paris Salon. Van Gogh sold only one painting during his lifetime. Rodin was repeatedly September 1989 • American Psychologist This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly. rejected by the Ecole des Beaux-Arts. The musical works of most renowned composers were initially greeted with derision. Stravinsky was run out of Paris by an enraged audience and critics when he first served them the Rite of Spring. Many other composers suffered the same fate, especially in the early phases of their career. The brilliant architect, Frank Lloyd Wright, was one of the more widely rejected architects during much of his career. To turn to more contemporary examples, Hollywood initially rejected the incomparable Fred Astaire for being only “a balding, skinny actor who can dance a little.” Decca Records turned down a recording contract with the Beatles with the nonprophetic evaluation, “We don’t like their sound. Groups of guitars are on their way out.” Whoever issued that rejective pronouncement must cringe at each sight of a guitar. It is not uncommon for authors of scientific classics to experience repeated initial rejection of their work, often with hostile embellishments if it is too discrepant from the theories in vogue at the time. For example, John Garcia, who eventually won well-deserved recognition for his fundamental psychological discoveries, was once told by a reviewer of his oft-rejected manuscripts that one is no more likely to find the phenomenon he discovered than bird droppings in a cuckoo clock. Verbal droppings of this type demand tenacious self-belief to continue the tortuous search for new Muses. Scientists often reject theories and technologies that are ahead of their time. Because of the cold reception given to most innovations, the time between discovery and technical realization typically spans several decades. It is widely believed that misjudgment produces dysfunction. Certainly, gross miscalculation can create problems. However, optimistic self-appraisals of capability that are not unduly disparate from what is possible can be advantageous, whereas veridical judgments can be selflimiting. When people err in their self-appraisals, they tend to overestimate their capabilities. This is a benefit rather than a cognitive failing to be eradicated. If selfefficacy beliefs always reflected only what people could do routinely, they would rarely fail but they would not mount the extra effort needed to surpass their ordinary performances. Evidence suggests that it is often the so-called normals who are distorters of reality, but they exhibit selfenhancing biases that distort appraisals in the positive direction. The successful, the innovative, the sociable, the nonanxious, the nondespondent, and the social reformers take an optimistic view of their personal efficacy to exercise influence over events that affect their lives (Bandura, 1986; Taylor & Brown, 1988). If not unrealistically exaggerated, such self-beliefs foster the perseverant effort needed for personal and social accomplishments. The findings of laboratory studies are in accord with the records of human triumphs regarding the centrality of the motivational effects of self-beliefs of efficacy in human attainments. It takes a resilient sense of efficacy to override the numerous dissuading impediments to significant accomplishments. September 1989 • American Psychologist Affective Processes People’s beliefs in their capabilities affect how much stress and depression they experience in threatening or taxing situations, as well as their level of motivation. Such emotional reactions can affect action both directly and indirectly by altering the nature and course of thinking. Threat is not a fixed property of situational events, nor does appraisal of the likelihood of aversive happenings rely solely on reading external signs of danger or safety. Rather, threat is a relational property concerning the match between perceived coping capabilities and potentially aversive aspects of the environment. People who believe they can exercise control over potential threats do not conjure up apprehensive cognitions and, therefore, are not perturbed by them. But those who believe they cannot manage potential threats experience high levels of stress and anxiety arousal. They tend to dwell on their coping deficiencies and view many aspects of their environment as fraught with danger. Through such inefficacious thought they distress themselves and constrain and impair their level of functioning (Bandura, 1988b, 1988c; Lazarus & Folkman, 1984; Meichenbaum, 1977; Sarason, 1975). That perceived coping efficacy operates as a cognitive mediator of anxiety has been tested by creating different levels of perceived coping efficacy and relating them at a microlevel to different manifestations of anxiety. Perceived coping inefficacy is accompanied by high levels of subjective distress, autonomic arousal, and plasma catecholamine secretion (Bandura, Reese, & Adams, 1982; Bandura, Taylor, Williams, Mefford, & Barchas, 1985). The combined results from the different psychobiological manifestations of emotional arousal are consistent in showing that anxiety and stress reactions are low when people cope with tasks in their perceived self-efficacy range. Self-doubts in coping efficacy produce substantial increases in subjective distress and physiological arousal. After perceived coping efficacy is strengthened to the maximal level, coping with the previously intimidating tasks no longer elicits differential psychobiological reactions. Anxiety arousal in situations involving some risks is affected not only by perceived coping efficacy but also by perceived self-efficacy to control intrusive perturbing cognitions. The exercise of control over one’s own consciousness is summed up well in the proverb: “You cannot prevent the birds of worry and care from flying over your head. But you can stop them from building a nest in your head.” Perceived self-efficacy in thought control is a key factor in the regulation of cognitively generated arousal. It is not the sheer frequency of aversive cognitions but the perceived inefficacy to turn them offthat is the major source of distress (Kent, 1987; Salkovskis & Harrison, 1984). Thus, the incidence of aversive cognitions is unrelated to anxiety level when variations in perceived thought control efficacy are controlled for, whereas perceived thought control efficacy is strongly related to anxiety level when the extent of aversive cognitions is controlled (Kent & Gibbons, 1987). 1177 This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly. The role of perceived self-efficacyand anxiety arousal in the causal structure ofavoidant behavior has also been examined extensively. The results show that people base their actions on self-perceptions of coping efficacy in situations they regard as risky. The stronger the perceived coping efficacy, the more venturesome the behavior, regardless of whether self-perceptions of efficacy are enhanced through mastery experiences, modeling influences, or cognitive simulations (Bandura, 1988b). Perceived self-efficacy accounts for a substantial amount of variance in phobic behavior when anticipated anxiety is partialed out, whereas the relationship between anticipated anxiety and phobic behavior essentially disappears when perceived self-efficacy is partialed out (Williams, Dooseman, & Kleifield, 1984; Williams, Kinney, & Falbo, in press; Williams, Turner, & Peer, 1985). In short, people avoid potentially threatening situations and activities, not because they are beset with anxiety, but because they believe they will be unable to cope with situations they regard as risky. They take self-protective action regardless of whether they happen to be anxious at the moment. The dual control of anxiety arousal and avoidant behavior by perceived coping efficacy and thought control efficacy is revealed in analyses of the mechanisms governing personal empowerment over pervasive social threats (Ozer & Bandura, 1989). One path of influence is mediated through the effects of perceived coping self-efficacy on perceived vulnerability and risk discernment, and the other through the impact of perceived cognitive control self-efficacy on intrusive aversive thoughts. Perceived self-inefficacy to fulfill desired goals that affect evaluation of one’s self-worth and to secure things that bring satisfaction to one’s life can give rise to bouts of depression (Bandura, 1988a; Cutrona & Troutman, 1986; Holahan & Holahan, 1987a, 1987b; Kanfer & Zeiss, 1983). When the perceived self-inefficacy involves social relationships, it can induce depression both directly and indirectly by curtailing the cultivation of interpersonal relationships that can provide satisfactions and buffer the effects of chronic daily stressors (Holahan & Holahan, 1987a). Depressive rumination not only impairs ability to initiate and sustain adaptive activities, but it further diminishes perceptions of personal efficacy (Kavanagh & Bower, 1985). Much human depression is also cognitively generated by dejecting ruminative thoughts (NolenHoeksema, 1987). Therefore, perceived self-inefficacy to exercise control over ruminative thought figures prominently in the occurrence, duration, and recurrence of depressive episodes (Kavanagh & Wilson, 1988). Other efficacy-activated processes in the affective domain concern the impact of perceived coping efficacy on basic biological systems that mediate health functioning (Bandura, in press-a). Stress has been implicated as an important contributing factor to many physical dysfunctions. Controllability appears to be a key organizing principle regarding the nature of these stress effects. Exposure to physical stressors with a concomitant ability to control them has no adverse physiological effects, whereas exposure to the same stressors without the ability to con1178 trol them impairs cellular components of the immune system (Coe & Levine, in press; Maier, Laudenslager, & Ryan, 1985). Biological systems are highly interdependent. The types of biochemical reactions that have been shown to accompany perceived coping inefficacy are involved in the regulation of immune systems. For example, perceived self-inefficacy in exercising control over cognitive stressors activates endogenous opioid systems (Bandura, Cioffi, Taylor, & Brouillard, 1988). There is evidence that some of the immunosuppressive effects of inefficacy in controlling stressors are mediated by release of endogenous opioids. When opioid mechanisms are blocked by opiate antagonists, the stress of coping inefficacy loses its immunosuppressive power (Shavit & Martin, 1987). In the laboratory research demonstrating immunosuppression through stress mediation, controllability is studied as a fixed dichotomous property in which animals either exercise complete control over physical stressors, or they have no control whatsoever. In contrast, most human stress is activated in the course of learning how to exercise control over recurring cognitive and social stressors. It would not be evolutionarily advantageous if acute stressors invariably impaired immune function, because of their prevalence in everyday life. Indeed, in a recently completed project, my colleagues and I found (Wiedenfeld et al., 1989) that stress aroused in the process of gaining coping efficacy over stressors enhances immune function. The rate of efficacy acquisition is a good predictor of whether exposure to acute stressors enhances or suppresses immune function. Selection Processes People can exert some influence over their life course by their selection of environments and construction of environments. So far, the discussion has centered on efficacyactivated processes that enable people to create beneficial environments and to exercise control over them. Judgments of personal efficacy also affect selection of environments. People tend to avoid activities and situations they believe exceed their coping capabilities, but they readily undertake challenging activities and select social environments they judge themselves capable of handling. Any factor that influences choice behavior can profoundly affect the direction of personal development because the social influences operating in the environments that are selected continue to promote certain competencies, values, and interests long after the decisional determinant has rendered its inaugurating effect. Thus, seemingly inconsequential determinants can initiate selective associations that produce major and enduring personal changes (Bandura, 1986; Snyder, 1986). The power of self-efficacy beliefs to affect the course of life paths through selection processes is clearly revealed in studies of career decision-making and career development (Betz & Hackett, 1986; Lent & Hackett, 1987). The more efficacious people judge themselves to be, the wider the range of career options they consider appropriate and the better they prepare themselves educationally for September 1989 • American Psychologist This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly. different occupational pursuits. Self-limitation of career development arises more from perceived self-inefficacy than from actual inability. By constricting choice behavior that can cultivate interests and competencies, self-disbeliefs create their own validation. It should be noted that the sociocognitive benefits of a sense of personal efficacy do not arise simply from the incantation of capability. Saying something should not be confused with believing it to be so. Simply saying that one is capable is not necessarily self-convincing, especially when it contradicts preexisting firm beliefs. No amount of reiteration that I can fly will persuade me that I have the efficacy to get myself airborne and to propel myself through the air. Action tendencies vary with the strength of self-beliefs of efficacy (Bandura, 1977). Efficacy beliefs exhibit a gradient of strength as a function of temporal and physical proximity to the relevant activity. One must consider the height and slope of the efficacy gradient and the threshold strength for acting on one’s self-belief. These characteristics of a self-belief system are affected by the authenticity of the efficacy information on which they are based. Self-efficacy beliefs that are firmly established are likely to remain strong regardless of whether one is far removed from the taxing or threatening activities or is about to perform them. Such beliefs are resilient to adversity. In contrast, weakly held self-beliefs are highly vulnerable to change: Self-doubts mount the nearer one gets to the taxing activities (Kent, 1987; Kent & Gibbons, 1987), and negative experiences readily reinstate selfdisbelief in one’s capabilities. Efficacy beliefs are the product of a complex process of self-persuasion that relies on cognitive processing of diverse sources of efficacy information. These include performance mastery experiences, vicarious experiences for judging capabilities in comparison with performances of others, verbal persuasion and allied types of social influences indicating that one possesses certain capabilities; and physiological states from which one may partly judge one’s capabilities, strength, and vulnerability. Information that is relevant for judging personal capabilities is not inherently enlightening. Rather, in the self-appraisal of efficacy these different sources of efficacy information must be cognitively processed, weighed, and integrated through self-reflective thought. Acting on one’s self-efficacy judgment produces confirming or disconfirming experiences that prompt further reappraisals of personal efficacy. Development of resilient self-efficacy requires some experience in mastering difficulties through perseverant effort. If people experience only easy successes, they come to expect quick results and their sense of efficacy is easily undermined by failure. Some setbacks and difficulties in human pursuits serve a useful purpose in teaching that success usually requires sustained effort. After people become convinced they have what it takes to succeed, they persevere in the face of adversity and quickly rebound from setbacks. By sticking it out through tough times, they emerge from adversity with a stronger sense of efficacy. September 1989 • American Psychologist Exercise of Agency Through Goal Representations Another distinctive human characteristic through which personal agency is exercised is the capacity of forethought. People do not simply react to immediate environmental influences like weathervanes, nor are they mechanically steered by implants from their past. Most human behavior, being purposive, is regulated by forethought. The future time perspective manifests itself in many different ways. People anticipate the likely consequences of their prospective actions, they set goals for themselves, and they plan courses of action likely to produce desired outcomes. Through the exercise of forethought and self-regulative standards, they motivate themselves and guide their actions anticipatorily. Theories that seek to explain human behavior solely as the product of external influences or the remnants of past stimulus inputs present a truncated image of human nature. This is because people possess self-directive capabilities that enable them to exercise some control over their thoughts, feelings, and actions by the consequences they produce for themselves. Psychosocial functioning is, therefore, regulated by an interplay of self-produced and external sources of influence. The capability for intentional and purposive action is rooted in symbolic activity. Future events cannot be causes of current motivation and action because that would entail backward causation in which the effect precedes the cause. However, by being represented cognitively in the present, conceived future events are converted into current motivators and regulators of behavior. Action is motivated and directed by cognized goals rather than drawn by remote aims. Forethought is translated into incentives and guides for action through the aid of selfregulatory mechanisms. Many theories of self-regulation are founded on a negative feedback control model. This type of system functions as a motivator and regulator of action through a discrepancy reduction mechanism. Perceived discrepancy between performance and an internal standard triggets action to reduce the incongruity. In negative feedback control, if performance matches the internal standard the person does nothing. A regulatory process in which matching a standard begets inertness does not characterize human self-motivation. Such a feedback control system would produce circular action that leads nowhere. Nor could people be stirred to action until they receive feedback of a shortcoming. Although comparative feedback is essential in the ongoing regulation of motivation, people can initially raise their level of motivation by adopting goals before they receive any feedback regarding their beginning effort (Bandura & Cervone, 1983). Negative feedback may help to keep them going on a preset course, but from time to time they must transcend the feedback loop to initiate new challenging courses for themselves. Different self-regulatory systems operate in the initiation and continued regulation of motivation. Human self-motivation relies on discrepancy pro1179 This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly. duction as well as on discrepancy reduction. It requires both proactive control and reactive or feedback control. People initially motivate themselves through proactive control by setting themselves valued challenging standards that create a state of disequilibrium and then mobilizing their effort on the basis of anticipatory estimation of what it would take to accomplish them. Feedback control comes into play in subsequent adjustments of effort to achieve desired results. After people attain the standard they have been pursuing, those who have a strong sense of efficacy generally set a higher standard for themselves. The adoption of further challenges creates new motivating discrepancies to be mastered. Similarly, surpassing a standard is more likely to raise aspiration than to lower subsequent performance to conform to the surpassed standard. Self-motivation thus involves a hierarchical dual control process of disequilibrating discrepancy production followed by equilibrating discrepancy reduction. An evaluative executive control system with a proactive component must therefore be superimposed on a negative feedback operation that keeps changing aspirational standards with progressive performance attainments. To capture the complexity of human self-regulation, such an executive control system must be invested with the evaluative agentive properties shown to play an important role in self-directedness. These properties are discussed next. Goals operate largely through self-referent processes, rather than regulating motivation and action directly. These processes provide the links between goals and action. Cognitive motivation based on goal systems is mediated by three types of self-reactive influences: (a) affective self-evaluation, (b) perceived self-efficacy for goal attainment, and (c) ongoing readjustment of internal standards. Goals create motivating involvement in activities by specifying the conditional requirements for positive self-evaluation. People seek self-satisfactions from fulfilling valued goals and are prompted to intensify their efforts by discontent with substandard performances. Perceived self-efficacy is another self-referent factor that plays an influential role in the self-regulation of motivation through goal systems. As previously noted, it is partly on the basis of self-beliefs of efficacy that people choose what challenges to undertake, how much effort to expend in the endeavor, and how long to persevere in the face of difficulties (Bandura, 1986, 1988a). In the face of negative discrepancies between personal standards and attainments, those who are assured of their capabilities heighten their level of effort and perseverance, whereas those who are beset by self-doubts about their capabilities are easily dissuaded by failure. The goals people set for themselves at the outset of an endeavor are subject to change, depending on the pattern and level of progress they are making (Campion & Lord, 1982). They may maintain their original goal, lower their sights, or adopt an even more challenging goal. Thus, the third constituent of self-influence in the ongoing regulation of motivation concerns the readjustment of internal standards in light of one’s attainments. These self-referent influences op1180 erating in concert account for the major share of variation in motivation through goal systems (Bandura & Cervone, 1986). In brief, the agentive properties of a self-motivational control system must include (a) predictive anticipatory control of effort, (b) affective self-evaluative reactions to one’s performances rooted in a value system, (c) selfappraisal of personal efficacy for goal attainment, and (d) self-reflective metacognitive activity concerning the adequacy of one’s efficacy appraisals and the suitability of one’s standard setting. Evaluation of perceived self-efficacy relative to task demands indicates whether the standards being pursued are within attainable bounds or are unrealistically beyond one’s reach. Exercise of Agency Through Anticipated Outcomes The ability to envision the likely outcomes of prospective actions is another way in which anticipatory mechanisms regulate human motivation and action. People strive to gain anticipated beneficial outcomes and to forestall aversive ones. However, the effects of outcome expectancies on performance motivation are partly governed by self-beliefs of efficacy. There are many activities that, if performed well, guarantee valued outcomes, but they are not pursued if people doubt they can do what it takes to succeed (Beck & L u n d , 1981; Betz & Hackett, 1986; Wheeler, 1983). Self-perceived inefficacy can thus nullify the motivating potential of alluring outcome expectations. The degree to which outcome expectations contribute to performance motivation independently of self-efficacy beliefs is partly determined by the structural relation between actions and outcomes in a particular domain of functioning. In activities in which the level of competence dictates the outcomes, the types of outcomes people anticipate depend largely on their beliefs of how well they will be able to perform in given situations. In most social, intellectual, and physical pursuits, those who judge themselves highly efficacious will expect favorable outcomes, whereas those who expect poor performances of themselves will conjure up negative outcomes. When variations in perceived self-efficacy are partialed out, the outcomes expected for given performances do not have much of an independent effect on behavior (Barling & Abel, 1983; Barling & Beattie, 1983; Godding & Glasgow, 1985; Lee, 1984a, 1984b; Williams & Watson, 1985). Expected outcomes contribute to motivation independently of self-efficacy beliefs when outcomes are not completely controlled by quality of performance. This occurs when extraneous factors also affect outcomes, or outcomes are socially tied to a minimum level of performance so that some variations in quality of performance above and below the standard do not produce differential outcomes. Hierarchical Dual Control Mechanisms in the Construction and Regulation of Action As already noted, motivation is self-regulated through the joint influence ofproactive and feedback mechanisms. The same dual control operates in the construction and September 1989 • American Psychologist This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly. regulation of complex patterns of behavior (Bandura, 1986, in press-b). Foresightful conceptions of actions guide the production of appropriate behavior and provide the internal standards for corrective adjustments in the development of behavioral proficiency (Carroll & Bandura, in press). These conceptions are formed on the basis of knowledge gained through observational learning, inferences from exploratory experiences, information conveyed by verbal instruction, and innovative cognitive syntheses of preexisting knowledge. The mechanism for transforming cognition into action operates through a conception-matching process. This involves both transformational and generative operations. Execution of a skill must be constantly varied to suit changing circumstances. Adaptive performance, therefore, requires a generative conception rather than a one-to-one mapping between representation and action. By applying an abstract specification of the activity, people can produce many variations on the skill. Conceptions are rarely transformed into masterful performance on the first attempt. Monitored enactments serve as the vehicle for transforming knowledge into skilled action. Performances are perfected by corrective adjustments during behavior production until a close match is eventually achieved between conception and action (Carroll & Bandura, 1985, 1987). Because errors can produce costly and injurious consequences, the prospects of healthy survival would be bleak if people had to rely solely on negative feedback to develop competencies. Negative feedback operates as a complementary but subordinate mechanism in the process of action construction. Dual control is similarly involved in the regulation of preestablished modes of action. Forethought guides the selection of actions, and the results produced by those actions verify the adequacy of the chosen course. A system of self-regulation combining proactive guidance with reactive adjustments is best suited for adaptive functioning, especially under changing circumstances. Psychological theories that rely exclusively on a negative feedback model provide only a fractional view of human self-regulation. Human action is, of course, regulated by multilevel systems of control. Cognitive guidance is critical during the acquisition of competences (Carroll & Bandura, in press). But after skills have been perfected, they no longer require cognitive control. Their execution is largely regulated by lower level sensorimotor systems (Carroll & Bandura, 1987). Partial disengagement of thought from proficient action frees cognitive resources for other purposes. If routinized behavior fails to produce expected results, the cognitive control system again comes into play. New courses of action are constructed and tested. Control reverts to the lower control system after an adequate means is found and becomes the habitual way of doing things. The Power of Forethought to Override Feedback Control Human adaptation and survival depend increasingly on the power of forethought to override immediate feedback September 1989 • American Psychologist control of action. We now possess the capacity to create technologies that can have pervasive effects not only on current life but also on that of future generations. Many technical innovations that provide current benefits also entail hazards and cumulative harmful effects that can eventually take a heavy future toll on human beings and the environment. The capacity to extrapolate future consequences from known facts enables people to take corrective actions to avert disastrous futures. It is the expanded time perspective and symbolization of futures afforded by cognition that increase the prospects of human survival. Had humans been ruled solely by immediate consequences, they would have long ago destroyed most of the ecological supports of life. Forethought often saves us from the perils of a foreshortened perspective. However, the power of anticipative control must be enhanced by developing better methods for forecasting distal consequences and stronger social mechanisms for bringing projected consequences to bear on current behavior to keep us off self-destructive courses. Distinction Between Self as Agent and as Object Social cognitive theory rejects the dichotomous conception of self as agent and self as object. Acting on the environment and acting on oneself entail shifting the perspective of the same agent rather than reifying different selves regulating each other or transforming the self from agent to object. In acting as agents over their environments, people draw on their knowledge and cognitive and behavioral skills to produce desired results. In acting as agents over themselves, people monitor their actions and enlist cognitive guides and self-incentives to produce desired personal changes. They are just as much agents influencing themselves as they are influencing their environment. The same is true for metacognitive activity. In their everyday transactions, people act on their thoughts and later analyze how well their thoughts have served them in managing events. However, the same person is doing the operative thinking and later evaluating the adequacy of his or her knowledge, thinking skills, and action strategies. The shift in perspective does not transform an individual from an agent to an object. One is just as much an agent reflecting on one’s experiences as in generating and executing the original courses of action. The same self performing multiple functions does not require creating multiple selves endowed with different roles. Human Agency and Psychoneural Processes Human agency does not imply psychophysical dualism. Thoughts are higher brain processes rather than psychic entities that exist separately from brain activities. Ideational and neural terminology are simply different ways of representing the same cerebral processes. The view that cognitive events are neural occurrences does not mean that psychological laws regarding psychosocial functioning are derivable from neurophysiological ones. 1181 This document is copyrighted by the American Psychological Association or one of its allied publishers. This article is intended solely for the personal use of the individual user and is not to be disseminated broadly. One must distinguish between biological laws governing the mechanics of cerebral systems and psychological laws of how cerebral systems can be orchestrated to serve different purposes. Psychological knowledge of how best to structure influences to create belief systems and personal competencies is not derivable from knowledge of the neurophysical mechanisms that subserve such changes. Thus, understanding the brain circuits involved in learning does not tell one much about how best to present and organize instructional contents, how to code them for memory representation, and how to motivate learners to attend to, cognitively process, and rehearse what they are learning. Nor does understanding of how the brain works furnish rules on how to create social conditions that cultivate the skills needed to become a successful parent, teacher, or executive. The optimal conditions must be specified by psychological principles. The influences needed to produce the neural occurrences underlying complex human behavior include events external to the organism acting together with selfgenerated ones. The laws of psychology specify how to structure environmental influences and to enlist cognitive activities to achieve given purposes. Although psychological laws cannot violate what is known about the physiological system that subserves them, they need to be pursued in their own right. Were one to embark on the road to reductionism, psychology would be reduced to biology, biology to chemistry, and chemistry to physics, with the final stop in atomic particles. Neither atomic particles, chemistry, nor biology will provide the psychological laws of human behavior. The construal of cognitions as cerebral processes raises the intriguing question of how people come to be producers of thoughts that may be novel, inventive, visionary, or that take complete leave of reality as in flights of fancy. One can originate fanciful but coherent thoughts as, for example, visualizing a hippopotamus gracefully riding the waves on a surfboard. Similarly, one can get oneself to cognize several novel acts and choose to execute one of them. Cognitive production, with its initiating and creative properties, defies explanation in terms of external cueing of preexisting cognitive products. Neither situational cues, knowledge structures, conditioned responses, nor prior brainwaves are likely to be highly predictive of the specific forms fanciful thoughts will take. Emergent cognitive events draw on existing cognitive structures but go beyond them. If thought processes are conceived of as cerebral processes, the relevant question is not how mind and body act on each other, but how people can bring into being cognitive or cerebral productions. The issues of interest concern the brain dynamics of cognitive generation. The novel scenario of the surfing hippopotamus was produced by the intentional exercise of personal agency. Intentionality and agency raise the fundamental question of how people activate the cerebral processes that characterize the exercise of agency and lead to the realization of particular intentions. In addition to asking how people originate thoughts and actions, one may also ask the intriguing 1182 question of how people occasion self-perceiving and selfreflecting cognitive activities. Human Agency, Freedom, and Determinism The notion of human agency also raises the fundamental issue of its relation to determinism. The term determinism is used here to mean the production of effects by events, rather than in the doctrinal sense that actions are completely determined by a prior sequence of causes independent of the individual. When viewed from the perspective of social cognitive theory, there is no incompatibility between human agency and determinism (Bandura, 1986). Freedom is not conceived negatively as the absence of external coercion or constraints. Rather, it is defined positively in terms of the exercise of self-influence. I have already examined how the exercise of personal agency is achieved through reflective and regulative thought, the skills at one’s command, and other tools of self-influence that affect choice and support selected courses of action. Self-generated influences operate deterministically on behavior the same way as external sources of influence do. Given the same environmental conditions, persons who have developed skills for accomplishing many options and are adept at regulating their own motivation and behavior are more successful in their pursuits than those who have limited means of personal agency. It is because self-influence operates deterministically on action that some measure of self-directedness and freedom is possible. Those who argue that people do not exercise any control over their motivation and action usually invoke a selective regression of causes in the analysis of self-regulation. They emphasize that external events influence judgments and actions, but neglect the portion of causation showing that the environmental events, themselves, are partly shaped by people’s actions. Environments have causes as do behaviors. In the model of reciprocal causation, people partly determine the nature of their environment and are influenced by it. Self-regulatory functions are personally constructed from varied experiences not simply environmentally implanted. Although people’s standards and conceptions have some basis in reality, they are not just ingrafts of it. Through their capacity to manipulate symbols and to engage in reflective thought, people can generate novel ideas and innovative actions that transcend their past experiences. They bring influence to bear on their motivation and action in efforts to realize valued futures. They may be taught the tools of self-regulation, but this in no way detracts from the fact that by the exercise of that capability they help to determine the nature of their situations and what they become. The self is thus partly fashioned through the continued exercise of self-influ…
Purchase answer to see full attachment