Psychological Review1973, Vol. 80, No. 4, 252-283TOWARD A COGNITIVE SOCIAL LEARNINGRECONCEPTUALIZATION OF PERSONALITY JWALTER MISCHEL 2Stanford UniversityDiverse data challenge and undermine ihe central assumptions of the traditional trait approach to personality. The implications for conceptions ofindividual differences and situations in the study of personality are examined.The issues discussed include the nature of behavioral "specificity," the acquiredmeaning of stimuli, the uses and misuses of traits, and the construction ofpersonality. To move toward a more adequate theoretical approach to per-sons, the following cognitive social learning variables are proposed as basicunits for the study of individuals: cognitive and behavioral construction com-petencies, encoding strategies and personal constructs, behavior-outcomeand stimulus-outcome expectancies, subjective stimulus values, and self-regulatory systems and plans. The specific interactions between these personvariables and psychological situations are analyzed within the framework ofa cognitive social learning approach.There has been a curious—indeed alarm-ing—bifurcation between progress in theo-ries regarding complex social behavior andcognition on the one hand, and in con-ceptualizations regarding the basic natureof personality on the other. Many of thetherapeutic implications of social learning(social behavior) theories have become evi-dent in the last few years. There havebeen notable advances in treatment tech-niques as well as significant reconceptuali-zations of the treatment process itself (e.g.,Bandura, 1969). These developments arejust starting to be accompanied by com-parable parallel developments in person-ality theory. In a second direction, therehas been vigorous progress in cognitivepsychology (e.g., Neisser, 1967). Butwhile cognitive and symbolic processes havereceived increasing attention both in the1 Parts of this manuscript are based on the Addressof the Chairman, Section III, Division 12, AmericanPsychological Association, Washington, D. C.,September 3, 1971. Preparation of this paper was fa-cilitated by National Institute of Mental HealthGrant M-6830 and National Science FoundationGrant GS-32582. Constructive comments have beenreceived from more colleagues and students thancan be listed here; the author is grateful for theirhelp.2 Requests for reprints should be sent to WalterMischel, Department of Psychology, Stanford Uni-versity, Stanford, California 94305.laboratory and in therapeutic applications,their implications for personality psychol-ogy have not yet been thoroughly exploredand their impact on the basic traditionalassumptions of personality psychology untilrecently has been limited.During the last 50 years, when basicconcepts were changing rapidly in mostfields of psychology, the most fundamentalassumptions about the nature of personalityseem to have been retained with few sub-stantial modifications. Of course therehave been many changes in the names andparticular characteristics of the trait dis-positions advocated by different theore-ticians and personality researchers in thelast few decades. But in spite of the hetero-geneity of hypothesized dimensions orstructures, perhaps the most fundamentalassumptions about them have remainedalmost monolithic until very recently. Thispaper briefly reviews the central assump-tions of global dispositional approaches topersonality, considers some of the mainmisconceptions, issues, and implicationsarising from recent challenges to thoseassumptions, and finally attempts a re-conceptualization of person variables in thelight of concepts from the study of cogni-tion and social learning.252RECONCEPTUALIZATION OF PERSONALITY253GLOBAL DISPOSITIONAL APPROACHESTO PERSONALITYAssumptions of Traditional TraitApproachesIt has generally been assumed thatpersonality dispositions or traits—the basicunits of personality study—are relativelystable, highly consistent attributes thatexert widely generalized causal effects onbehavior. Whether one uses the languageof factors, or of habits, or of basic atti-tudes, or of dynamics and character struc-ture, this fundamental assumption has beenshared: personality comprises broad under-lying dispositions which pervasively in-fluence the individual's behavior acrossmany situations and lead to consistencyin his behavior (e.g., Allport, 1937).3 Thesedispositions are not directly observed butare inferred from behavioral signs (traitindicators), either directly or indirectly(Mischel, 1968). Guided by this assump-tion, personality research has been a questfor such underlying broad dimensions, forbasic factors, or for pervasive motives, orfor characteristic life styles. In personalityassessment the trait assumptions regardingstructure are seen in the existence of hun-dreds of tests designed to infer dispositionsand almost none to measure situations.The same belief in global traits that mani-fest themselves pervasively is perhaps bestseen in the projective test assumption thatresponses to vague or minimal stimuli willreveal individual differences in fundamentalgeneralized dispositions (MacFarlane &Tuddenham, 1951).Empirical Status of AssumptionsGiven the pervasiveness of the con-sistency assumption of dispositional per-sonality theory, its empirical status be-comes especially important. There havebeen several recent reviews of that evi-dence (e.g., Mischel, 1968, 1969, 1971;Peterson, 1968; Vernon, 1964). The data3 In social psychology, the "attitude" has beenthe unit endowed with properties parallel to thoseassigned to the trait in the field of personality, and itappears to be subject to very similar criticisms andproblems (e.g., Abelson, 1972).cannot be summarized adequately here,but several themes emerge. To recapitu-late briefly, impressive consistencies oftenhave been found for intellective features ofpersonality and for behavior patterns suchas cognitive styles and problem-solvingstrategies that are strongly correlated withintelligence (e.g., Witkin, 1965). Con-sistency also is often high when peoplerate their own traits, as in questionnairesand other self-reports (e.g., E. L. Kelly,1955). Temporal continuity also has beendemonstrated often when the individual'sbehavior is sampled at different timeperiods but in similar situations. Whenone goes beyond cognitive variables topersonality dimensions and when onesamples personality by diverse methods andnot just by self-report questionnaires, thedata change and undermine the