Personality and Social Psychology Review 1998 Vol. 2, No.2, 137—154 Copyright ~ 1i98 by Lawrence Eribatim Associates, Inc. Social Metacognition: An. E rpansicnis R view John T. Jost Graduate School of Business Stanford University Arie W. Kruglanski and Thomas 0. Ne sort Department of Psychology University of Maryland at College Park In this article we review recent social psychological contributions to the metacognitive movement. It is argued that social psychologists have long contributed to the study of “thinking about thinking,” even though their work has not yet been recognized as metacognitive. The present “expansionist” survey suggests that the domain of social metacognition should include (a) beliefs about one’s own mental states and processes as well as beliefs about those of other people, (b) momentary sensations as well as enduring folk theories, and (c) descriptive beliefs about how the mind works and normative beliefs about how it ought to work. The contents and origins of metacogni- tion are inherently social; at the same time, metacognitions are comprised of cognitive elements and are governed by the principles and laws applicable to human thinking in general. Accordingly, whereas metacognitions about self-knowledge may be de- rived from different informational sources than metacognitions about other people, the processes whereby different types of metacognitions are formed, activated, and applied are essentially the same. Focusing on the social nature of metacognition and the profound relevance of cultural expectations on cognitive performance makes clear the benefits of systematically exploring the cognitive—social interface in reference to metacognitive phenomena. Human beings possess the striking reflexive capac- ity to hold mental states about mental states, that is, to engage in thinking about thinking. This has been her- aIded as a grand evolutionary achievement of our spe- cies and its kin (Povinelli, 1993), and its social and psychological significance may scarcely be exagger- ated. Daily life is replete with complex determinations about the reliability of our own thoughts, feelings, and beliefs as well as attributions about the thoughts, feel- ings, and beliefs of others around us (Nelson, Kruglan- ski, & Jost, 1998). For centuries, philosophers have Portions of this article were written while John T. Jost was supported as a Postdoctoral Research Associate at the University of Maryland by National Institutes of Mental Health Grant ROl—MH52578 administered to Arie W. Kruglanski. Additional fundingfor this project was provided by National Science Foundation Grant SBR—9417422 and Research Scientist Award K05—MH01213 to Arie W. Kruglanski and by National Institutes of Mental Health Grant ROI—MH32205 and Research Scientist Award K05—MHIO7S to Thomas 0. Nelson. We thank Michael W. Morris for helpful discussions of this material as well as Marcos Brodsky and Katy Jakle for assistance with manuscript preparation. Requests for reprints should be sent to John T. Jos:, Graduate Sehool of Business, Stanford University, Stanford, CA 94305—5015. E-mail: Jos~John@gsb.stanford.edu. speculated about the dependability of self-knowledge and about the problems inherent in drawing inferences about the contents of other minds (Rosenthal, 1991). Contemporary psychologists, too, have devoted consid- erable theoretical and research attention to the topic of metacognition, which recently cracked the “top 100 topics” of psychological research (Nelson, 1992, p. ix). Although a good number of edited volumes on the subject now exist (e.g., Antaki & Lewis, 1986; Forrest- Pressley, MacKinnon, & WaIler, 1985; Metcalfe & Shimamura, 1994; Nelson, 1992; Yzerbyt, Cones, & Dardenne, in press), there is no consensually agreed on definition of metacognition. Kitchener (1983) noted that the term “has been variously referred to as ‘cognt- tive monitoring’ (Flavell, 1979), ‘executive processes’, self-communication’, and ‘knowledge about knowl- edge”’ (p. 222). Typically, metacognitive researchers investigate mental processes whereby individuals “con- trol” and “monitor” their own mental activity (Nelson & Narens, 1994). Although this body of work has advanced psycholocical understandin~ considerably, it restricts its focus in three significant ways. Cirst, it adriresse:: :etacoenitive jud~menrs about :he sri but ignores the role of metacogni ions about other people. Second. it emphasizes fescrior~ ye beliefs about how the