The neural correlates of direct and reflected self-knowledge Kevin N. Ochsner, a, * Jennifer S. Beer, b Elaine R. Robertson, c Jeffrey C. Cooper, c John D.E. Gabrieli, c John F. Kihsltrom, d and Mark D’Esposito d,e a Department of Psychology, Columbia University, 369 Schermerhorn Hall, 1190 Amsterdam Avenue, New York, NY 10027, USA b Department of Psychology and Center for Mind and Brain, University of California, Davis, CA 95616, USA c Department of Psychology, Stanford University, CA 94305, USA d Department of Psychology, University of California, Berkeley, CA 94704, USA e Helen Wills Neuroscience Institute, University of California, Berkeley, CA 94704, USA Received 18 December 2004; revised 20 June 2005; accepted 23 June 2005 Socrates said that in order to lead a balanced life one must, ‘‘know thyself.’’ In two fMRI experiments, the present study examined the mechanisms mediating two ways in which the self can be known: through direct appraisals (i.e., an individual’s own self-beliefs) and reflected appraisals (i.e., an individual’s perception of how others view him or her). Experiment 1 examined the common and distinct neural bases of direct appraisals of the self, close others, and normative judgments of trait desirability. All three judgment types activated medial prefrontal cortex (MPFC) to a similar degree. Experiment 2 examined the common and distinct neural bases of (1) direct appraisals of self, a close other or a non-close other, and (2) reflected appraisals made from the perspective of a close or a non-close other. Consistent with Experiment 1, all judgment types activated MPFC. Direct appraisals of the self as compared to others more strongly recruited MPFC and right rostrolateral PFC. Direct appraisals as compared to reflected appraisals recruited regions associated with a first-person perspective (posterior cingulate), whereas reflected as compared to direct appraisals recruited regions associated with emotion and memory (insula, orbitofrontal, and temporal cortex). These results support models suggesting that MPFC mediates meta-cognitive processes that may be recruited for direct and reflected self appraisals depending upon the demands of a specific task. D 2005 Elsevier Inc. All rights reserved. Keywords: Medial prefrontal cortex; Self; Self-reference; Person; Percep- tion theory of mind Introduction How do we know what we are like? Am I friendly? Am I outgoing? Am I aggressive or anxious or eager? To answer these questions, contemporary social psychological theories suggest that there are at least two appraisals of the self upon which we draw (Baumeister, 1998; Fiske, 2004; Kenny and DePaulo, 1993; Shrauger and Schoeneman, 1999). Direct appraisals draw primarily upon semantic self-knowledge abstracted from previous experi- ences, although in some cases, self appraisals might draw upon episodic knowledge of specific life events (Kihlstrom et al., 2003; Klein et al., 1989, 1996; Markus, 1977; Tulving et al., 1988). By contrast, reflected appraisals draw on our beliefs about how we are seen by others. The person Fseen_ in reflected appraisals has been termed ‘‘The Looking Glass Self’’ (Cooley, 1902; Mead, 1934). 1 For much of the past century, behavioral researchers have asked two distinct but related questions about the mechanisms underlying direct and reflected appraisals. The first question is whether direct appraisals of the self are Fspecial,_ insofar as they draw upon different informational bases and inferential processes used to directly appraise other individuals. For example, it is possible that evaluating your own friendliness recruits highly accessible memory representations that are more structured and complex than those used to evaluate the friendliness of a work colleague (Dweck et al., 2003; Higgins and Bargh, 1987; Symons and Johnson, 1997). At best, there is mixed evidence to support this claim. On one hand, memory is enhanced for traits evaluated in relation to the self in comparison to other targets. This memory advantage has been argued to reflect something Fspecial_ about the unique organizational structure of self-knowledge (Maki and McCaul, 1985; McCaul and Maki, 1984; Rogers et al., 1977). On the other hand, similar levels of memory are observed for stimuli whose representations are equated for complexity of organization and/or elaboration (Klein and Kihlstrom, 1986; Symons and Johnson, 1997). 1053-8119/$ - see front matter D 2005 Elsevier Inc. All rights reserved. doi:10.1016/j.neuroimage.2005.06.069 * Corresponding author. E-mail address: ochsner@psych.columbia.edu (K.N. Ochsner). Available online on ScienceDirect (www.sciencedirect.com). 1 A variant on this second possibility is that self-knowledge derives not just partially, but almost completely from our perceptions of others (Cooley, 1902; Lewin, 1951; Mead, 1934; cf. Bem, 1967). www.elsevier.com/locate/ynimg NeuroImage 28 (2005) 797 – 814