Using Private Knowledge to Predict How One Is Viewed by Others

People have more information about themselves than others do, and this fundamental asymmetry can help to explain why individuals have difficulty accurately intuiting how they appear to other people. Determining how one appears to observers requires one to utilize public information that is available to observers, but to disregard private information that they do not possess. We report a series of experiments, however, showing that people utilize privately known information about their own past performance (Experiments 1 and 2), the performance of other people (Experiment 3), and imaginary performance (Experiment 4) when intuiting how they are viewed by others. This tendency can help explain why people’s beliefs about how they are judged by others often diverge from how they are actually judged. People devote considerable attention to wondering (and, at times, worrying) about how they appear in the eyes of others. Despite this attention, people’s beliefs about how others view them are often mistaken (Gilovich & Savitsky, 1999; Kenny & DePaulo, 1993). For example, several investigations have demonstrated that people overestimate the extent to which observers judge them harshly after a failure or embarrassing blunder, and that this miscalibration stems partly from people’s failure to consider information that observers take into account. People account insufficiently for observers’ tendency to empathize and commiserate with those who commit embarrassing blunders (Epley, Savitsky, & Gilovich, 2002); they do not consider ‘‘nonfocal’’ information as fully as observers do (Savitsky, Epley, & Gilovich, 2001); and they fail to account for observers’ tendency to adjust their inferences to reflect mitigating situational constraints (Van Boven, Kamada, & Gilovich, 1999). The research we report in this article explored the inverse notion: that difficulties intuiting how one is viewed by observers are produced not only by failing to consider information considered by observers, but also by actively utilizing information that observers fail to consider (indeed, have no access to). Beliefs about how one is perceived can be biased by a tendency to evaluate one’s own performance in light of contextual information that is unavailable to observers, and then to use this selfevaluation as a guide for intuiting their impressions. A woman may know, for example, that she is less fit than she used to be, more attractive than most of her friends, and less productive than she might wish. Acquaintances who know nothing of her past, her friends, or her wishes, however, can hardly use such comparisons when forming their impressions, and instead can base their impressions only on information that is currently available to them. And yet, like jurors who find it difficult to disregard inadmissible evidence once it has been provided, people may find it difficult to disregard what they know is private information when intuiting how others view them. Private knowledge, like inadmissible testimony in court, can influence how people encode and evaluate an event so profoundly that correcting for the private nature of this information can prove difficult, if not impossible. This difficulty can help explain why people err in estimating how they appear in the eyes of others. The influence of private information may be a particularly powerful determinant of error in daily life because of a basic asymmetry in the amount of information people possess about themselves versus the amount that others possess about them (Nisbett, Caputo, Legant, & Marecek, 1973). People have privileged access to their own internal thoughts and feelings, and observe themselves across time, continuously from one moment to the next, but they are viewed by others in isolated episodes. An individual therefore experiences events in his or her own life within a rich situational context that allows each event to be understood in light of information that is often unavailable to others. When this private information is discrepant with one’s public performance—which is available to others—there is likely to be Address correspondence to John R. Chambers, Department of Psychology, P.O. Box 112250, University of Florida, Gainesville, FL 32611, e-mail: [email protected]. PSYCHOLOGICAL SCIENCE 542 Volume 19—Number 6 Copyright r 2008 Association for Psychological Science a divergence between how one expects to be judged and how one is actually judged. Consider a speaker who delivers the same lecture on two consecutive evenings. Although he or she can compare performances from one night to the next, audience members cannot. The speaker is therefore likely to overestimate how positively the second lecture will be evaluated if it seems better than the first, but to overestimate how negatively it will be evaluated if it seems worse. This hypothesis calls to mind several supportive findings. First, considerable evidence suggests that people consult their own mental states as a proxy for other people’s mental states (Epley, Keysar, Van Boven, & Gilovich, 2004; Kahneman & Lovallo, 1993; Meltzoff & Brooks, 2001; Nickerson, 1999). This egocentrism can lead individuals to overestimate the extent to which others have access to their internal thoughts and personal attributes (Gilovich, Savitsky, & Medvec, 1998; Savitsky & Gilovich, 2003; Vorauer & Ross, 1999), as well as the extent to which others share their beliefs and subjective perceptions (Keysar, 1994; Ross, Greene, & House, 1977). Because people often have little awareness of how contextual information has influenced how they encode and evaluate events, they may make little effort to undo such influence when intuiting other people’s impressions (Brenner, Rottenstreich, & Sood, 1999; Fischhoff, 1975; Hsee, 1996; Hsee & Zhang, 2004). Even when contextual influence is rendered obvious, individuals’ attempts to correct their egocentrically encoded judgments tend to be insufficient (Epley et al., 2004; Keysar & Barr, 2002). Second, our hypothesis resonates with research showing that more information can, ironically, decrease accuracy (Hall, Ariss, & Todorov, 2007). Indeed, it is often knowledge individuals have about themselves that undermines the accuracy of their self-related judgments (Buehler, Griffin, & Ross, 1994; Epley & Dunning, 2000, 2006). We argue here that individuals’ private knowledge of the context in which their performances occur can bias their beliefs about how others view them. In other words, we suggest that people have difficulty knowing how they are viewed by others because they simply know too much about themselves.