Perceptions of Intrapersonal and Interpersonal Insight

Perceptions of Intrapersonal and Interpersonal Insight

Our discussion of naive realism and insider versus outsider perspectives gives rise to the following four hypotheses about lay perceptions of self-knowledge and knowledge of others (see Pronin, Kruger, Savitsky, & Ross, 2001):

1. People perceive their own self-knowledge and insight to be more accurate and complete than that of other people.

2. People perceive their knowledge of other people to be more accurate and complete than other people’s knowledge of them.

3. People perceive the discrepancy between their self-knowledge and other people’s knowledge of them to be greater than the corresponding discrepancy between other people’s self-knowledge and their knowledge of those other people.

4. People perceive their group’s knowledge of other groups to be more accurate and complete than other groups’ knowledge of their group.

In an initial exploration of the first three of these hypotheses, Pronin et al. (2001) simply asked roommates to complete a survey. In this survey, respondents evaluated how well they knew themselves, how well they knew their roommate, how well their roommate knew them, and how well their roommate knew himself or herself, with respect to domains ranging from specific traits (e.g., competitiveness) and behaviors (e.g., frequency of looking in the mirror) to more general knowledge about thoughts, feelings, and motives. The results provided evidence for all three hypotheses. In virtually every domain, people rated their self-knowledge to be superior to their roommate’s self-knowledge, and their knowledge about their roommate to be greater than vice versa. They also indicated that although they sometimes knew their roommates as well as, if not better than, their roommates knew themselves, there were no domains in which their roommates matched or exceeded the participants’ own self-knowledge. Figure 36.4 in the source summarizes this study and three related studies described below; the figure is not included here. Further analysis suggested that participants were most skeptical of the accuracy of their roommates’ self-knowledge relative to their own when the traits in question were negative (e.g., messiness, susceptibility to flattery, inability to handle stress) rather than neutral or positive (e.g., risk-taking, giving weight to moral considerations), and when they pertained to private thoughts and feelings rather than overt behavior.

Subsequent studies by Pronin et al. (2001) revealed that people show this asymmetry not only in assessments of intrapersonal and interpersonal insight on the part of roommates with whom they interact on a daily basis, but also in assessments of how much they learn about strangers, and vice versa, from observing small, dubiously informative samples of behavior. In one such study, previously unacquainted individuals were given a chance to get to know each other by taking turns asking relatively innocuous personal questions of their choosing (e.g., “Where do you hope to see yourself in ten years?” or “What would your ideal vacation be?”). Participants generally believed that they learned more about the other person from the brief encounter than the other person learned about them. This perception was revealed both in ratings of how much each learned about the other (in terms of traits, preferences, etc.) and also in the length of written descriptions about what they had learned.

A similar result was obtained in a third study in which participants both furnished completions of word fragments and read the completions of another participant. Participants saw their “partner’s” responses as more self-revealing than they did their own. They typically saw their own responses as either situationally determined (that is, as suggested by the fragment, or cued by a previous response) or as an essentially random and uninformative reflection of their momentary state of mind, while they saw their partners’ responses as reflective of personal traits or dispositions. In other words, they seemed well aware of the folly involved in attaching much significance to their spontaneous and rather haphazard answers, but less aware of the folly involved in making inferences about their partners on the basis of the same type of information. As in the brief encounter study, they recognized their partners’ susceptibility to bias, misinterpretation, and especially over-interpretation when their partners made inferences about them on the basis of the meager information provided, but they did not recognize the likelihood that they would be equally susceptible to those failings when they made inferences about their partners on a similar basis.

Interestingly, participants in this study showed this self–other asymmetry only when they assessed the diagnosticity of their partners’ completions before providing and assessing their own completions. In other words, the tendency to make overly confident trait inferences about the actor whose responses one was considering seemed, at least in this study, to be forestalled by the prior opportunity to ponder the determinants of one’s own responses.

In a final study, Pronin et al. (2001) found evidence for similar biases and asymmetries in perceptions of accuracy and bias at the group level. That is, individuals claimed that the political groups to which they belonged (e.g., liberals versus conservatives, pro-choice versus pro-life advocates) knew and understood their counterparts better than vice versa. The same tendency was also true for men’s and women’s perceptions about each other’s insights, although the relevant asymmetry in this case was masked somewhat by the fact that both sexes agreed that women understood men better than men understood women. The women simply thought this gap was greater!

What accounts for the impression or belief that we know them better than vice versa? One possibility relevant to our present concerns, suggested by Quattrone and Jones (1980), is the conviction that the members of other groups (i.e., “they”) are simply more homogeneous and therefore more easily characterized than the members of our group (i.e., “we”). This conviction, in turn, may arise from a tendency to view other groups’ responses (just as we view other individuals’ responses) as more dictated by self-interest, ideology, or other top-down influences than those of our own group. In any case, there is ample research evidence that individual group members believe that members of the other group fail to appreciate their group’s diversity or variability, and see them in a stereotyped manner (Linville, Fischer, & Salovey, 1989; Park & Judd, 1990). Indeed, a pilot study conducted in our own laboratory (Pronin & Berger, 1999) suggests that students who live in two residences subject to campus-wide stereotypes (i.e., as the “jock frat” and “hippie house”) each believed that their house would recognize the diversity in the responses of residents of the other house more than vice versa. A more specific source of the belief that they are more misinformed and less knowledgeable about us than vice versa, is further suggested by this pilot study. Respondents believed that inhabitants of the other house gave more credence to stereotypes and propaganda about their house than vice versa.

These findings concern perceived insight. Believing that another person reveals more than we do, or that our group understands theirs better, does not establish that we actually know more. The word-fragment study also suggests that considering the circumstances behind our own responses can affect how confidently we interpret someone else’s.