Personality and metaperception.

Abstract

The purpose of this study was to investigate the relationship between normal and disordered personality and the ability to see oneself as one is seen by others. The subjects were 121 undergraduates from the University of Michigan Introductory Psychology subject pool and 46 adults from the local community. They completed four self-report measures: the Millon Clinical Multiaxial Index - III (MCMI-III), the NEO Five-Factor Inventory (NEO-FFI), the Goldberg 100 Unipolar Markers, and the Inventory of Interpersonal Problems - Circumplex (IIP-C). In addition, the subjects participated in structured interactive tasks with three different partners who were also subjects in the study. At the end of each interaction, subjects rated their partners on a set of adjectives, and guessed how they would be rated by their partners on the same adjectives. The subject and partner ratings were used to compute accuracy scores for each subject. The accuracy scores were subjected to a factor analysis which yielded two factors: an introverted/neurotic factor and a likable/competent factor. The scores on the self-report measures served as the independent variables and the accuracy factors served as the dependent variables in a series of hierarchical multiple regressions. Subjects who received high scores on the MCMI-III Avoidant and Dependent Personality disorder scales, high scores on the Neuroticism scale of the NEO-FFI, or low scores on the Emotional Stability scale of the Goldberg 100 Unipolar Markers overestimated the degree to which they were perceived by their partners to be introverted and neurotic. Subjects who received high scores on the Agreeableness and Surgency scales of the Goldberg 100 Unipolar Markers, or high scores on the Extraversion scale of the NEO-FFI overestimated the favorability of their partners' ratings on both accuracy factors. Subjects who reported high levels of interpersonal problems as measured by the total score of the IIP-C were likely to underestimate the positivity of their partners' ratings on both accuracy factors. There were no effects due to group or sex for any of the findings.