Verbal reports about causal influences on social judgments: Private access versus public theories

Female subjects were asked to make four judgments about a young woman after reading her "job application portfolio. " Five characteristics of the young woman were manipulated orthogonally (e.g., physical appearance, academic credentials). Subjects were asked to report how each of the five manipulated factors had influenced each of their judgments. "Observer subjects, " who had access only to very impoverished descriptions of each of the five factors, were asked to predict how each of the factors would influence each of the judgments. Results showed that (a) subject reports about the effects of the factors on the judgments were in general highly inaccurate; (b) observer predictions were ex-tremely similar to subject reports; (c) for the single judgment for which sub-jects showed substantial accuracy (a judgment about intelligence), observer predictions were as accurate as subject reports. The results indicate that what-ever introspective access to cognitive processes may exist, it is not sufficient to produce generally accurate reports about such processes, nor even to produce reports that differ much from those of poorly informed outside observers. Re-ports about cognitive processes may be based less on introspection than on a priori causal theories about the effects of particular stimuli on particular responses. Nisbett and Wilson (1977) have argued recently that people may have little or no direct introspective access to complex mental processes such as those involved in judgment, The research reported here was supported by Grant BNS 75-23191 from the National Science Foundation. The authors are indebted to Timothy D. Wilson for advice, criticism, and assistance; to Keith Smith for statistical help; and to Sharon Gurwitz, Ronald Lemley, Hazel Markus, and Lee Ross for comments on the manuscript. Requests for reprints should be sent to Richar