A closer examination of causal inference: The roles of consensus, distinctiveness, and consistency information

Assumed individuals, in making attributions for behavior, expect to encounter information patterns indicating stimulus, person, or circumstance causation. Each of these patterns is characterized by certain levels of consensus, distinctiveness, and consistency. When given information about a specific instance of behavior, individuals relate the information to these expected patterns and interpret the behavior in terms of the attributions implied by the patterns with which it is consistent. From these assumptions, predictions were made regarding the interpretation of various combinations of specified information and the information judged likely to exist in cases where it is not specified. Stimulus materials were statements describing interpersonal behaviors (Study 1) or achievements (Study 2), each followed by 1 to 3 sentences supplying consensus, distinctiveness, and/or consistency information. 216 undergraduates were required either to make inferences about causality or to characterize the missing information. Results of both procedures in both studies are consistent with the predictions. Several biases in the interpretive process were distinguished. These are interpreted as consequences of expectations that (a) behavior frequently covaries with situation and (b) in achievement as compared with interpersonal settings, a person's response to a particular stimulus is more consistent with his or her responses to similar stimuli and with other individuals' responses to the particular stimulus.