Person perception in multiple-cue probability learning task as a function of cognitive complexity and inferential set

This study was conducted in order to investigate the relationship between cognitive complexity, one measure of cognitive differentiation, and the accuracy and process of interpersonal learning and perception. Additionally, the effect of differing degrees of ego-involvement with the task was explored. Forty-eight female subjects, divided into high (complex) and low (simple) complexity groups on the basis of the Role Construct Repertory Test, rated the overall personal adjustment of 45 stimulus people under ego-involving and neutral inferential set conditions. Judgments were made both before and after training designed to Improve discrimination. Brunswlk's Lens Model of Inference was employed in analyzing the data, and the results clearly corroborated previous findings that an additive linear regression model accounts for the major portion of variance in subjects' response systems to a multiple-cue task. Although the predicted inverse relationship between cognitive complexity and accuracy of judgment failed to attain statistically significant levels, in all cases the post-learning group means were in the expected direction, with complex subjects being more accurate, more linear, and utilizing cues to better advantage than simple subjects. A similar situation was found for the ego-involved subjects as opposed to those in the neutral set condition. The results also offered tentative evidence for a complexity difference in process of judgment which leads to differential improvement in the task, serendipitously. It was observed that a stereotypic judgment process based on overly-relevant cue dimensions biased the pre-learning trials by permitting excessively high achievement in the operant state.