Abstract
The attenuation of distinctiveness-based illusory correlations in judgments of groups was investigated. Subjects read about desirable and undesirable behaviors performed by members of either two or three groups. The behavioral information about two of the groups (A and B) was identical in all conditions, Group B and undesirable behavior occurring infrequently. When only information about Groups A and B was presented, results replicated the usual distinctiveness-based illusory correlation. That is, judgments of Group B were biased to be more negative than those of Group A. In three other conditions, behavioral descriptions of a third group (C) were also presented. The pattern of information about Group C differed, but in each case it was designed to undermine the basis on which illusory correlations are formed. In all three conditions the additional information about a third group attenuated, but did not eliminate, the illusory correlation evident in the two-group case. The attenuation involved changes in judgments of Group B members but not Group A members.
Get full access to this article
View all access options for this article.
