Abstract
An attempt was made to extend Dustin and Baldwin's (1966) model of impression formation to sets of six traits. Ss rated the probability that three undesirable traits (U) would describe a person, given that he possessed three desirable traits (D). Ss then rated their liking for the person, assuming that he actually did possess all six traits. The results showed a curvilinear relationship between probability ratings and liking ratings, thus only partially supporting the model. When the perceived relation between D and U traits was relatively improbable, the greater the improbability, the greater the liking. When the perceived relation was relatively probable, the greater the probability, the greater the liking. The results supported both a need for novelty and a need for consistency hypothesis.
Get full access to this article
View all access options for this article.