抄録
The path-analytic averaging (PAA) model of information integration predicts that the two most important findings in the information integration literature, cross-over interaction and the set-size effect, are mutually exclusive. Previous tests of the PAA model have examined cross-over interaction and the set-size effect separately; the present experiment tested both simultaneously, by asking 40 undergraduates to rate the desirability of candidates for secretarial positions based on one or two items of information. Findings supported the exclusivity prediction in experimental conditions in which no fan spread of the two-item slopes occurred. When fan spread was elicited, however, exclusivity was not so clearly supported.