The contact hypothesis predicts that cooperative interaction with members of a disliked group results in increased liking for those members and generalizes to more positive attitudes toward the group. The authors sought to provide evidence consistent with the hypothesis that contact affects attitude in part by eliciting a more positive portrait of the typical group member. Undergraduates participated in a 1-hr dyadic learning session (scripted cooperative learning, jigsaw cooperative learning, or individual study) with a confederate portrayed as a former mental patient. Students initially expected the confederate to display traits similar to those of a typical former mental patient. After the sessions, initially prejudiced students in the 2 cooperative conditions described the typical mental patient more positively and adopted more positive attitudes and wider latitudes of acceptance toward the group. Connections between intergroup attitudes and impression formation are discussed.