ABSTRACT In this article, we report investigations of four role-playing experiments and one laboratory manipulation that examine the effects of confession on forgiveness and other related judgments. The basic paradigm in the simulation studies was to reveal that a political figure or student in a class confessed either following or not following an accusation, or denied personal responsibility for the act. Among the variables manipulated were the attributions for the wrongdoing and the spontaneity of the confession. The dependent variables in one or more investigations included the perceived personal character of the trangressor, attributions of responsibility for the act, affective reactions of sympathy and anger, forgiveness, and behavioral judgments such as sanctioning and voting likelihood. In the laboratory manipulation study, a mixed-motive game setting was used in which a confederate confessed to having prior knowledge that resulted in his winning the game. We then examined whether this admission influenced subsequent cooperation and competition, as well as the other players' perceptions of the confederate's personality and character. Confession was found to have strong beneficial effects, particularly when given without a prior accusation and in ambiguous causal situations.