Contact researchers have overlooked (a) the mechanisms that explain the association between negative contact and prejudice, (b) the effects of positive and negative contact on outcomes beyond prejudice, and (c) the importance of testing contact effects cross-culturally. In the present article, we addressed these gaps in the literature by drawing on data from White Americans (N = 207; Study 1), Hong Kong Chinese (N = 145; Study 2), and Buddhist Thais (N = 161; Study 3). Specifically, we examined positive and negative contact as predictors of old-fashioned and modern prejudice toward, and negative metaperceptions about, Black Americans, Mainland Chinese, and Muslim Thais, respectively. We also tested intergroup anxiety as a mediator of the associations between positive and negative contact, and all intergroup outcomes. Across three studies, positive contact predicted reduced intergroup anxiety, prejudice, and negative metaperceptions, while negative contact predicted increased intergroup anxiety, prejudice, and negative metaperceptions. Negative contact, however, was the more consistent predictor of intergroup attitudes. Intergroup anxiety emerged as a robust mediator of the relationships between both types of contact and all intergroup outcomes. We thus present the first test of a model of positive and negative contact that holds across both Western and non-Western contexts.