In 16 independent samples from five countries involving ~7,700 participants, we employ a mixture of cross-sectional, longitudinal, and quasi-experimental methods to examine the effect of intergroup contact on (a) the blatant dehumanization of outgroups, and (b) the perception that outgroup members dehumanize the ingroup (meta-dehumanization). First, we conduct a meta-analysis across 12 survey samples collected from five countries regarding eight different target groups (total N = 5,388) and find a consistent effect of contact quality on dehumanization and meta-dehumanization. Second, we use a large longitudinal sample of American participants ( N = 1,103) to show that quality of contact with Muslims at Time 1 predicts dehumanization of Muslims and meta-dehumanization 6 months later. Finally, we show that sustained semester-long “virtual contact” between American and Muslim college students is associated with reduced American students’ ( N = 487) dehumanization of, and perceived dehumanization by, Muslims.