The proposed project seeks to mitigate the negative consequences of populism and deep political polarization for democracy. It builds on prior collaborative research on the causes of polarization and its consequences for democratic erosion around the world. I propose to conduct experiments to test hypotheses drawn from social, moral and political psychology about political tribalism (in-group/out-group formation) in order to determine the effectiveness of different interventions in mitigating the antipathy and distrust of the “Other” typical of severe political and societal polarization. Outcome variables include changes in a) attitudes toward issues and groups, b) willingness to cooperate and compromise across groups and c) changes in support for democratic and liberal norms and practices. The project will use Crowdflower and Amazon Mechanical Turk to conduct on-line survey experiments in Europe and the United States.