There is increasing concern that affective polarization has the potential to undermine democratic norms and social cohesion. In this research plan, we extend the idea of affective polarization, not through partisanship, but through identification with two opinion-based groups, namely on the issues of migration and vaccination, and investigate whether such affective divisions across political camps have negative implications for social cohesion and democratic norms as pivotal cornerstones of democracy. We argue that individuals may develop opinion-based identification with others who share the same opinion, and that this identification may extend to affective biases in non-political situations. In turn, affective biases may undermine citizens’ support for democratic norms and tolerance. To investigate these arguments, we conduct a threefold empirical study. First, we use a conjoint experiment that allows us to investigate how opinion-based identities causally shape social interactions outside the political realm. Second, we use cross-sectional survey data from Switzerland to scrutinize the role of identity in shaping affective polarization in the context of political issues beyond partisanship. Third, to provide causal evidence on the relationship between different forms of affective polarization and democratic norms, we propose an original survey experiment that modifies the traditional trust game in order to causally induce variation in affective polarization. This should allow us to investigate how opinion-based and partisan affective polarization causally shape tolerance and democratic norms.