Despite the prominence of emotion disturbance in bipolar disorder, few studies have assessed emotion differentiation. The present investigation used an experience-sampling approach to test the utility of emotion differentiation in predicting bipolar mood-related difficulties. Across two studies, emerging adults participated during a normative first year of college (Spring 2019; Study 1; n=136) or during their first year of college marked by a naturalistic global pandemic stressor (Spring 2020; Study 2; n=136). Study 1 results suggested that emotion differentiation was not associated with trait bipolar risk. Study 2 suggested that global emotion differentiation was associated with increased trait bipolar risk, but not current mood symptom severity. These results suggest that relationships between emotion differentiation and dimensions of bipolar risk may vary by context. Discussion focuses on the implications for translational interventions.