An Integrated Model of Cognition, Emotion, and Vote Choice: Evidence from Taiwan

Political behavior research often separates cognitive evaluations from emotional reactions, treating voters as either rational performance evaluators or emotionally driven decision makers. This study advances an integrated cognition–emotion model in which broad political evaluation (e.g., retrospective judgment) and discrete emotional response—anxiety and anger—operate as distinct but interacting pathways shaping electoral choice. Using post-election survey data from Taiwan’s 2024 presidential contest and controlling for lagged vote choice and feelings toward candidates, I show that negative assessments of Taiwan’s international status and cross-Strait relations significantly reduce support for the incumbent. Anxiety about potential conflict independently lowers incumbent support and amplifies the electoral penalty associated with negative evaluations. Anger toward external coercion, by contrast, increases support for the incumbent but does not strengthen the effect of cognitive judgments. Across all robustness checks, the core findings remain substantively identical. These findings offer a framework for understanding how citizens interpret political conditions and how these interpretations—cognition and emotions–shape their vote choices.