Political psychology in an era of personalization: Misinformation research, generative AI, and the promise of tailored experiments

Abstract

Voters increasingly inhabit personalized information environments, where content, algorithms, and social cues vary across people. Yet popular approaches to studying and addressing features of information environments such as misinformation exposure often flatten this heterogeneity: Measures and interventions are developed for populations rather than individuals. Using misinformation as an illustrative case, I argue that population‐based methods for measuring and intervening on beliefs can obscure why some beliefs matter politically while others do not. I advocate for tailored approaches that adapt research designs to groups and individuals. Tailoring can increase relevance, facilitate persuasion, and create fruitful conditions for testing influential theories of opinion formation and change. I conclude by discussing the promise of a personalized political psychology that aligns measures and interventions with the idiosyncratic ways voters form and express opinions.