Polarization in Social Contexts: Multiple-World Experiments on the Dynamics of Opinion Formation, Fact Perception and Collective Action

• Applicant: Professor Dr. Marc Keuschnigg
• Subject Area: Empirical Social Research ; Sociological Theory
• Term: since 2026
• Project identifier: Deutsche Forschungsgemeinschaft (DFG) - Project number 561512351

In today's interconnected world, global crises like climate change, violent conflicts, and mass migration have accentuated the perception of growing polarization. Everyday observations suggest an increase in radical attitudes and behaviors that potentially fray the social fabric of societies. At the same time survey data indicate that polarization is less severe than news, social media, and political rhetoric suggest. This proposal argues that both the apparent lack of polarization in surveys and the intense polarization seen in public discourse can coexist. People may report less polarized views in isolation, but exhibit heightened polarization in social contexts marked by in-group and out-group boundaries. In such social environments, we expect affective polarization to take hold of opinion formation, factual beliefs, and their readiness for collective action. In a total of three work packages (WPs), each using a large-scale online experimental design (n>2000), we will investigate how different social contexts impact affective polarization and its effects on opinions, beliefs, and actions. These experiments will be complemented by agent-based simulations to explore broader scenarios beyond experimental constraints. WP1 investigates how opinions, i.e., personal convictions that lack a definite ground truth, form and evolve in different social contexts. It establishes our methodology of macrosociological multiple-world experiments and explain why similar populations can appears both mildly and highly polarized, depending on the social circumstances of opinion formation and expression. WP2 examines factual issues with established truths, investigating how affective polarization can lead to diverging interpretations of perceived “truths” that impact the accuracy of beliefs across different populations. We expect to see certain social influences being capable of deviating beliefs away from the ground truth, demonstrating the fragility of commonly shared interpretations of facts. This will extend our understanding of the social mechanisms behind the spread of misinformation and counter-narratives. WP3 addresses polarization’s impact on collective action, studying how affective polarization can disrupt cooperative behavior across political divides, potentially influencing collective responses to societal challenges. We further consider how unsuccessful collective action can exacerbate existing affective polarization, creating a feedback loop that further entrenches division. Together, our project aims to provide new insights into the mechanics of polarization in both opinion and factual realms. By integrating sociological theories with controlled online experiments and agent-based modeling, this research seeks to provide a richer understanding of how individual and group behaviors are influenced by network structures and influence dynamics. The findings are expected to inform strategies for mitigating polarization's negative effects.

• DFG Programme: Research Grants
• International Connection: Italy
• Cooperation Partners: Giulia Andrighetto, Ph.D.; Professor Dr. Arnout van de Rijt