Simulating and Experimenting with Social Media Mobilization Using LLM Agents

2025-10-30

AI Summary

This study investigates how political mobilization messages on social media diffuse through peer influence and affect voter turnout. We propose a high-fidelity multi-agent simulation framework integrating U.S. Census demographic data, empirically grounded Twitter network topology, and heterogeneous large language model (LLM) agents powered by GPT-4.1 to jointly model individual cognition and collective dynamics. Our key contribution is the first embedding of LLM-based agents into an empirically calibrated social network—enabling reproducible, experimentally controllable simulations for counterfactual analysis and sensitivity testing. Experiments successfully replicate core empirical findings, including the superiority of social over informational mobilization and significant peer spillover effects. The framework demonstrates strong mechanistic interpretability and theoretical consistency, offering a novel methodological foundation for computational political science and digital governance research.

Technology Category

• Multiagent Systems: Agent-Based Simulation and Emergent Behavior
• Cognitive Modeling & Cognitive Systems: Social Cognition And Interaction
• Humans and AI: Voting

Application Category

• Social Networks and Social Media: Generative AI / large language models and their impact on social systems
• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation
• Responsible Web: Machine-in-the-loop, human agency and autonomy

Abstract

Online social networks have transformed the ways in which political mobilization messages are disseminated, raising new questions about how peer influence operates at scale. Building on the landmark 61-million-person Facebook experiment citep{bond201261}, we develop an agent-based simulation framework that integrates real U.S. Census demographic distributions, authentic Twitter network topology, and heterogeneous large language model (LLM) agents to examine the effect of mobilization messages on voter turnout. Each simulated agent is assigned demographic attributes, a personal political stance, and an LLM variant ( exttt{GPT-4.1}, exttt{GPT-4.1-Mini}, or exttt{GPT-4.1-Nano}) reflecting its political sophistication. Agents interact over realistic social network structures, receiving personalized feeds and dynamically updating their engagement behaviors and voting intentions. Experimental conditions replicate the informational and social mobilization treatments of the original Facebook study. Across scenarios, the simulator reproduces qualitative patterns observed in field experiments, including stronger mobilization effects under social message treatments and measurable peer spillovers. Our framework provides a controlled, reproducible environment for testing counterfactual designs and sensitivity analyses in political mobilization research, offering a bridge between high-validity field experiments and flexible computational modeling.footnote{Code and data available at https://github.com/CausalMP/LLM-SocioPol}

Problem

• Simulating social media mobilization effects using LLM agents
• Examining peer influence on voter turnout through agent interactions
• Bridging field experiments with computational modeling of political mobilization

Innovation

• Agent-based simulation with real census data
• LLM agents with political sophistication variants
• Reproduces mobilization patterns from field experiments