POSIM: A Multi-Agent Simulation Framework for Social Media Public Opinion Evolution and Governance

2026-03-24

AI Summary

This study addresses the limitations of existing models in capturing the cognitive processes, irrational behaviors, and multi-stage dynamics of social media users during opinion evolution—factors that hinder effective governance. To overcome these challenges, we propose a multi-agent simulation framework that integrates large language models (LLMs) with the Belief–Desire–Intention (BDI) architecture. Our approach embeds user interactions, social network structures, and recommendation mechanisms within a virtual social environment, while leveraging a Hawkes point process to drive the temporal evolution of events. Notably, we incorporate irrational factors into the LLM-driven cognitive model for the first time, using real interaction chains to uncover an “empathy paradox”: empathy-driven interventions may inadvertently amplify negative sentiment. Validated against Weibo data, the framework successfully reproduces opinion dynamics from individual to collective levels and enables prospective governance intervention experiments.

Technology Category

• Multiagent Systems: Agent-Based Simulation and Emergent Behavior
• Cognitive Modeling & Cognitive Systems: Social Cognition And Interaction
• Knowledge Representation and Reasoning: Reasoning with Beliefs

Application Category

• Social Networks and Social Media: Media and governance, opinion dynamics, filter bubbles, polarization
• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation
• Web Mining and Content Analysis: Sentiment analysis and opinion mining

Abstract

Modeling social media public opinion evolution is essential for governance decision-making. Traditional epidemic models and rule-based agent-based models (ABMs) fail to capture the cognitive processes and adaptive behaviors of real users. Recent large language model (LLM)-based social simulations can reproduce group-level phenomena like polarization and conformity, yet remain unable to recreate the irrational interactions and multi-phase dynamics of real public opinion events. We present POSIM (Public Opinion Simulator), a multi-agent simulation framework for social media public opinion evolution and governance. POSIM integrates LLM-driven agents with a Belief--Desire--Intention (BDI) cognitive architecture that accounts for irrational factors, places them in a virtual social media environment with social networks and recommendation mechanisms, and drives temporal dynamics through a Hawkes point process engine that captures the co-evolution of agents and the environment across event phases. To validate the framework, we collect real-world public opinion datasets from the Weibo platform covering the full interaction chain of users. Experiments show that POSIM successfully reproduces key characteristics of public opinion evolution from individual mechanisms to collective phenomena, and its effectiveness is further supported by multiple statistical metrics. Building on POSIM, governance-oriented guidance and intervention experiments uncover a counterintuitive empathy paradox: empathetic guidance deepens negative sentiment instead of easing it under certain conditions, offering new insights for governance strategy design. These results demonstrate that the proposed framework can fully serve as a computational experimentation platform for proactive strategy evaluation and evidence-based governance. All source code is available at https://github.com/DeepCogLab/posim/.

Problem

• public opinion evolution
• social media simulation
• irrational behavior
• multi-phase dynamics
• governance

Innovation

• LLM-based multi-agent simulation
• BDI cognitive architecture
• Hawkes point process
• public opinion evolution
• computational governance