Opinion dynamics and mutual influence with LLM agents through dialog simulation

TL;DR

This paper introduces a simulation framework using large language model agents in structured dialogs to study opinion dynamics, bridging classical models and modern AI systems, especially useful when real-world data is scarce.

Contribution

The paper presents a novel LLM-based simulation framework that incorporates classical opinion dynamics models like DeGroot and Friedkin-Johnsen, enabling scalable analysis of opinion formation.

Findings

Framework effectively simulates opinion influence and anchoring effects.

Bridges classical opinion models with modern LLM multi-agent systems.

Provides a scalable tool for opinion dynamics research with limited data.

Abstract

A fundamental challenge in opinion dynamics research is the scarcity of real-world longitudinal opinion data, which complicates the validation of theoretical models. To address this, we propose a novel simulation framework using large language model (LLM) agents in structured multi-round dialogs. Each agent's dialog history is iteratively updated with its own previously stated opinions and those of others analogous to the classical DeGroot model. Furthermore, by retaining each agent's initial opinion throughout the dialog, we simulate anchoring effects consistent with the Friedkin-Johnsen model of opinion dynamics. Our framework thus bridges classical opinion dynamics models and modern multi-agent LLM systems, providing a scalable tool for simulating and analyzing opinion formation when real-world data is limited or inaccessible.