Modelling Opinion Dynamics at Scale with Deep MARL

2026-06-05

AI Summary

Traditional models of opinion dynamics rely on handcrafted local interaction rules, which struggle to uncover the genuine emergence mechanisms of macroscopic phenomena such as consensus and polarization in large-scale social networks. This work proposes a deep multi-agent reinforcement learning (MARL) framework wherein agents autonomously learn opinion interaction strategies through simple reward signals. We develop a GPU-accelerated consensus and truth-discovery game environment enabling efficient simulation with thousands of agents. Innovatively extending other-play to general-sum social interactions mitigates unrealistic conventions, while a graph attention mechanism recovers agent influence solely from network topology. Validation on a Bluesky subnet shows that high conformity best matches human behavioral data; however, in large static social media networks, such behavior substantially reduces collective accuracy and incentivizes strategic lying, whereas small dynamic networks remain unaffected or even benefit.

Abstract

Modelling opinion dynamics typically relies on hand-crafted local interaction rules to study emergent macroscopic phenomena such as consensus and polarisation. In contrast, multi-agent reinforcement learning (MARL) enables agents to learn such behaviours directly by optimising simple rewards. To explore the potential of MARL for opinion dynamics, we introduce a GPU-accelerated consensus and truth-finding game that scales to populations of up to 1000 agents, comparable to many real-world social sub-networks. To prevent unrealistic conventions, we extend other-play to general-sum social interactions. We next validate our model on a subset of the Bluesky network by recovering agent importance structures from graph topology alone via a learned attention layer, finding that highly conforming populations most closely match human data. In large social media networks such high levels of conformity significantly reduce collective accuracy and promote dishonest agents that lie to fit in. By contrast, small, dynamic hunter-gatherer networks are less affected; here, conformity can even improve collective agreement. This suggests a mismatch between evolved human conformity heuristics and modern social media environments as a potential contributor to misinformation.

Problem

• opinion dynamics
• conformity
• misinformation
• social networks
• collective accuracy

Innovation

• Deep MARL
• Opinion Dynamics
• Other-Play
• Attention Mechanism
• Collective Accuracy