Trust, Lies, and Long Memories: Emergent Social Dynamics and Reputation in Multi-Round Avalon with LLM Agents

TL;DR

This paper investigates how LLM agents develop reputation and deception strategies over multiple rounds of Avalon, revealing emergent social behaviors driven by memory and reasoning effort.

Contribution

It demonstrates that cross-game memory leads to reputation formation and that increased reasoning effort enhances strategic deception in LLM agents.

Findings

Reputation is role-conditional and influences team selection.

Higher reasoning effort correlates with more sophisticated deception tactics.

Agents reference past interactions to inform current social strategies.

Abstract

We study emergent social dynamics in LLM agents playing The Resistance: Avalon, a hidden-role deception game. Unlike prior work on single-game performance, our agents play repeated games while retaining memory of previous interactions, including who played which roles and how they behaved, enabling us to study how social dynamics evolve. Across 188 games, two key phenomena emerge. First, reputation dynamics emerge organically when agents retain cross-game memory: agents reference past behavior in statements like "I am wary of repeating last game's mistake of over-trusting early success." These reputations are role-conditional: the same agent is described as "straightforward" when playing good but "subtle" when playing evil, and high-reputation players receive 46% more team inclusions. Second, higher reasoning effort supports more strategic deception: evil players more often pass early missions to build trust before sabotaging later ones, 75% in high-effort games vs 36% in low-effort games. Together, these findings show that repeated interaction with memory gives rise to measurable reputation and deception dynamics among LLM agents.