High Volatility and Action Bias Distinguish LLMs from Humans in Group Coordination

TL;DR

This paper compares human and large language model group coordination in a common-interest game, revealing that LLMs exhibit high volatility and action bias, unlike humans who adapt and stabilize over time.

Contribution

It introduces a behavioral diagnostic framework highlighting key differences in coordination strategies between humans and LLMs, emphasizing the coordination gap.

Findings

Humans improve and stabilize their behavior over time in the game.

LLMs often fail to improve and show excessive switching, impairing convergence.

Rich feedback benefits humans significantly but has limited effect on LLMs.

Abstract

Humans exhibit remarkable abilities to coordinate in groups. As large language models (LLMs) become more capable, it remains an open question whether they can demonstrate comparable adaptive coordination and whether they use the same strategies as humans. To investigate this, we compare LLM and human performance on a common-interest game with imperfect monitoring: Group Binary Search. In this n-player game, participants need to coordinate their actions to achieve a common objective. Players independently submit numerical values in an effort to collectively sum to a randomly assigned target number. Without direct communication, they rely on group feedback to iteratively adjust their submissions until they reach the target number. Our findings show that, unlike humans who adapt and stabilize their behavior over time, LLMs often fail to improve across games and exhibit excessive switching, which impairs group convergence. Moreover, richer feedback (e.g., numerical error magnitude) benefits humans substantially but has small effects on LLMs. Taken together, by grounding the analysis in human baselines and mechanism-level metrics, including reactivity scaling, switching dynamics, and learning across games, we point to differences in human and LLM groups and provide a behaviorally grounded diagnostic for closing the coordination gap.