Are LLM Agents Behaviorally Coherent? Latent Profiles for Social Simulation

ICLR 2026 Conference Submission22303 Authors

Keywords: LLM, Agent, Simulation

Abstract:

The impressive capabilities of Large Language Models (LLMs) have fueled the
notion that synthetic agents can serve as substitutes for real participants in human-
subject research. To evaluate this claim, prior research has largely focused on
whether LLM-generated survey responses align with those produced by human
respondents whom the LLMs are prompted to represent. In contrast, we address
a more fundamental question: Do agents maintain internal consistency, retaining
similar behaviors when examined under different experimental settings? To this
end, we develop a study designed to (a) reveal the agent’s internal state and (b)
examine agent behavior in a conversational setting. This design enables us to
explore a set of behavioral hypotheses to assess whether an agent’s conversational
behavior is consistent with what we would expect from its revealed internal state.
Our findings show significant internal inconsistencies in LLMs across model families
and at differing model sizes. Most importantly, we find that, although agents may
generate responses matching those of their human counterparts, they fail to be
internally consistent, representing a critical gap in their capabilities to accurately
substitute for real participants in human-subject research.

Primary Area: applications to neuroscience & cognitive science

Submission Number: 22303