Reason-Mediated Behavioral Models for Auditing LLM Social Simulators

-->

Computer Science > Artificial Intelligence

arXiv:2607.24649v1 (cs)

[Submitted on 27 Jul 2026]

Title: Reason-Mediated Behavioral Models for Auditing LLM Social Simulators

Authors: Atharva Pandey , Gautam Jajoo

View a PDF of the paper titled Reason-Mediated Behavioral Models for Auditing LLM Social Simulators, by Atharva Pandey and 1 other authors

View PDF HTML (experimental)

Abstract: Large language models are increasingly used as social simulators, including as synthetic survey respondents. Most evaluations ask whether simulated outcomes resemble human outcomes. We argue that this is necessary but too weak: a simulator can match the final answer while using the wrong rationale-derived reason pattern. We study this problem through a 94-person sunscreen concept test in which each respondent evaluated three product concepts and wrote open-ended rationales. We map those rationales into signed reason states $Z$, where positive signs support adoption and negative signs block it. This gives a practical audit: holding respondent descriptors $D$, category context $K$, and concept treatment $X$ fixed, do human rationale-derived reasons help predict behavior $Y$, and can an LLM simulate the same reason state without seeing the human rationale or outcome? Human rationale-derived reasons substantially improve held-out prediction of purchase intent. LLM-simulated reasons are more

…