Enhancing LLMs with Cognitive-Affective Personality Inference for Simulating Human Social-Psychological Behavior

2026-10-06

AI Summary

This study addresses the limitation of static personality prompting in large language models (LLMs), where conflating stable traits with situational interpretations leads to poor alignment in socio-psychological simulations. To overcome this, we propose SPIN, a framework that pioneers a decoupled inference pipeline separating personality compilation from state induction. Through three zero-shot invocations, SPIN disentangles the personality core, cognitive-affective states, and decision outputs, explicitly modeling intermediate states to precisely distinguish stable traits from dynamic situational responses. Experimental results demonstrate that SPIN achieves state-of-the-art behavioral alignment across multiple LLMs and two categories of socio-psychological tasks. These findings validate the critical role of the decoupled architecture in enhancing the simulation fidelity of LLMs for computational social psychology.

Abstract

Large language models are increasingly used to simulate human participants in social and behavioral studies, yet static persona prompting typically maps a participant profile and an experimental scenario directly to a response, entangling stable dispositions with situation-specific interpretations. To address this limitation, we introduce \textbf{SPIN}, a cognitive-affective personality system-inspired inference pipeline for simulating human social-psychological behavior. Specifically, SPIN implements this structured inference process through three zero-shot LLM calls that compile a task-blind participant core, elicit condition-specific cognitive-affective states, and read out decisions from those states, thereby reusing stable personality structure while routing each trial-specific response through an explicit state representation. We evaluate SPIN on two reconstructed social-psychological study families spanning uncertainty reasoning and pluralistic ignorance, across four base LLMs. Compared with blank, demographic, narrative, and chain-of-thought prompt variants, SPIN consistently delivers the strongest overall alignment performance across base LLMs and study families. Ablations and state analyses further show that both personality compilation and structured state elicitation contribute to the gains, and that the elicited states shift interpretably across informational and normative conditions. These results suggest that structured personality-state inference can improve benchmark-level behavioral alignment beyond richer persona descriptions or generic multi-step reasoning.

Problem

• Large Language Models
• Human Behavior Simulation
• Social-Psychological Behavior
• Static Persona Prompting
• Cognitive-Affective Personality

Innovation

• Cognitive-Affective Personality System
• Zero-shot Inference Pipeline
• Social-Psychological Behavior Simulation
• Structured State Elicitation
• Behavioral Alignment