The Need for a Socially-Grounded Persona Framework for User Simulation

2026-01-12

arXiv.org

AI Summary

This work proposes SCOPE, a novel framework that integrates 141 fine-grained sociopsychological attributes—such as values and identity—into synthetic character modeling, moving beyond the limitations of coarse demographic proxies commonly used in existing approaches. Leveraging empirical data from 124 U.S. participants, the authors construct large language model–driven synthetic agents and evaluate them on the SimBench benchmark. Results demonstrate that demographic information alone accounts for only approximately 1.5% of behavioral variance, whereas incorporating sociopsychological dimensions substantially improves behavioral prediction accuracy, character similarity, and fairness. The SCOPE-generated agents outperform both default prompting strategies and NVIDIA’s Nemotron characters, underscoring the critical role of non-demographic factors in realistic user simulation.

Technology Category

• Humans and AI: Game Design — Virtual Humans, NPCs and Autonomous Characters
• Cognitive Modeling & Cognitive Systems: Simulating Human Behavior
• Application Domains: Humanities & Computational Social Science

Application Category

• User Modeling, Personalization and Recommendation: Psychology-informed user models and recommender systems
• Social Networks and Social Media: Generative AI / large language models and their impact on social systems
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions

Abstract

Synthetic personas are widely used to condition large language models (LLMs) for social simulation, yet most personas are still constructed from coarse sociodemographic attributes or summaries. We revisit persona creation by introducing SCOPE, a socially grounded framework for persona construction and evaluation, built from a 141-item, two-hour sociopsychological protocol collected from 124 U.S.-based participants. Across seven models, we find that demographic-only personas are a structural bottleneck: demographics explain only ~1.5% of variance in human response similarity. Adding sociopsychological facets improves behavioral prediction and reduces over-accentuation, and non-demographic personas based on values and identity achieve strong alignment with substantially lower bias. These trends generalize to SimBench (441 aligned questions), where SCOPE personas outperform default prompting and NVIDIA Nemotron personas, and SCOPE augmentation improves Nemotron-based personas. Our results indicate that persona quality depends on sociopsychological structure rather than demographic templates or summaries.

Problem

• persona
• social simulation
• sociopsychological
• user simulation
• large language models

Innovation

• socially-grounded persona
• sociopsychological framework
• user simulation
• behavioral alignment
• bias reduction