Persona-Based Simulation of Human Opinion at Population Scale

TL;DR

This paper presents SPIRIT, a framework that creates psychologically grounded personas from social media to simulate individual opinions and behaviors at scale, surpassing traditional demographic-based models.

Contribution

The paper introduces SPIRIT, a novel method for inferring semi-structured personas from social media to enable realistic simulation of human opinions and actions.

Findings

SPIRIT-conditioned simulations better match self-reported responses.

The approach reproduces human-like heterogeneity in responses.

Persona banks serve as virtual respondent panels for opinion studies.

Abstract

What does it mean to model a person, not merely to predict isolated responses, preferences, or behaviors, but to simulate how an individual interprets events, forms opinions, makes judgments, and acts consistently across contexts? This question matters because social science requires not only observing and predicting human outcomes, but also simulating interventions and their consequences. Although large language models (LLMs) can generate human-like answers, most existing approaches remain predictive, relying on demographic correlations rather than representations of individuals themselves. We introduce SPIRIT (Semi-structured Persona Inference and Reasoning for Individualized Trajectories), a framework designed explicitly for simulation rather than prediction. SPIRIT infers psychologically grounded, semi-structured personas from public social media posts, integrating structured attributes (e.g., personality traits and world beliefs) with unstructured narrative text reflecting values and lived experience. These personas prompt LLM-based agents to act as specific individuals when answering survey questions or responding to events. Using the Ipsos KnowledgePanel, a nationally representative probability sample of U.S. adults, we show that SPIRIT-conditioned simulations recover self-reported responses more faithfully than demographic persona and reproduce human-like heterogeneity in response patterns. We further demonstrate that persona banks can function as virtual respondent panels for studying both stable attitudes and time-sensitive public opinion.