mlsanityData to 14 Oct 2026

Behavioral History Outperforms Descriptions of the Person for LLM Synthetic Personas

arXiv 2610.03998 ↗First posted 2 Oct 2026cs.AI cs.CL cs.CYcited by 0

Not on the map. Why?

Abstract (arXiv)

Large language models (LLMs) are increasingly used as synthetic personas representing survey respondents. Their validity as substitutes for particular respondents depends on whether they reproduce individuals' decisions. We examine what information helps synthetic respondents predict each individual's later choices, using five conditions that add progressively richer information: no personal information, demographics, personality traits, cognitive scores, and finally the respondent's earlier survey choices as behavioral history. We use a two-wave panel of 845 US adults who completed measures of 14 behavioral biases (spanning risk, time preferences, overconfidence, and reasoning), so each respondent's earlier answers provide a human test-retest benchmark; in the behavioral-history condition, all items that score the target bias are withheld. At the population level, the average number of biases per respondent in every condition is close to the human average (7.1-8.1 biases, against 7.1 for humans). This aggregate similarity masks differences in variance: persona descriptions recover only 53-67% of human between-person variation, whereas adding behavioral history restores it to approximately the human level. At the individual level, description-based personas achieve only 7-12% of the informedness observed in human test-retest responses, while adding behavioral history raises this to 28%. The condition including behavioral history has the highest estimated informedness in all 17 demographic groups, whereas description-based conditions provide little or no information for some groups. Synthetic responses also exhibit stronger education- and income-related differences than human responses. For LLM synthetic personas, a respondent's past answers add more to individual-level prediction than a description of who they are.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 33 of its 61 references that are in the map. First posted on arXiv 2 Oct 2026.

Builds on

33 of 61 references.

References · 33

  1. 2023Whose Opinions Do Language Models Reflect?cited by 773
  2. 2022Out of One, Many: Using Language Models to Simulate Human Samplescited by 752
  3. 2022Using Large Language Models to Simulate Multiple Humans and Replicate Human Subject Studiescited by 515
  4. 2024LLM Agents Grounded in Self-Reports Enable General-Purpose Simulation of Individualscited by 249
  5. 2023Bias Runs Deep: Implicit Reasoning Biases in Persona-Assigned LLMscited by 199
  6. 2024Quantifying the Persona Effect in LLM Simulationscited by 169
  7. 2024Large language models that replace human participants can harmfully misportray and flatten identity groupscited by 144
  8. 2025LLM Generated Persona is a Promise with a Catchcited by 81
  9. 2023Questioning the Survey Responses of Large Language Modelscited by 74
  10. 2023The Challenge of Using LLMs to Simulate Human Behavior: A Causal Inference Perspectivecited by 59
  11. 2024Large Language Models Assume People are More Rational than We Really arecited by 54
  12. 2025SimBench: Benchmarking the Ability of Large Language Models to Simulate Human Behaviorscited by 52
21 more
  1. 2024LLM economicus? Mapping the Behavioral Biases of LLMs via Utility Theorycited by 47
  2. 2023Diminished Diversity-of-Thought in a Standard Large Language Modelcited by 45
  3. 2024Large Language Models: An Applied Econometric Frameworkcited by 32
  4. 2024Evaluating language models as risk scorescited by 27
  5. 2024Language Models Trained to do Arithmetic Predict Human Risky and Intertemporal Choicecited by 12
  6. 2026Assessing the Reliability of Persona-Conditioned LLMs as Synthetic Survey Respondentscited by 10
  7. 2025How Personality Traits Shape LLM Risk-Taking Behaviourcited by 9
  8. 2025Missing the Margins: A Systematic Literature Review on the Demographic Representativeness of LLMscited by 9
  9. 2026Behavioral Economics of AI: LLM Biases and Correctionscited by 9
  10. 2025Evaluating and Aligning Human Economic Risk Preferences in LLMscited by 8
  11. 2026The Chameleon's Limit: Investigating Persona Collapse and Homogenization in Large Language Modelscited by 8
  12. 2026Plausible but Not Valid: A Psychometric Audit of LLMs as Synthetic Survey Respondentscited by 8
  13. 2026When Synthetic Users Fail: A Cross-Domain Benchmark of LLM-Simulated Human Survey Responsescited by 5
  14. 2026When Can Digital Personas Reliably Approximate Human Survey Findings?cited by 4
  15. 2025Pay What LLM Wants: Can LLM Simulate Economics Experiment with 522 Real-human Persona?cited by 2
  16. 2026The Collapse of Heterogeneity in Silicon Philosopherscited by 2
  17. 2026From Demographics to Survey Anchors: Evaluating LLM Agents for Modeling Retirement Attitudescited by 2
  18. 2026Synthetic Personalities: How Well Can LLMs Mimic Individual Respondents Using Socio-Economic Microdata?cited by 2
  19. 2026When Persona Attributes Improve Population Alignment in Large Language Modelscited by 2
  20. 2026Item-Mean Surrogates: Why Richer Persona Data Fail to Improve LLMs as Human Surrogatescited by 1
  21. 2026When Can LLM Digital Twins Reduce Human Measurement? From Behavioral Fidelity to Statistical Substitutabilitycited by 1

Built on by

Not cited yet.