mlsanityData to 14 Oct 2026

Using Cognitive Models to Improve Language Model Simulation of Human Persuasion Games

arXiv 2606.17657 ↗First posted 16 Jun 2026cs.AIcited by 2

Not on the map. Why?

Abstract (arXiv)

People make decisions differently in strategic interactions. Some update beliefs like a Bayesian; others exhibit biases like motivated reasoning. Although creators of large language models use simulated humans for safety evaluations and training, they often fail to cover this breadth of human behavior. We argue that cognitive science and economics provide a convenient tool for doing so, making use of mathematical models of human decision-making. We propose an approach that we call Equation-to-Behavior Prompting for guiding large language models to match cognitive models, and evaluate this approach on persuasion games based on legal decision-making. We find that large models can approximate equation-based specifications -- Bayesian updating, affine distortion, motivated updating, and Grether's α-β model -- using prompting, but small models fail to do so. However, training small models with reinforcement learning to adhere to mathematical rules, Equation-to-Behavior RL, reduces belief error by 26.5% in out-of-distribution parameterizations. We show that these simulations can help create diverse training environments; training small models to consider different kinds of decision-makers improves average belief change by 2.5%--12% over Bayesian-only training, even when persuading GPT-5-mini. Our work could improve human simulations for training and evaluation in increasingly realistic settings, and could also enable novel research into more complicated mathematical models of human decision-making.

Key facts. Cited by 2 papers in the map (data through 14 Oct 2026). Builds on 35 of its 72 references that are in the map. First posted on arXiv 16 Jun 2026.

Builds on

35 of 72 references.

References · 35

  1. 2024DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Modelscited by 9,379
  2. 2024GPT-4o System Cardcited by 4,588
  3. 2021An Explanation of In-context Learning as Implicit Bayesian Inferencecited by 938
  4. 2024Debating with More Persuasive LLMs Leads to More Truthful Answerscited by 266
  5. 2024LLM Agents Grounded in Self-Reports Enable General-Purpose Simulation of Individualscited by 249
  6. 2024Personalizing Reinforcement Learning from Human Feedback with Variational Preference Learningcited by 155
  7. 2024Personalized Language Modeling from Personalized Human Feedbackcited by 132
  8. 2023Generative agent-based modeling with actions grounded in physical, social, or digital space using Concordiacited by 126
  9. 2024On the Conversational Persuasiveness of Large Language Models: A Randomized Controlled Trialcited by 109
  10. 2024How Well Can LLMs Negotiate? NegotiationArena Platform and Analysiscited by 100
  11. 2025SPIRAL: Self-Play on Zero-Sum Games Incentivizes Reasoning via Multi-Agent Multi-Turn Reinforcement Learningcited by 100
  12. 2023Turning large language models into cognitive modelscited by 67
23 more
  1. 2024Is In-Context Learning in Large Language Models Bayesian? A Martingale Perspectivecited by 57
  2. 2024Large Language Models Assume People are More Rational than We Really arecited by 54
  3. 2025Chasing Moving Targets with Online Self-Play Reinforcement Learning for Safer Language Modelscited by 39
  4. 2025Consistently Simulating Human Personas with Multi-Turn Reinforcement Learningcited by 39
  5. 2024A Mechanism-Based Approach to Mitigating Harms from Persuasive Generative AIcited by 34
  6. 2024Incoherent Probability Judgments in Large Language Modelscited by 30
  7. 2026NitroGen: An Open Foundation Model for Generalist Gaming Agentscited by 27
  8. 2025Measuring General Intelligence with Generated Gamescited by 24
  9. 2025Open Character Training: Shaping the Persona of AI Assistants through Constitutional AIcited by 24
  10. 2025The Levers of Political Persuasion with Conversational AIcited by 20
  11. 2025Toward Efficient Exploration by Large Language Model Agentscited by 17
  12. 2024Efficacy of Language Model Self-Play in Non-Zero-Sum Gamescited by 15
  13. 2025Evaluating Generalization Capabilities of LLM-Based Agents in Mixed-Motive Scenarios Using Concordiacited by 12
  14. 2026AI Gamestore: Scalable, Open-Ended Evaluation of Machine General Intelligence with Human Gamescited by 12
  15. 2025Levels of Analysis for Large Language Modelscited by 10
  16. 2025Towards Strategic Persuasion with Language Modelscited by 10
  17. 2025Are Large Language Models Sensitive to the Motives Behind Communication?cited by 8
  18. 2026Commercial Persuasion in AI-Mediated Conversationscited by 7
  19. 2025Emergent Alignment via Competitioncited by 5
  20. 2025Post-Training LLMs as Better Decision-Making Agents: A Regret-Minimization Approachcited by 5
  21. 2025Information Design With Large Language Modelscited by 4
  22. 2024Persuasion, Delegation, and Private Information in Algorithm-Assisted Decisionscited by 3
  23. 2025Friend or Foe: Delegating to an AI Whose Alignment is Unknowncited by 2

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 2.

2 papers.

  1. 2026RealCompanion: Benchmarking Human Understanding from Reasoning over Longitudinal Real-World Conversationscited by 0
  2. 2026Understanding and Reducing Misalignment in Persuasion Optimizationcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2023Human Choice Prediction in Language-based Persuasion Games: Simulation-based Off-Policy Evaluation
  2. 2025Noise, Adaptation, and Strategy: Assessing LLM Fidelity in Decision-Making
  3. 2026The Internal Anatomy of Strategic Choice in Large Language Models
  4. 2024Can LLMs Replace Economic Choice Prediction Labs? The Case of Language-based Persuasion Games
  5. 2025Using Reinforcement Learning to Train Large Language Models to Explain Human Decisions
  6. 2024Simulating Human Strategic Behavior: Comparing Single and Multi-agent LLMs
  7. 2024Large Model Strategic Thinking, Small Model Efficiency: Transferring Theory of Mind in Large Language Models
  8. 2025Bias-Adjusted LLM Agents for Human-Like Decision-Making via Behavioral Economics