Discovering Differences in Strategic Behavior Between Humans and LLMs
Not on the map. Why?
Abstract (arXiv)
As Large Language Models (LLMs) are increasingly deployed in social and strategic scenarios, it becomes critical to understand where and why their behavior diverges from that of humans. While behavioral game theory (BGT) provides a framework for analyzing behavior, existing models do not fully capture the idiosyncratic behavior of humans or black-box, non-human agents like LLMs. We employ AlphaEvolve, a cutting-edge program discovery tool, to directly discover interpretable models of human and LLM behavior from data, thereby enabling open-ended discovery of structural factors driving human and LLM behavior. Our analysis on iterated rock-paper-scissors reveals that frontier LLMs can be capable of deeper strategic behavior than humans. These results provide a foundation for understanding structural differences driving differences in human and LLM behavior in strategic interactions.
Key facts. Cited by 2 papers in the map (data through 7 Oct 2026). Builds on 35 of its 96 references that are in the map. First posted on arXiv 10 Feb 2026.
Builds on
35 of 96 references.
References · 35
- 2022Training language models to follow instructions with human feedbackcited by 15,712
- 2014On the Properties of Neural Machine Translation: Encoder-Decoder Approachescited by 1,303
- 2022Using Large Language Models to Simulate Multiple Humans and Replicate Human Subject Studiescited by 506
- 2025Reasoning Models Don't Always Say What They Thinkcited by 388
- 2022Social Simulacra: Creating Populated Prototypes for Social Computing Systemscited by 300
- 2023Large Language Models Empowered Agent-based Modeling and Simulation: A Survey and Perspectivescited by 253
- 2023Can Large Language Models Transform Computational Social Science?cited by 245
- 2022Neural Theory-of-Mind? On the Limits of Social Intelligence in Large LMscited by 220
- 2023Playing repeated games with Large Language Modelscited by 213
- 2023EconAgent: Large Language Model-Empowered Agents for Simulating Macroeconomic Activitiescited by 154
- 2024How Well Can LLMs Negotiate? NegotiationArena Platform and Analysiscited by 93
- 2023Can Large Language Models Serve as Rational Players in Game Theory? A Systematic Analysiscited by 91
23 more
- 2024Nicer Than Humans: How do Large Language Models Behave in the Prisoner's Dilemma?cited by 62
- 2024CogBench: a large language model walks into a psychology labcited by 60
- 2024Large Language Models and Cognitive Science: A Comprehensive Review of Similarities, Differences, and Challengescited by 48
- 2024Hypothetical Minds: Scaffolding Theory of Mind for Multi-Agent Tasks with Large Language Modelscited by 45
- 2024Decision-Making Behavior Evaluation Framework for LLMs under Uncertain Contextcited by 41
- 2024Centaur: a foundation model of human cognitioncited by 34
- 2025Virtual Agent Economiescited by 17
- 2025Language Agents as Digital Representatives in Collective Decision-Makingcited by 15
- 2024Capturing the Complexity of Human Strategic Decision-Making with Machine Learningcited by 14
- 2025Game Theory Meets Large Language Models: A Systematic Survey with Taxonomy and New Frontierscited by 14
- 2025Generating Computational Cognitive Models using Large Language Modelscited by 12
- 2025Strategic Tradeoffs Between Humans and AI in Multi-Agent Bargainingcited by 11
- 2025Game Reasoning Arena: A Framework and Benchmark for Assessing Reasoning Capabilities of Large Language Models via Game Playcited by 10
- 2024Do Large Language Models Learn Human-Like Strategic Preferences?cited by 8
- 2025Strategic Intelligence in Large Language Models: Evidence from evolutionary Game Theorycited by 6
- 2025AI Testing Should Account for Sophisticated Strategic Behaviourcited by 6
- 2023Population-based Evaluation in Repeated Rock-Paper-Scissors as a Benchmark for Multiagent Reinforcement Learningcited by 5
- 2025To Mask or to Mirror: Human-AI Alignment in Collective Reasoningcited by 5
- 2023Assessing Large Language Models' ability to predict how humans balance self-interest and the interest of otherscited by 4
- 2014Social cycling and conditional responses in the Rock-Paper-Scissors gamecited by 3
- 2024An Experimental Study of Competitive Market Behavior Through LLMscited by 3
- 2025Redefining Research Crowdsourcing: Incorporating Human Feedback with LLM-Powered Digital Twinscited by 3
- 2025An Interpretable Automated Mechanism Design Framework with Large Language Modelscited by 2
Built on by
Citing papers per year: 2026: 2.
2 papers.