Year: 2026
Type: preprint
Abstract: Mentalization - the ability to infer others' beliefs and intentions to guide one's own choices - is a key cognitive function underlying human social interactions. Large language models (LLMs) demonstrate behaviour consistent with humans on theory-of-mind tasks, yet whether these models can guide adaptive behaviour through mentalization is unknown. Here we use two economic games with cognitive computational modeling to uncover the latent strategies underlying mentalization in LLMs. We tested individual LLM agents across four model families, DeepSeek, GPT-4.1, GPT-5 and Gemini 2.0 Flash (N = 2,099), against opponents of varying sophistication and examined whether a prompting strategy designed to elicit strategic reasoning improved performance. We benchmarked results against human participants (N = 251) as a comparative measure. Across both games, LLMs showed clear behavioural and computational signatures of mentalizing that differed markedly by model provider and size. Strategic prompting generally improved performance by inducing more sophisticated reasoning, yet the extent of the benefit differed across the two tasks. Last, GPT-5 agents flexibly adapted their recursive depth of reasoning to increasingly sophisticated opponents, demonstrating superior performance to human participants. Collectively, we demonstrate different capacities for mentalization across LLMs, and highlight cognitive computational modeling as a formal method for assessing comparative intelligence across humans and machines.
Source: arXiv (Cornell University)
Institutions: Virginia Tech, University of Birmingham
Subfield: Developmental and Educational Psychology
Field: Psychology
Domain: Social Sciences
Open Access status: green
Awards: G1089062559, G1318905810, G1740661754, Neurocomputational mechanisms of prosocial behaviour in health, development and disorder, PROFUSE: Fusing brain and computational mechanisms of prosocial behaviour across health, development and disorder, G4159152195, Modelling variability in the social brain across the lifespan, DTP Bid Led by University of Birmingham, Modelling variability in the social brain across the lifespan, G558895819, Neurocomputational mechanisms of prosocial behaviour in health, development and disorder, G6484005240, G6601929330, Hypatia: Health Modelling, Made Simple, Hypatia: Health Modelling, Made Simple, G8326345491
OpenAlex ID: W7204584626
Cites:
Cited by:
Related to: