On Emergent Social World Models -- Evidence for Functional Integration of Theory of Mind and Pragmatic Reasoning in Language Models

TL;DR

This study explores whether language models develop interconnected social world models by integrating Theory of Mind and pragmatic reasoning, using behavioral and causal experiments inspired by cognitive neuroscience.

Contribution

It provides new ToM localizer data, methodological improvements, and empirical evidence supporting the functional integration of social cognition components in language models.

Findings

Evidence suggests interconnected social world models in LMs

Methodological advances improve localization of social cognition in models

Results support the emergence of integrated social reasoning capabilities

Abstract

This paper investigates whether LMs recruit shared computational mechanisms for general Theory of Mind (ToM) and language-specific pragmatic reasoning in order to contribute to the general question of whether LMs may be said to have emergent "social world models", i.e., representations of mental states that are repurposed across tasks (the functional integration hypothesis). Using behavioral evaluations and causal-mechanistic experiments via functional localization methods inspired by cognitive neuroscience, we analyze LMs' performance across seven subcategories of ToM abilities (Beaudoin et al., 2020) on a substantially larger localizer dataset than used in prior like-minded work. Results from stringent hypothesis-driven statistical testing offer suggestive evidence for the functional integration hypothesis, indicating that LMs may develop interconnected "social world models" rather than isolated competencies. This work contributes novel ToM localizer data, methodological refinements to functional localization techniques, and empirical insights into the emergence of social cognition in artificial systems.