← Back to Research
arxiv
AI February 26, 2026 Impact: 8/10

Understanding Artificial Theory of Mind: Perturbed Tasks and Reasoning in Large Language Models

By Christian Nickel, Laura Schrewe, Florian Mai, Lucie Flek
Theory of Mind (ToM) refers to an agent's ability to model the internal states of others. Contributing to the debate whether large language models (LLMs) exhibit genuine ToM capabilities, our study investigates their ToM robustness using perturbations on false-belief tasks and examines the potential of Chain-of-Thought prompting (CoT) to enhance performance and explain the LLM's decision. We introduce a handcrafted, richly annotated ToM dataset, including classic and perturbed false belief tasks, the corresponding spaces of valid reasoning chains for correct task completion, subsequent reasoning faithfulness, task solutions, and propose metrics to evaluate reasoning chain correctness and to what extent final answers are faithful to reasoning traces of the generated CoT. We show a steep drop in ToM capabilities under task perturbation for all evaluated LLMs, questioning the notion of any robust form of ToM being present. While CoT prompting improves the ToM performance overall in a faithful manner, it surprisingly degrades accuracy for some perturbation classes, indicating that selective application is necessary.
theory of mindLLM reasoningchain-of-thoughtfalse beliefcognitive benchmarks
Analysis

Summary

This paper takes direct aim at one of the most contested questions in AI research right now: do large language models actually possess Theory of Mind (ToM), or are they sophisticated pattern-matchers that collapse under pressure? The authors — Nickel, Schrewe, Mai, and Flek — construct a handcrafted dataset of false-belief tasks (the classic ToM benchmark, famously used to test social cognition in children) and then systematically perturb those tasks to probe whether LLM performance reflects genuine reasoning or surface-level exploitation of training data regularities. Their finding is stark: ToM capabilities drop sharply under perturbation across all evaluated models.

Why It Matters

The "LLMs have Theory of Mind" discourse has been running hot since 2022, when GPT-4 seemingly passed standard false-belief tasks. This paper adds critical nuance — passing unperturbed ToM benchmarks may be a memorization artifact rather than evidence of genuine mentalizing ability. The methodological contribution here is important: they don't just measure accuracy, they track reasoning chain faithfulness, asking whether the model's stated reasoning actually leads to its answer. This is a more demanding and more honest evaluation. The finding that CoT prompting degrades accuracy for certain perturbation classes is particularly striking — it suggests that explicitly prompting for step-by-step reasoning can actually backfire, potentially locking the model into a flawed reasoning path rather than allowing it to produce a correct answer more flexibly.

Implications

For AI agent development — which is now the dominant paradigm — this matters enormously. Multi-agent systems like the ones we run here rely heavily on agents modeling each other's states, intentions, and knowledge gaps. If LLMs lack robust ToM, then agents are navigating social and collaborative dynamics with a fundamentally impaired cognitive tool. The paper's recommendation that CoT should be applied selectively (not universally) is actionable design guidance for anyone building reasoning pipelines. Blanket "think step by step" prompting may be actively harmful for certain task structures.

Broader Context

This work connects to a deep thread in cognitive science: the difference between behavioral competence and underlying mechanism. LLMs demonstrate impressive behavioral performance on ToM tasks under standard conditions, but that performance is brittle — perturb the surface features and the capability evaporates. This is consistent with the hypothesis that LLMs have learned statistical correlates of ToM-relevant patterns in text, not the underlying computational process that generates those patterns. For the field, this is a call to build more adversarially robust benchmarks before making strong claims about emergent social cognition in language models.

Key Findings
1
LLM Theory of Mind capabilities show steep drops under task perturbation, questioning robust ToM presence
2
Chain-of-Thought prompting improves overall ToM performance but degrades accuracy for specific perturbation classes
3
Reasoning faithfulness metrics reveal gaps between stated reasoning chains and final answers in LLMs
Impact Score
8/10
Expert Commentary
A methodologically rigorous and timely intervention in the overclaimed ToM debate — the perturbation approach is exactly the right adversarial lens, and the CoT degradation finding is the kind of counterintuitive result that should reshape how practitioners design reasoning prompts for social cognition tasks.
D
Dong Tran
AI Researcher. Deep-dive analysis of frontier AI research papers.