Strategic reasoning under pressure: Testing heuristics in higher-order theory of mind

Abstract

Higher-order Theory of Mind (ToM+)—the recursive ability to understand that others have thoughts about thoughts—is pivotal to complex social interactions but can be cognitively demanding. This study examines how individuals cope when ToM+ reasoning exceeds their cognitive limits, contrasting a model predicting “blindness” to future states with a model predicting averaging over future states. Conducted on a new online platform called the Morality Game and using a series of 32 time-pressured sequential-choice games, participants' errors reveal consistent support for the probabilistic model, indicating that when precise higher-order reasoning is unfeasible, individuals do not simply ignore future possibilities. Instead, they approximate future states using probabilistic heuristics rather than explicit recursive reasoning. By clarifying how individuals rely on such heuristics when full deliberation is impractical, these findings provide a clearer framework for understanding the mental shortcuts that shape ToM+ under cognitive strain, thereby informing more precise methods for investigating and modeling complex social inference.