Abstract
In daily life, we continually evaluate the physical structure of our environments to plan our decisions and actions – e.g., packing a grocery bag in a way that will not crush the items at the bottom or building a sandcastle that does not topple. The physical scenarios that we regularly encounter can vary dramatically in their contents, yet we can intuitively understand their underlying physical structure and predict their future behavior. How? We might draw on domain-general mental resources to reason about physical behavior in an ad-hoc fashion, but an intriguing alternative is that we possess a dedicated set of mental resources for physical inference, allowing us to flexibly apply the principles of classical mechanics to any scene we encounter. Here, we tested for a dedicated mental resource for intuitive physics by evaluating individual differences in performance on a battery of intuitive physics tasks. We presented participants with five tasks, spanning a broad range of physical scenarios: 1) stability judgments of block towers; 2) trajectory predictions for bouncing pucks; 3) mass inferences from observed collisions; 4) predictions of collision outcomes based on objects’ masses, velocities, and geometry; and 5) mass inferences from observation of people’s interactions with the objects of interest. A factor analysis revealed that this collection of tasks draws on a common underlying mental resource that is distinct from other facets of cognition such as spatial reasoning and working memory. Further, we found that a reduced task battery (lasting ~15 minutes) captured ~90% of the reliable variance of the full battery, providing a measure that can be efficiently applied in a variety of contexts to evaluate intuitive physics abilities (e.g., in clinical populations, in longitudinal studies, or in training studies). Our findings point to a specialized mental resource for intuitive physics that is distinct from other cognitive domains.