Efficient Visual Heuristics in the Perception of Physical Object Properties

A key to interacting with the physical world is the ability to infer object properties, such as elasticity, from vision, allowing us to anticipate an objects behavior in advance. This kind of perceptual inference challenges current AI systems--highlighting the complexity of the underlying computations. How does the human brain solve this task? Here, we propose a resource-rational model based on learned statistics of object motion to explain how humans judge elasticity. We created 100,000 physics-based simulations of bouncing objects with different elasticities and found that even tiny changes in initial conditions (e.g., orientation) yield starkly different trajectories. Yet, across these simulations, we identified 23 motion features that capture natural variations in elasticity. Although a weighted combination of these features reliably predicts physical elasticity, surprisingly, humans do not seem to employ cue combination when judging elasticity. Instead, we found that observers flexibly switch between different cues, i.e., heuristics. A series of experiments designed to carefully tease apart several competing heuristics, suggests that observers switch between different computationally efficient yet highly informative heuristics depending on the information available in the stimulus.