A stochastic world model on gravity for stability inference

Abstract

The fact that objects without proper support will fall to the ground is not only a natural phenomenon, but also common sense in mind. Previous studies suggest that humans may infer objects’ stability through a world model that performs mental simulations with a priori knowledge of gravity acting upon the objects. Here we measured participants’ sensitivity to gravity’s direction, the most critical parameter of gravity in stability inference, to investigate how the world model works. We found that the world model was not a faithful replica of Newton’s law of gravity but rather encoded gravity’s direction as a Gaussian distribution, with the vertical direction as the maximum likelihood. The world model with this stochastic feature fit nicely with participants’ subjective sense of objects’ stability and explained the illusion that taller objects are perceived as more likely to fall. Furthermore, a computational model with reinforcement learning revealed that the stochastic feature likely originated from agent-environment interaction, and computer simulations illustrated the ecological advantage of the stochastic over deterministic representation of gravity’s direction in balancing accuracy and speed for efficient stability inference. In summary, the stochastic world model on gravity provides an example of how a priori knowledge of the physical world is implemented in the brain that helps humans operate flexibly in open-ended environments.