Search for: All records

How do people perform general-purpose physical reasoning across a variety of scenarios in everyday life? Across two stud ies with seven different physical scenarios, we asked participants to predict whether or where two objects will make contact. People achieved high accuracy and were highly consistent with each other in their predictions. We hypothesize that this robust generalization is a consequence of mental simulations of noisy physics. We designed an “intuitive physics engine” model to capture this generalizable simulation. We find that this model generalized in human-like ways to unseen stimuli and to a different query of predictions. We evaluated several state-of-the-art deep learning and scene feature models on the same task and found that they could not explain human predictions as well. This study provides evidence that human’s robust generalization in physics predictions are supported by a probabilistic simulation model, and suggests the need for structure in learned dynamics models. 

Many models of intuitive physical reasoning posit some kind of mental simulation mechanism, yet everyday environments frequently contain far more objects than people could plausibly represent with their limited cognitive capacity. What determines which objects are actually included in our representations? We asked participants to predict how a ball will bounce through a complex field of obstacles, and probed working memory for objects in the scene that were more and less likely to be relevant to the ball’s trajectory. We evaluate different accounts of relevance and find that successful object memory is best predicted by how frequently a ball’s trajectory is expected to contact that object under a probabilistic simulation model. This suggests that people construct representations for mental simulation efficiently and dynamically, on the fly, by adding objects “just in time”: only when they are expected to become relevant for the next stage of simulation.