Probabilistic Simulation Predicts Human Performance on Viscous Fluid-Pouring Problem

Abstract

The physical behavior of moving fluids is highly complex, yet people interact with them daily with relative ease. To investigate how humans achieve this remarkable ability, the present study extended the classical water-pouring problem (Schwartz & Black, 1999) to examine how humans consider physical properties of fluids (e.g., viscosity) and perceptual variables (e.g., volume) in a reasoning task. We found that humans do not rely on simple heuristic rules to reason about fluid dynamics. Instead, they rely on perceived viscosity and volume to make their judgments. Computational results from a probabilistic simulation model reliably account for human sensitivity to latent fluid attributes and their performance on our task. In contrast, non-simulation models based on statistical learning fail to fit human performance. Our results provide converging evidence supporting mental simulation strategies in physical reasoning, and outline experimental conditions that rectify the dissociation between explicit prediction and tacit judgment.


Back to Table of Contents