Proceedings of the Annual Meeting of the Cognitive Science Society

In this paper, we bring together research on active learningand intuitive physics to explore how people learn about“microworlds” with continuous spatiotemporal dynamics.Participants interacted with objects in simple two-dimensionalworlds governed by a physics simulator, with the goal ofidentifying latent physical properties such as mass, and forcesof attraction or repulsion. We find an advantage for activelearners over passive and yoked controls. Active participantsspontaneously performed several kinds of “natural exper-iments” which reveal the objects’ properties with varyingsuccess. While yoked participants’ judgments were affectedby the quality of the active participant they observed, they didnot share the learning advantage, performing no better thanpassive controls overall. We discuss possible explanations forthe divergence between active and yoked learners, and outlinefurther steps to categorize and explore active learning in thewild.