Chimpanzees infer the location of a reward on the basis of the effect of its weight

SummaryThe extent to which animals in general, and non-human primates in particular, understand physical causality is currently unclear [1,2]. One way to assess an animal's causal understanding is to test its ability to analyze a causal chain backwards — to infer cause from an effect [3]. In the study reported here, chimpanzees saw a given outcome (effect) of an action and had to infer the preceding event (cause) in order to solve the problem. More specifically, subjects saw a banana being hidden inside one of two opaque cups mounted on opposite sides of a balanced beam, but they were kept ignorant about the banana's exact location. Subsequently, the subjects witnessed the balance beam tilting to one side after the experimenter released it from its equilibrium position (the Balance condition). The chimpanzees preferentially (and from trial one) selected the lower, compared to the upper, cup. Two control conditions demonstrated that the chimpanzee subjects lacked an intrinsic preference for the lower cup when there was no movement involved (the Wedge condition) or when the balance beam was tilted by the experimenter's action (the Non-causal balance condition). We conclude that the chimpanzee subjects of our experiments demonstrated evidence of causal inference based on an object's weight