Proceedings of the Annual Meeting of the Cognitive Science Society

There are conflicting theories about how people reason through cause and effect. A key distinction between two prominent accounts pertains to whether, in judging an event’s causal relevance, people preferentially consider what actually happened (as predicted by process theories) or whether they also consider what could have happened under different conditions (as predicted by counterfactual theories). Toward adjudicating between these theories, the current work used eye tracking and Gaussian Process modeling to investigate how people form causal judgments retrospectively and in the absence of ongoing visual input. Participants played a virtual ball-shooting game: after choosing to move left or right, they encoded a video of the actual outcome and then were prompted to mentally simulate either (a) what actually happened, (b) what could have happened, or (c) what caused the outcome to happen while looking at a blank screen. During causal judgment, we found evidence that participants visually mentally simulated counterfactual possibilities: they moved their eyes in similar patterns as when they imagined a counterfactual alternative. Altogether, these results favor counterfactual theories of causal reasoning, demonstrate how visual mental simulation can support this reasoning, and provide a novel methodological approach for using eye movements to investigate causal reasoning and counterfactual thinking more broadly.