Proceedings of the Annual Meeting of the Cognitive Science Society

According to existing accounts of causation, people rely on asingle criterion to identify the cause of an event. Thephenomenon of causal illusions raises problems for suchviews. Causal illusions arise when a particular factor isperceived to be causal despite knowledge indicatingotherwise. According to what we will call the Dual-ProcessHypothesis of Causal Identification, identifying a causeinvolves two cognitive processes: 1) an automatic, intuitiveprocess that identifies possible causes on the basis ofperceptual cues (spatial and temporal) and 2) a slow,reflective process that identifies possible causes on the basisof causal inference, in particular, a consideration of possiblemechanism. Consistent with this hypothesis, we found that inresponse to a causal illusion shown in a naturalistic setting,people’s initial judgments of causation were higher than theirultimate judgments of causation (Experiment 1). Using anonline measure of the time-course of people’s causaljudgments, we found that people initially view animations ofcausal illusions as causal before concluding that they are non-causal (Experiment 2). Finally, we obtained similar resultsusing a deadline procedure, while also finding that the lowerthe cognitive reflectiveness (as measured by the CRT), thestronger people’s impressions of causation were (Experiment3). Implications for different classes of theories of causationare discussed.