IntroductionMagic, understood as a set of effects designed to produce experiences of impossibility, offers a distinctive methodological potential for cognitive science because it makes it possible to manipulate, with ecological validity, processes such as attention, perception, reasoning, and memory. Despite increasing research interest over the last two decades, the available evidence has not been systematically mapped across cognitive domains, methodological approaches, and the neural correlates underlying magic perception or processing. The aim of this scoping review was to map and synthesize the empirical evidence on the relationship between illusionism and human cognitive processes across four cognitive domains: attention and misdirection, visual perception, high-level cognition, and epistemic emotion and memory. Neural correlates of the magical experience were examined as a fifth synthesis category.MethodsA systematic search was conducted in Web of Science, Scopus, PubMed, and APA PsycINFO, with no date restriction, and reported following PRISMA-ScR guidelines. Of the 1,633 records identified, 55 studies met the eligibility criteria. Findings reported in these studies were mapped and synthesized narratively.ResultsThe most consistent findings concerned the attentional domain, indicating that covert attention contributes substantially to misdirection across several experimental contexts, with convergent findings reported by multiple laboratories. With regard to high-level cognition, the reviewed evidence does not clearly support the idea that magicians possess privileged psychological knowledge. In epistemic emotion and memory, responses to physically impossible events are related to the early development of intuitive physical knowledge. At the neural level, the limited evidence available suggests that observing magical effects involves networks associated with conflict detection, predictive processing, and expectation violation, including frontal regions (DLPFC, ACC, IFG) and the caudate nucleus, providing a preliminary biological correlate to the cognitive phenomena described above.DiscussionOverall, the findings support the view that magic constitutes a valid and promising methodological tool for the study of human cognition, although the heterogeneity of the studies explored so far and the geographical concentration of the research limit the generalizability of the conclusions.