Occult Belief: Seeing Is Believing

Abstract

In Experiment 1 an experimental analog was developed to an apparent context effect, in which a performance that arouses little excitement when performed by a magician may induce occult beliefs when performed by an alleged psychic. An accomplice presented three common magician's tricks, which resembled performances, to six introductory psychology classes. In the two classes comprising the Psychic condition, the instructors skeptically introduced the accomplice as an alleged psychic; in the Weak Magic condition, as an amateur magician; in the Strong Magic condition, as an amateur magician who would perform stunts which resembled psychic phenomena, but were really not. Belief was assessed through free-form written response, in the form of feedback to the performer. These instructional sets succeeded in manipulating proportion of occult belief. However, proportion of occult belief was above 50% and far exceeding magic beliefs in each experimental condition, even though, as indicated by a manipulation check, subjects in the Magic conditions heard and understood the instructors' assertions that the accomplice was a magician who would be faking a psychic performance. In paper-and-pencil tests in Experiments 2 and 3, where the performance was verbally described, different instructional sets which were similar to those of Experiment 1 but which reminded subjects explicitly of the logical plausibility of the trickery hypothesis, were presented in a within-S and a between-S design. In analogy with Experiment 1, psychic belief was prevalent and only slightly reduced by logically contradictory information. It was concluded that prevalent prior belief accounted for the tendency to ascribe psychic powers to the performer, and that reasoning fallacies accounted for the resistance of belief to a variety of disconfirming cognitive inputs.