The effects of prior beliefs in reasoning: An associational interpretation

Abstract

Van Duyne (1976) reports that the subjects’ performance on the Wason selection task is biased by their prior belief in the truth of the rule with which they are reasoning. Specifically, the subjects are reported as reasoning more logically if they believe the rule to be ‘sometimes true’ rather than ‘always true’. Van Duyne's analysis rests, however, upon an interpretation of the subjects’ verbal justifications; a reanalysis of his data shows no effects upon actual response frequencies. However, it was felt that a more powerful test might be obtained by use of rules explicitly thought to be true and false. Accordingly, the subjects were asked to generate statements that they thought to be always and usually true and always and usually false; and were subsequently presented with a selection task on each of the four types of statement. Falsity did have a significant effect upon response frequencies, in the direction of better logical performance. It is suggested that this result is not caused by the subjects’ intention to validate their beliefs (as Van Duyne proposes) but that the subjects’ prior experience leads to the logical answer having more associative cues on ‘false’ sentences.