Abstract
The present series of experiments examined a recent hypothesis by Margolis (1987) to explain the poor performance usually obtained on Wason's four-card selection task. According to Margolis, the difficulty stems mainly from scenario ambiguity in which the context is misperceived, and partially from semantic ambiguity in which the meaning of the if … then rule is misinterpreted. Reliable facilitation on an abstract version of the task was observed when these two ambiguities were eliminated. In order to observe the facilitation, however, the consequent part of the rule had to be phrased to prevent matching. These findings are discussed in terms of Evans's heuristic and analytic processing model of reasoning and Cheng and Holyoak's pragmatic reasoning schemas theory.