The effect of premise order in conditional reasoning: a test of the mental model theory

Abstract

The difference in difficulty between modus ponens ( if p then q; p; therefore q) and modus tollens ( if p then q; not-q; therefore not-p) arguments has been traditionally explained by assuming that the mind contains a rule for modus ponens, but not for modus tollens. According to the mental model theory, modus tollens is a more difficult deduction than modus ponens because people do not represent the case not-q in their initial model of the conditional. On the basis of this theory, we predicted that conditions in which reasoners are forced to represent the not-q case should improve correct performance on modus tollens. In particular, we predicted that the presentation of the minor premise ( not-q ) as the initial premise should produce facilitation. Experiment 1 showed that this is the case: whereas the inversion of the premise order did not affect modus ponens, it produced a significant increase of valid conclusions for modus tollens. Experiment 2 showed that this facilitation does not depend on the negative form (contrary vs. contradictory) of the minor premise. Experiments 3 and 4 (and/or some of their replications) demonstrated that facilitation also occurs when participants are asked to find the cases compatible with not-q or to evaluate a p conclusion. No premise order effect was found for sentences which make explicit the not-q case right from the start, i.e. p only if q conditionals and biconditionals (Experiments 5 and 6). Finally, Experiments 7 and 8 showed that the conditional fallacies are not significantly affected by the premise order.