Modeling individual differences in contrapositive reasoning with continuous latent state and trait variables

Abstract

Recent developments in the psychology of judgment and reasoning have emphasized the importance of individual differences in responses to reasoning experiments. This shift in emphasis, from the modal performance to variations in performance, raises some methodological concerns about the reliability and validity of the scores obtained when measuring reasoning performance. Differences in these scores might reflect measurement errors or reactions to pragmatic and semantic contextualization rather than differences in reasoning ability. This article shows how latent state-trait modeling can be applied to the responses collected in standard reasoning experiments. A latent state-trait analysis of contrapositive reasoning (if p then q; not-q; what follows?) revealed that the variables used to assess contrapositive reasoning ability showed remarkably good reliability and are only weakly affected by situation-specific influences. Consequently, they are appropriate for assessing stable interindividual differences in abilities. Moreover, the hypothesis of the existence of a general task-independent reasoning ability has to be rejected because the different task-specific abilities are only moderately correlated (between .27 and .63).