Motivational relevance modulates the predictive validity of the implicit association test

With two experiments, we investigated the impact of the relevance of objects for current needs on the measurement of implicit attitudes with the Implicit Association Test (IAT; Greenwald, McGhee, & Schwartz, 1998). Experiment 1 (n = 103) showed that an IAT measuring implicit preference for beverages was predictive of both behavioral choice and explicit preference for thirsty, but not for quenched participants. This effect was specific for preferences toward thirst-related objects: Thirst did not increase the ability of a food-IAT to predict explicit preferences for foods. Experiment 2 (n = 142) conceptually replicated the moderation effects with a different need (i.e., to clean one's teeth), another type of product (toothpaste), and a different behavior (online shopping). It also ruled out abstraction as a mechanism underlying the effect of stimulus relevance. The discussion highlights the possible underlying mechanisms and the implications of these results for theory and measurement of implicit attitudes.