Prejudiced or Just Smart?: Intelligence as a Confounding Factor in the IAT Effect

Abstract

Implicit Association Tests (IATs) are well-known measures of implicit cognition, particularly attitudes. Previous studies reported that IATs are affected by method-specific variance: IAT effects of more intelligent people may appear smaller due to their reduced task-switch costs. In contrast, based on a theoretical framework that assumes IAT effects to depend on successful recoding of the congruent IAT task, larger IAT effects are expected for more intelligent people. We tested the hypothesis that intelligence can bias IAT effects with different IATs. General processing speed was also assessed. Two studies indicated faster and more intelligent participants to have larger IAT effects in some, but not all, IATs.