Not all implicit measures of attitudes are created equal: Evidence from an embodiment perspective

We hypothesize that while evaluative priming involves proprioceptive cues, the IAT is representational due to its structural features and the specific algorithm upon which the IAT-effect rests. As predicted, evaluative priming is shown to rely on differential facial muscle activity while the IAT as a measurement instrument is not influenced by proprioceptive information. Evaluative priming does not yield differential responsiveness for congruent and incongruent trials when facial muscle activity is inhibited whereas the IAT-effect is shown to be impervious to such inhibition. Implications for the underlying mechanisms of implicit measures are discussed.