Independent Effects of Emotional Expression and Group Membership in the Evaluative Priming Task

Research on automatic evaluative responses to faces varying in emotional expression and ethnicity has yielded conflicting results. Some paradigms, like the Approach/Avoidance task, demonstrated interactive evaluation. In contrast, recent studies using the Evaluative Priming Task (EPT) yielded independent effects of expression and ethnicity. One key difference between these paradigms is the task relevance of the faces. In the EPT faces served solely as primes without direct relevance to the task. To examine whether increased task relevance could engender interactive processing in the EPT, we utilized a modified version of the "bona fide pipeline" EPT. In this adaptation, participants categorized the valence of target words succeeding prime faces followed by probe faces. Participants then judged whether the prime and probe faces depicted the same person, thereby adding task relevance to the prime faces. Experiment 1 revealed independent priming effects of emotion and ethnicity. Since error data and inverse efficiency scores provided evidence for an interactive evaluation, we replicated Experiment 1 using a sequential Bayes testing strategy. Experiment 2 confirmed that the effects of emotion and ethnicity remain independent, indicating that increased task relevance did not yield the integrated processing of emotion and ethnicity as initially hypothesized.