Everaert, T. , Spruyt, A. & De Houwer, J.
Ghent University
We present a series of studies that suggest that, in contrast to popular beliefs, automatic affective processing takes place only if attention is assigned to the affective stimulus dimension (Spruyt, De Houwer, & Hermans, 2009). In four experiments, we encouraged one group of participants to attend the affective stimulus dimension (the affective group) and another group to attend a non-affective, semantic stimulus dimension (the non-affective group). We used different tasks to measure affective processing in each study. In Experiment 1 and 2, we used more traditional measures of affective processing: the emotional Stroop task and the dot probe task. In Experiment 3, we performed a multidimensional scaling procedure on participants’ similarity judgments. This procedure yielded the weights participants assigned to the affective stimulus dimension and a non-affective stimulus dimension. In Experiment 4, we employed EEG to compare the size of the P3a component elicited by an affective and non-affective oddball stimulus. In all experiments, automatic affective processing was more pronounced in the affective group. In the non-affective group, we observed similar effects with regard to the non-affective stimulus dimension. Feature-specific attention allocation thus appears to play a crucial role in automatic affective processing.