Abstract
Loading cognitive control processes such as working memory decreases performance in response conflict tasks by making it harder to prioritize targets over distractors (deFockert et al., 2001). Here we test whether the loss of cognitive control is dependent on how working memory is loaded. Specifically, if working memory is occupied by items that draw capacity away from either targets or distractors in a selective attention task, then we should observe different patterns of dual-task interference effects. Subjects performed a matching task while maintaining items in working memory. The primary matching task was to make a same/different judgment for two face targets, each occluding the center of a larger house distractor, presented side-by-side. The subjects only performed face matching, but the background house distractor pair was either congruent or incongruent to the face target responses (same/ different). These response conflict trials were performed with three types of concurrent working memory tasks: face memory, house memory and no memory. In face memory blocks, subjects maintained two faces in working memory while performing the matching task. In house memory blocks, subjects maintained two houses in working memory. The main result of interest was the response time difference between incongruent and congruent trials that reflect response conflict elicited by the house distractors. Significant response conflict was observed overall. Interestingly, this response conflict was reduced when subjects were maintaining houses in working memory compared to when they were maintaining faces or nothing at all. This suggests that the house working memory load reduced the capacity for perceptual processing of house distractors, attenuating response conflict. Thus, distractor interference does not always increase with increased working memory load, but is rather dependent on whether memory items share the same resources with targets or with distractors.