Evidence in favor of a resource depletion account of the attentional blink

The attentional blink (AB) refers to subjects' impaired ability to detect the second of two different target stimuli in a rapid serial visual presentation (RSVP) stream if they appear within 200–600 ms of one another. Traditionally, theoretical accounts of the phenomenon have postulated that the AB occurs due to limited attentional processes (such as the consolidation of information into working memory) being devoted to the first target at the expense of the second target (Resource Depletion accounts, e.g., Chun & Potter, 1995). Recently, however, several other mechanisms have been proposed to be responsible for the effect. One prominent alternative account is the temporary loss of control hypothesis (Di Lollo et al., 2005; Kawahara et al., in press), which proposes the AB occurs due to subjects' inability to maintain the current attentional set when presented with distractors. Here we tested whether resource depletion contributes to the AB. In Experiments 1 and 2, subjects viewed RSVP (100 ms/item) streams containing three sequentially presented letter targets amongst digit distractors. Across blocks the Target 2 stimulus was either drawn form the same category as Targets 1 and 3 (uniform condition) or from a different category (digits, varied condition). Replicating Di Lollo et al. (2005), we found that Target 1 and Target 3 performance differed in varied trials, but not (no AB) in uniform trials. However, an AB was observed under the uniform conditions when we increased the processing resources devoted to the first target, either explicitly by emphasizing Target 1 accuracy or implicitly when Target 1 was the first of the three targets featurally distinct from the distractors. A third experiment demonstrated that this AB was not due to a deficit in maintaining multiple targets in working memory. These results suggest that resource depletion does indeed contribute to the AB.