Abstract
When two targets are embedded in a rapid serial visual presentation stream of distractors, perception of the second target is impaired when the inter-target lag is relatively short (less than 500 ms). Stimuli concurrently presented with the stream can affect this phenomenon, which is called attentional blink (AB). Previous studies have yielded conflicting results concerning the direction of the effect of added distractors on the AB: some studies (e.g., Kawahara, 2003; Visser et al., 2004) report an increased AB, while others (e.g., Choo and Kim, 2006) report a decreased AB. The present study explored the boundary conditions of the exaggeration/reduction effects of distractors on the AB and investigated possible underlying mechanisms by manipulating the spatial configuration, timing, and type of distractors. Specifically, Experiments 1–3 manipulated the spatial uncertainty and spatial switching between the targets independently and found that location uncertainty of at least one of the targets was a critical factor in explaining the apparent inconsistency between the studies of Visser et al. (2004) and Choo and Kim (2006). The results of Experiments 4–6 indicated that the magnitude of the AB deficit increased, regardless of the type of distractors, when spatial uncertainty of the target locations was involved. Interestingly, the reduction effect occurred when task-irrelevant distractors were presented, but the effect depended on the similarity between the distractors and targets: if the irrelevant items were too similar or too dissimilar, the reduction effect did not occur. Moreover, the reduction effect occurred only when the second target was presented at subthreshold level. These results suggest the possible contribution of stochastic resonance or the centre-surround attentional mechanism in producing the effects of distractors on the AB deficit.