The attentional blink and task switching within and across modalities

The attentional blink (AB) is a robust deficit obtained for a second visual target (T2) appearing within 200 600 ms of a correctly identified first target (T1). In most AB studies both targets appear among distractors in a rapid serial visual presentation (RSVP), and a key variable is the lag or SOA between the two targets. In the present chapter we review research bearing on the basis for the AB deficit and related deficits, including cross-modal versions of the AB procedure and tasks that require a task switch (a switch in target criterion) between T1 and T2. We conclude that the standard AB deficit is restricted to visual targets and can be distinguished from an additional deficit that results from a task switch between T1 and T2. The latter effect is found with cross-modal and auditory stimuli as well as visual stimuli, and is additive with the AB effect (when both targets are visual). We propose that the standard AB effect occurs at a different stage of processing than the more central taskswitching deficit and shows features that distinguish it from the latter. The visual AB effect represents a limit in the speed with which visual targets — which are vulnerable to masking — can be consolidated into working memory or awareness: A second target may be lost while queuing for access to the consolidation process. We review studies that show a clear dissociation between AB and task-switching deficits, consistent with complementary findings of Allport and Hsieh (this volume) on criterion shifting in RSVP target search. The evidence suggests that there are multiple bottlenecks in processing which individually or together limit performance when two target stimuli must be processed close in time.