When a series of three successive to-be-reported items (targets) is displayed in a rapid serial visual presentation (RSVP) stream of distractors, it has been shown that no attentional blink--a marked impairment in the report of the second of two targets, typically observed when the targets appear within 200-600 ms of one another--occurs in target accuracy. The present study examines three recently introduced computational models that provide different explanations of this protracted sparing effect. Using a standard RSVP design and these models, we provide empirical data and simulations that illustrate that structural limitations affect the processing of successive targets. In addition, we compare the candidate mechanisms that might underlie these limitations.
