The impact of negative attentional set upon target processing in RSVP: an ERP study

This study investigates whether the negative attentional set, a form of top-down attentional bias, can be set up on a trial-by-trial basis and impair online target processing in an RSVP (Rapid Serial Visual Presentation) task in which two targets are to be identified. Using the N2pc (N2 posterior contralateral) - a component in the event-related potential (ERP) evoked by lateralized targets - as an index of attentional selection, we demonstrated that the online processing of the second target (T2) can be inhibited by a category-specific negative attentional set elicited by a special distractor (D1) prior to the first target (T1) and that this attentional set can be set up at an abstract, conceptual level. A digit T2 was presented on the left or right following a central RSVP letter stream which had a unique red letter T1. Another digit or a Chinese number character was presented prior to T1 as D1, which had to be ignored. Relative to the D1 absent condition, either type of D1 impaired T2 performance and delayed the N2pc response to T2. D1 elicited a frontocentral N2 peaking at about 300 ms post-onset of D1, suggesting that D1 is indeed an inhibition-evoking stimulus. A further behavioral experiment ruled out the possibility that D1 impairs T2 performance via attentional capture or a category-unspecific, general negative attentional set.