Abstract
Observers are able to search visual displays for multiple possible target objects, held in memory. In this so-called "Hybrid Search", reaction time increases with the log of the number of items held in memory (Wolfe 2012). What resources are consumed during the time required for the memory search? To investigate this question, we studied the effect that memory set size had on the Attentional Blink (AB) when identification of the first target (T1) required search of the memory set. In 3 experiments, observers memorized 2, 4, 8 or 16 items. In the subsequent AB experiment, the T1 task was to report the presence of a member of the memory set. In Experiment 1, the memory set and all members of the RSVP stream were letters. The T2 task was to identify a single red letter. In Experiment 2, the memory set and all members of the RSVP stream were objects. The T2 task was to judge which of four possible objects appeared in the red frame. We observed a significant interaction between AB magnitude and memory set size in both experiments: larger memory set size was associated with a greater deficit in reporting T2 at early lags. In Experiment 3, T1 was a memory set object that was embedded in a stream of numbers and a single red number served as T2. Unlike previous experiments, AB magnitude did not interact with memory set size. Taken together, our data suggest that the size of the memory set for T1 influences subsequent processing only when other items in the RSVP stream are similar to the target item. Simply invoking a memory search with T1 is not enough to impair subsequent processing of visual items if those items are in a different category (e.g. an object in a stream of digits).
Meeting abstract presented at VSS 2013