Abstract
3 experiments investigate whether prior exposure to a scene results in a faster (lower RT) and/or more efficient search (reduction in the slope of the set size function). In Experiment 1, observers performed 6 consecutive searches of the same visual array (set sizes 3, 6, or 12), with a different target presented (orally) for each search. In Experiments 2 and 3, each array was searched once. Prior exposure was manipulated, on some trials, by a 2 s preview. Some of the arrays were also embedded in visual noise. Prior exposure produced faster RTs but did not affect search efficiency. The partial visual quality × preview interaction suggests that prior exposure allowed the storage of a limited number of items. If the target was one of those items, the observer responded without searching; if not, the observer searched the remaining items as if there was no preview. Results are consistent with a view of visual representations as fairly sparse.