Abstract
When searching for a target, people implicitly remember the configuration of the search display, showing faster reaction times when the display is later repeated versus new (random) displays. This implicit configuration memory, called contextual cuing, occurs regardless of stimulus color or jittering, but disappears when the target is moved to a different location in a repeated display (Chun & Jiang, 1998). The relationship between targets and surrounding distractors is clearly important to contextual cuing and suggests that the closeness of distractors to the target may affect observers' memory for context. Targets are more difficult to detect as target-distractor proximities decrease within an individual display (Motter & Holsapple, 2000; Eriksen & Eriksen, 1974), but this effect has not been examined using repeated displays. In the experiments presented here, we investigate how stimulus density affects contextual cueing. We used a traditional search task comprised of rotated T's and L's where a left or right rotated T was the target and stimuli were either monochromatic or contained an equal number of red, yellow, green, and blue stimuli (as in Chun & Jiang, 1998). There were two display densities, sparse and dense, with an equal number of stimuli in each. For the sparse displays, the entire display was used, and for the dense displays, the stimuli were restricted to 1 of 4 quadrants. Each block of trials contained an equal number of new and repeated displays of both densities. Overall, subjects' reaction times were faster for dense displays, possibly because of the smaller search area. Interestingly, the contextual cueing effect was larger for dense displays. One possible explanation is that there are more items close to the fovea in the dense displays, requiring fewer fixations to perceive the configuration.