Abstract
In a color-singleton search task, Os located an odd-colored target (e.g., a red diamond among greens) and discriminated its shape (whether the diamond was “chipped” on its left or right corner). These trials were intermixed with passively-viewed, uniform-colored arrays. We have previously reported (VSS, 2002) that Os are faster to perform the color-singleton search task when the distractor color was previewed than when the target color was previewed — the distractor-color preview effect (DCPE). Our current research addresses three questions: First, we previously measured the RT difference only between these two types of repetition. Thus, it was unclear whether this difference reflected a benefit of distractor-color preview or a cost of target-color preview. We measured the DCPE relative to an achromatic baseline and found evidence of both a cost and benefit of color preview, consistent with our previous claim that preview of colored items reduces the salience of same-colored items during subsequent search. Second, we also found that the DCPE was tolerant to changes in local item features (e.g., shape, size, etc.). Interestingly, connecting preview items into certain global forms eliminated the DCPE. We thus hypothesized that the DCPE required processing of color in a search-relevant context. This hypothesis was evaluated by a speeded classification task. Third, in determining the mechanism by which increased target salience produces the DCPE, we previously ruled out an account based on low-level color adaptation. We now report an analogous, context-dependent preview effect for motion-singleton search, suggesting the preview effect is not confined to color processing. We will discuss the potential effect on singleton detection and focal attention processes.