Abstract
When an observer monitors a stream of possible targets at fixation, a peripheral distractor can capture attention under some circumstances. Previous research has shown that a distractor sharing the defining feature of the target is more likely to capture attention than one that does not, an effect known as contingent capture (e.g., Folk, Leber, & Egeth, 2002). This has often been attributed to, and taken as evidence that, the operation of feature-based attention occurs in parallel across the visual field, despite knowledge that the target will be at fixation. The current experiments test the validity of this perspective by exploring the spatial specificity of this effect. In these tasks, observers viewed an RSVP stream at fixation (10 letters/s) and reported the identity of a target letter that was defined by a specific color. On some trials, a peripheral distractor appeared 200 or 500 ms before the appearance of the central target. In the contingent condition, the distractor possessed the target color, while in the noncontingent condition it did not. Critically, the eccentricity of this peripheral distractor was manipulated. Attentional capture is quantified as failure to report the identity of the target letter in the central stream. Experiment 1 tested the magnitude of attentional capture at a range of distractor eccentricities (1°, 1.5°, 2.5°, or 4.5°). Noncontingent distractors showed no attentional capture, regardless of eccentricity. Notably, attentional capture in the contingent condition was reduced as distractor eccentricity increased across this range of parafoveal eccentricities. This effect can be attributed to an inhomogeneity in the bottom-up color signal available in the periphery, or it may reflect an interaction between feature-based attention and the distribution of spatial attention during the task. In Experiment 2, we manipulated the amount of attentional focus required to do the central task and found evidence suggesting the latter hypothesis.
Meeting abstract presented at VSS 2013