Abstract
Abstract: :
Purpose:The saliency model of visual search (Rosenholtz, Vision Research, 1999) predicts that background color should have an effect on asymmetries in visual search based on color. One set of experiments (Rosenholtz, Nagy, and Bell, submitted to VSS, 2002) showed that switching background color from white to red reversed the direction of search asymmetries for targets and distractors differing in saturation (L excitation) and varying from white to reddish in appearance. In the work reported here observers again searched for targets among distractors on white and red backgrounds. Targets and distractors differed in saturation (S excitation) varying from white to bluish in appearance (saturation condition), or they differed in hue varying from reddish to bluish in appearance (hue condition). Methods:Observers searched for a single known target stimulus among 53 homogeneous distractor stimuli all of the same color and luminance. Targets differed from the distractors only in chromaticity, but both targets and distractors were more luminous than the background. The stimuli were small disks (0.14 deg. diam.) and were presented in random locations within an area 4.25 deg in diameter on a color monitor. Target stimuli were included on 66% of the trials in each block. On each trial observers were required to determine whether a target was present and depress a response button as rapidly as possible. Response times for the initial button push were measured. Observers subsequently indicated whether the target was present or absent. Accuracy of 90% or better was enforced. Results:On a white background, response times were shorter when the bluer member of each pair of colors served as the target color in the saturation condition, but there was little asymmetry in the hue condition. When the same stimuli were presented on a red background there was little asymmetry in the saturation condition but an asymmetry was introduced in the hue condition with the bluer stimuli producing shorter response times. When an asymmetry was present, the size of the asymmetry typically increased as the difference between the pair of target-distractor colors and the background color increased. Conclusion:Results confirm that background color clearly affects asymmetries in visual search based on color. Both the presence and size of asymmetries in color search appear to depend heavily on the relationships among target and distractor chromaticities and background chromaticity.
Keywords: 362 color vision • 626 visual search