<p>When searching visual scenes, we use low-level visual information from objects’ defining features such as color and luminance contrasts. What is the relative influence of color and luminance for saccade target selection? Basic perceptual research suggests that we are not very sensitive to peripheral color, yet color is thought to be an important basic feature guiding visual search. Previous gaze-contingent research shows that targets can be localized faster in color than in grayscale scenes, therefore the availability of color in the visual periphery indeed helps visual search. However, object boundaries are typically defined by both color and luminance contrasts. Here we study the isolated roles of color and luminance during object-in-scene search by presenting either color-only or luminance-only contrasts in peripheral vision, using a gaze-contingent moving-window display with three varying window sizes. We found that peripheral target selection and search performance were more efficient with luminance contrasts, whereas color was used only sparingly beyond the parafovea. We conclude that color contrasts in peripheral vision are only efficiently used in scene search when they are jointly occurring with luminance contrasts.</p>

错误:搜索内容不能为空,请输入英文关键词
错误:关键词超出字数限制,请精简
高级检索

Color needs luminance for visual selection during scene search

  • Anke Cajar,
  • Jochen Laubrock

摘要

When searching visual scenes, we use low-level visual information from objects’ defining features such as color and luminance contrasts. What is the relative influence of color and luminance for saccade target selection? Basic perceptual research suggests that we are not very sensitive to peripheral color, yet color is thought to be an important basic feature guiding visual search. Previous gaze-contingent research shows that targets can be localized faster in color than in grayscale scenes, therefore the availability of color in the visual periphery indeed helps visual search. However, object boundaries are typically defined by both color and luminance contrasts. Here we study the isolated roles of color and luminance during object-in-scene search by presenting either color-only or luminance-only contrasts in peripheral vision, using a gaze-contingent moving-window display with three varying window sizes. We found that peripheral target selection and search performance were more efficient with luminance contrasts, whereas color was used only sparingly beyond the parafovea. We conclude that color contrasts in peripheral vision are only efficiently used in scene search when they are jointly occurring with luminance contrasts.