Abstract
We usually have subjective experience that the entire visual field is in rich color, although color sensitivity is much reduced in the peripheral retina. This "pan-field color" phenomenon (Balas & Sinha, 2007) can be demonstrated by a failure to detect a large achromatized region in either the central or peripheral visual field, with briefly presented scene images. This study investigated the effects of color diagnosticity of scenes on the occurrence of pan-field color. Color diagnosticity refers to the degree to which a scene is associated with specific colors. If pan-field color is mediated by color completion based on color statistics of the scenes, color diagnosticity could enhance its occurrence. We used three window conditions with high and low color-diagnostic scene images subtending 28.6° × 24.3°; Full-color, color-center (only the central 21° region was colored), and gray-center (the central 13° region was achromatized) conditions. The coloring of the scene was also manipulated (normal and complementary colors) to control color-based scene segmentation. The scene image was presented for 70 ms, followed by a mask. The observer's task was to indicate whether the image had an achromatized region (scotoma). Results showed that observers more frequently missed the scotoma (i.e., pan-field color) in the normal than in the complementary color condition. However, the occurrence of pan-field color was similar with high and low color-diagnostic scenes in both color- and gray-center conditions. Analysis using signal detection theory showed that the scenes in normal color led to a higher sensitivity to the scotoma but also mediated a greater response bias for missing the scotoma. An additional experiment with longer stimulus duration showed that high color diagnosticity provided an advantage for detecting the scotoma, so that reducing the occurrence of pan-field color. Overall, these findings suggested that response bias, rather than color completion, mainly mediates pan-field color.
Meeting abstract presented at VSS 2017