Abstract
The watercolor illusion (WCI), a color-spreading illusion induced by contrasting outer and inner borders, results in a perception of a pale illusory diffusion of a hue similar to the lighter border. This illusion is a strong figural cue similar to, and often stronger than, other Gestalt cues for figure-ground organization. In our present study, we examined the effect of the WCI on figure-ground assignment for regions with matched and non-matched luminance contrast. The goal was to examine the ability of the WCI to bias figure-ground assignment when luminance varies between regions of an image. Images consisted of a square divided into two parts by a vertical wavy contour. Each of these images had no WCI, WCI left, and WCI right conditions. Participants reported whether a probed region (either left or right) appeared to be the figure. Results showed a strong effect of the WCI when luminance was matched as white (Experiment 1) or gray (Experiment 3), demonstrating the WCI acts as a strong figural cue and is able to bias reversible stimuli. However, the WCI failed to bias figure-ground with contrasting luminance regions (Experiment 2). This study is the first to demonstrate how the WCI interacts with contrasting luminance regions in reversible figure-ground stimuli. These results enhance our understanding of color-spreading mechanisms and how they interact with luminance, contrast, and perceptual organization.The watercolor illusion (WCI), a color-spreading illusion induced by contrasting outer and inner borders, results in a perception of a pale illusory diffusion of a hue similar to the lighter border. This illusion is a strong figural cue similar to, and often stronger than, other Gestalt cues for figure-ground organization. In our present study, we examined the effect of the WCI on figure-ground assignment for regions with matched and non-matched luminance contrast. The goal was to examine the ability of the WCI to bias figure-ground assignment when luminance varies between regions of an image. Images consisted of a square divided into two parts by a vertical wavy contour. Each of these images had no WCI, WCI left, and WCI right conditions. Participants reported whether a probed region (either left or right) appeared to be the figure. Results showed a strong effect of the WCI when luminance was matched as white (Experiment 1) or gray (Experiment 3), demonstrating the WCI acts as a strong figural cue and is able to bias reversible stimuli. However, the WCI failed to bias figure-ground with contrasting luminance regions (Experiment 2). This study is the first to demonstrate how the WCI interacts with contrasting luminance regions in reversible figure-ground stimuli. These results enhance our understanding of color-spreading mechanisms and how they interact with luminance, contrast, and perceptual organization.