Abstract
The constructive nature of the brain, such as when visual percepts occur in regions with no direct stimulation, can be investigated through the motion processing system. Here, we investigated the perception of illusory visual phantoms, a form of completion that occurs when two low-contrast collinear drifting gratings are separated by a homogeneous region, leading to the ghostly impression of stripes that extend across the gap. Previous reports have suggested that a motion aftereffect (MAE) can be induced by visual phantoms (Weisstein, 1977; Science); however, our preliminary investigations with static test patterns failed to replicate these reports. We therefore explored whether visual phantoms might elicit an MAE using a dynamic counterphasing test pattern, as higher-level forms of motion perception can often elicit a dynamic MAE (dMAE) even if no static MAE occurs. Our displays consisted of two collinear horizontal gratings (12° × 6°, 0.25 cyc/deg), separated by a vertical (5°) gap, that drifted (2 Hz) during an adaptation phase. A counterphasing grating appeared in the blank gap location during test. Either the darkest portion of the inducer gratings matched the background luminance to elicit vivid phantoms, or the mean luminance of the inducer gratings matched the background luminance as a control condition. Three contrast levels were tested: 2.73%,13.64%, and 63.64%. Participants reported whether the test grating moved upward, downward or whether motion was ambiguous. dMAE was strongly biased in the motion direction opposite to that of the adapting motion direction, but curiously, it appeared equally strong for the visual phantom condition and phantom control condition. These results suggest that even though salient visual phantoms are perceived in some conditions but not others, the perceptual representation of the phantoms does not seem to provide direct input to the visual motion processing system.