Abstract
After adapting to a moving pattern, subsequently viewed static patterns appear to move in the direction opposite to adapting motion (Wohlgemuth, 1911). Motion aftereffects (MAE) can also produce shifts in the perceived spatial position of static test stimuli (Snowden, 1998). Here, we extend these results by showing how the MAE shifts the position of non-static, translating stimuli. Participants adapted to a large sine-wave grating, presented in their peripheral vision, which drifted vertically either upwards or downwards for 5 seconds. This adapting stimulus was then replaced by a small Gabor patch that moved horizontally for 1 second. Participants then reported the angle of the patch’s motion. This procedure was repeated 6 times per trial, each time following 5 seconds of adaptation, allowing us to measure changes in the perceived angle after various adaptation durations. In all cases, the perceived angle of the Gabor’s drift deviated from its actual horizontal trajectory in the direction opposite to adapting motion. Our results suggest that the MAE causes positional shifts in a non-static, translating stimulus and that these offsets accumulate over the 1 second test presentation, generating a linear, angled trajectory. The angle of offset from horizontal also increased significantly with adaptation duration, from 9.5° following 5 seconds of adaptation, to 15° following 30 seconds of adaptation, equivalent to a vertical position shift of 0.96 dva across test presentation. This is larger than the 0.5 dva shift reported by Snowden (1998) under relatively similar adaptation conditions but with a static test. The angled trajectories we report here reflect some combination of the Gabor’s actual external motion vector and a vector opposite to the adapting stimulus, the MAE. The mechanism underlying this combination of motion vectors may be similar to that observed in the double-drift illusion (Cavanagh & Tse, 2019; Tse & Hsieh, 2006).