Abstract
The motion after-effect (MAE) is generally observed after prolonged inspection of a moving stimulus, typically requiring tens of seconds of adaptation. However, neurophysiological studies have reported adaptation effects following briefly presented adapting stimuli (e.g., Muller et al., 1999). The putative functional roles of this rapid neural adaptation are improved coding efficiency and increased stimulus discriminability. Here, we investigated whether psychophysical adaptation effects can be observed at the timescale of reported neurophysiological effects.
In the main experiment, observers viewed a large, high-contrast adapting grating presented for 67ms (1c/deg, 15deg/s, temporal sum of contrast was 0, precluding afterimage generation). As confirmed by control experiments, this exposure duration was below the motion discrimination threshold, likely because of suppressive center-surround interactions (Tadin et al., 2003). Nevertheless, when presented with a stationary test stimulus, observers reported perceiving illusory motion in the direction opposite to the sub-threshold adapting motion. This was observed for adaptor-test ISIs of 0ms, 150ms, 400ms, but not for 1000ms.
In other experiments, we showed that this rapid MAE exhibits partial interocular transfer and is low-pass tuned to the spatial frequency of the adaptor, with the strongest MAE when the adapter and test have the same spatial frequency. Interestingly, removing the test stimulus and shortening adaptor presentation to 50ms (in order to generate a negative afterimage) also yielded a strong MAE, i.e., the resulting afterimage appeared to move.
In summary, we found that adaptation to brief moving stimuli yields perceivable MAEs, even when the exposure duration is below that required for above chance motion discrimination. Evidently, the MAE is not merely a perceptual illusion that follows prolonged exposure to a moving stimulus, but rather a process that can occur essentially every time we experience motion.