The waterfall illusion can be transferred between vision and touch

If you look at a waterfall for about 30 seconds, and then shift your gaze to a nearby stationary object, such as a rock or a tree, that object will seem to drift slowly upwards. This well known optical illusion demonstrates a phenomenon called the motion after-effect, which is thought to occur as a result of adaptation – the brain compensates for movement in one direction, causing us to momentarily perceive a stationary objects to be moving in the other.

Although illusory motion can also be induced in the sense of touch, the brain is thought to process visual and tactile motion separately. But now researchers from MIT have found that not only can moving visual stimuli induce a tactile motion after-effect, but also that moving tactile stimuli can induce a visual motion after-effect. The findings, which are published in Current Biology, show that the senses of vision and touch are closely linked, and that each can influence the other. 

Several recent studies have shown that tactile stimuli elicit strong responses in regions of the brain known to process visual motion. Following on from this, Talia Konkle of the McGovern Institute for Brain Science and her colleagues investigated whether motion after-effects could be transferred from vision to touch and vice versa. Participants sat in front of a monitor which displayed visual stimuli consisting of alternating black and white horizontal bars which moved up or down. Immediately after each visual stimulus, patterns of moving or stationary tactile stimuli were applied to the tip of the participants’ right index finger, with a small piezoelectric device.

All of the participants consistently perceived the stationary tactile stimuli applied to their fingertip to be moving, in the opposite direction to that of the movement of the visual stimulus presented immediately beforehand. The motion after-effects induced by the visual stimuli had been transferred to the sense of touch, producing a tactile version of the waterfall illusion. In another set of experiments, the researchers found that the after-effects could also be transferred the other way. When moving tactile stimuli were applied to their fingertips, the participants perceived a stationary horizontal bar which was then presented to them on the computer screen to be moving. Again, the bar was perceived to be moving up following a downward sweeping motion applied to the fingertip, and down following an upward stroke.

The findings demonstrate that the visual and tactile sensory pathways interact with each other, even at the earliest stages of processing. Thus, after-effects can be transferred between the two senses, so that the perceptual experience in one modality can be influenced by a stimulus presented previously to the other modality. This suggests that both sensory pathways contain adaptive filters which are tuned to specific properties of stimuli in the other pathway, and which determine how subsequent stimuli in their own modality will be processed and perceived.

It is already known that vision modulates the sense of touch, but this study is the first to show that the relationship between the two sense is reciprocal, and that touch can also modulate the sense of vision. Rather than being processed separately, visual and tactile motion stimuli are apparently processed in at least partly overlapping neural pathways, which contain nodes that process both types of information. The results do not reveal which parts of the brain might encode these shared neural representations, but visual areas V5 amd MT+, which have already been shown to be activated by moving tactile stimuli, are good candidates.


Konkle, T. et al (2009). Motion Aftereffects Transfer between Touch and Vision. Curr. Biol. DOI: 10.1016/j.cub.2009.03.035.

4 thoughts on “The waterfall illusion can be transferred between vision and touch

  1. A startling motion after effect that I have experienced numerous times may fall into this category. Pulling into a parking space, I am applying the brakes and at the exact instance I am percieving the car stopping; the car next to me is backing out. Pressing harder and harder on the brake to stop the percieved continued forward motion, I realize that the vehicle next to me is moving and I have already stopped. For this to occur, I believe that an element of perfect coincidental timing and a measure of situational unawareness is involved.

Comments are closed.