Amber Callan | Visual Perceptual Aftereffects: McCollough Effect and Waterfall Illusion
portfolio_page-template-default,single,single-portfolio_page,postid-420,qode-quick-links-1.0,ajax_fade,page_not_loaded,,vertical_menu_enabled,side_area_uncovered_from_content,qode-theme-ver-11.1,qode-theme-bridge,wpb-js-composer js-comp-ver-5.1.1,vc_responsive

In my Psychology of Perception class, we learned about how the user perceives the world around them. All of the senses can be tricked, but if you know how the senses work, you can either manipulate them to achieve the outcome you want or learn to avoid making these perceptual mistakes.

Therefore, I decided to study two visual perceptual aftereffects in depth to get a better idea of how our visual system works

After extended exposure to a stimulus, the body begins to adjust and adapt. While every system has tricks and responses for unique environments, the visual system prominently displays some of the more interesting adaptations called perceptual after effects. Two widely-known types of visual after effects are negative after images and motion after effects. For negative after images, once an image is focused on for an extended period of time, the receptors in the retina become fatigued. When you break that focus and look at other objects, the tired receptors may distort them and bring up falsely colored images. In motion aftereffects, the visual system goes from seeing a moving picture to a still one. When looking at the still picture afterwards, the motion still appears to be there. Both after images and motion aftereffects are false information displayed by the visual cortex, yet each uses their own pathway to create a unique perception.

First off, negative after images are the simplest visual aftereffect yet still one of the hardest phenomena to explain. Once the eye is exposed to an image and then exposed to a blank surface, it appears as if the person “sees” the same image but with the opposite colors as they previously saw. This is because the receptors in the retina of the eye tire easily if exposed to the same stimulus (Pomerantz, Class). When the stimulus gets switched, the receptors still hold onto their coding of the previous image because they have gotten used to seeing it for so long. Then false colors and other perceptions can appear in a new image though they are not physically present. If the cones perceive a red image, the afterimage appears to be green while if cones look at a yellow image, the afterimage will appear to be blue. The same works with black and white. As an example, look at the American flag on the image below for thirty seconds and then glance at the bottom blank part of the page. Typically viewers will see the American flag in red and blue even though they just saw it in yellow and green.



An example of opponent-process theory. If the flag is looked at for a significant amount of time and then the viewer switches their gaze to a white surface (such as the bottom of this sheet of paper), they tend to see the American flag normally with blue by the stars and red in the stripes.


The real question on why these aftereffects happen was answered in the 18th century when Thomas Young proposed the “trichromatic theory”. In the trichromatic theory, viewers have three types of cones (short, medium and long) that code for unique wavelengths; red, green, and blue (Svaetichin, 1956). While this theory explains how we see in color, it cannot explain why afterimages occur. This is where Hering’s theory of opponent process elaborates and builds on the ideas Young originally thought of. It attempts to explain after images using three separate channels, shown below. The short, medium and long cones are used in different combinations to perceive two channels of color and one channel of light (Hurvich, 1957). It shows that red and green plus blue and yellow are each opposites in their wavelengths. Therefore in those channels, you could perceive a variety of colors, but the most striking ones would be the opposite pairings. This is also why looking at opposites such as orange and blue causes the eye more stress than colors closer together on the wheel (McCamy, Marcus & Davidson, 1976). In the brightness channel of black and white, images are perceived as one of the opposites; light or dark. They work alongside the color channels in parallel motion to see the color of an image in addition to its brightness.



The basic assumption of opponent-process theory is that color systems are in pairs. The chromatic systems differentiate between color contrast while the achromatic system differentiates between brightness contrast. Source: RGB Opponent. (2008). Retrieved November 18, 2013, from


While most images work as a basic afterimage, the McCollough effect is a more specific example of the negative afterimage (McCollough, 1965). Looking at two images of stripes (one vertical and one horizontal) with different colors before looking at black and white comprehensive images will show the black and white lines as colored (shown in the first image below). This uses the basic premise of the opponent process theory but the two different orientations of the lines create a unique experience (Humphrey, 1998). The after effect combines both previously viewed images and shows the opposing colors in the corresponding lines. Therefore if a viewer saw green stripes horizontally in the original image, they would see red stripes in the afterimage. If the orientation of the lines is changed ever so slightly, the effect disappears (shown in the second image below). At an extreme 45 degree angle, the effect disappears completely. However, at angles in between 0, 45 and 90 degrees, the greater the angle difference, the less the effect shows (McCollough & Webster, 2011). The timing of the McCollough effect changes the strength of the afterimage as well. If the exposure time increases then the lasting aftereffect will increase (Riggs, White, & Eimas, 1974). Some studies even suggest that the McCollough effect has a specific timescale for a quick and a slow experiment. In the quick one, the effect can rise and fall within thirty seconds, but in the long experiment the effect can be strong for around five to eight minutes (Vul, Krizay & MacLeod, 2008).



Figure 1. If a person were to look at the two images on the right individually for a significant amount of time, they would then see the left image with green vertical stripes and red horizontal stripes. This is the most used figure to illustrate the “McCollough Effect”.



This is a prime example of a motion aftereffect. If you look at the middle of the circle for 15 seconds and stare at another image or object, it will appear to be moving! If the middle circle or the outside lines were to be moving, there would still be an opposite detected motion in the other direction.


Some have tried to explain the McCollough effect through associative and neural adaptation (Bedford, 1995). The associative adaptation describes two individual elements that are never connected, but when brought together in a full image, the connection is formed. When the two striped patterns are looked at alone, they still linger in the visual system until connected (Barlow, 1990). The neural adaptation model solely describes on the neurons that are focused on the stimuli and uses their fatigue as a reason for the McCollough effect (Stromeyer, 1978). Others have thought of the McCollough effect in terms of an “error-correcting device” which adjusts the internal representation of an image to external world possibilities (Dodwell & Humphrey, 1990).

Next, motion aftereffects use the same parameters as the negative after images. If the receptors are exposed to a moving image for an extended period of time, their receptors will adjust to the motion. Then if the image is switched to a still image, viewers still tend to see the opposite motion in the moving image displayed on the new image (Mather, Verstraten & Anstis 1998). This follows along with afterimages in that the receptors adjust and then are required quickly to return to a new baseline. Some researchers believe that once the motion is coded as a baseline for all possible directions of motion that any other object is seen in reference to that motion (Barlow & Hill, 1963). Others believe that the neurons that code for one direction of motion are more excitable for the opposite direction. Therefore once they fatigue and a motionless stimulus is presented, those neurons excite and the illusion of an opposite motion occurs (Bajaj, 2013).

A more specific example of a motion aftereffect is the waterfall illusion. If a viewer looks at a waterfall for a minute and then glances at motionless rocks on the side, the rocks will appear to be moving upwards (Schrater & Simoncelli 1998).  First recognized by Robert Addams in 1834, the waterfall illusion can be found in nature and has been studied by countless researchers. Some believe that this effect is created by having two simultaneous stimuli next to each other and rapidly switching the coding between the two creates the illusion (Crane 1988). They have compared this to the Necker cube (the shell of a 3D cube) where the concavity of the object switches from concave to convex depending on which orientation a viewer thinks about. Others think that the waterfall and rocks are individual stimuli and that the fatigue is the only cause of the illusion. The first theory lends itself to an associative adaption theory while the second is closely related to the neural adaptation model.

Negative after images and motion aftereffects both display false information through the visual system so humans perceive something that is not present. Most viewers tend to realize that what they are seeing is not present when they experience these false perceptions. However this still does not stop the effect from happening (Loftus 1992). They both are a direct result of tired neurons and can be interpreted through fatigue. The neurons adapt to one stimulus and when the stimulus is switched, they slowly adapt to the new one. In both cases of after images and motion aftereffects, the lasting effect of the first stimulus impacts the next. The opponent process theory helps to explain both of these phenomena. In the McCollough effect, the red and green shown transform into their opposites in the cohesive image. In the waterfall illusion, the top to bottom motion from the waterfall translates into a bottom to top motion on the rocks.



A further explanation of the McCollough effect with pre and post adaptations. The above diagram displays the channels with unique orientations and the preferred color for that orientation. The decreasing bar size indicates the adapting gratings
Source: Celeste McCollough Howard and Michael A. Webster. (2011). McCollough effect. Scholarpedia, 6 (2), 8175.


The difference in these two perceptual aftereffects is in what exactly they are coding. For the negative after image, the coding is for color while the motion aftereffect clearly codes for motion. The pathways for these two processes are different in the brain as well. In the brain, vision is processed in the occipital lobe and follows either the dorsal or ventral pathway, shown in the image below. For motion, the visual cortex is activated along with the V5 extrastriate cortex which is a region of the occipital lobe close to the primary visual cortex (Guy, 2008). Researchers have also found that small areas surrounding the V5 extrastriate cortex show activity on brain activity measurements for unique motion tasks (Zeki, Watson, and Frackowiak, 1993). This suggests that the cortex might be more specialized than we can measure. For color perception, the brain activity can mainly be found in the visual cortex and surprisingly in the language processing lobe of the brain (Siok, et. al).



The dorsal and ventral steam of motion processing. The dorsal stream is responsible for detecting motion as it shows activity during the waterfall illusion and other motion after effects.


In conclusion, visual perceptual aftereffects may differ in pathway and coding material but are very similar in their explanation. Because neurons generally have the same function, it would make sense that they excite and inhibit in similar patterns. Neurons weaken over time through fatigue and cannot perform as well when they are constantly in demand. The pathways differ because the information that is being sensed differs. For motion, the sensations must travel from the eyes to the V5 cortex while for color. For color, the sensations travel to the primary visual cortex and to the language center. All in all, they make for an amazing optical illusion.



[columns] [span6]

Addams, R. (1834). An account of a peculiar optical phenomenon seen after having looked at a moving body. London and Edinburgh Philosophical Magazine and Journal of Science, 5, 373–374

Bajaj K. (2013). Visual Motion Aftereffect in an Excitable Brain Tissue – Explaining the waterfall illusion, ISBN 978-93-5126-149-0

Barlow, H.B. (1990). A theory about the fundamental role and synaptie mechanism of visual after-effects. In C. Blakemore (Ed.). Vision: coding and efficiency. Cambridge, UK: Cambridge University Press.

Barlow, H.B., & Hill, R.M. (1963). Evidence for a physiological explanation of the waterfall illusion. Nature, 200, 1345-1347.

Bedford, F. L. (1995). Constraints on perceptual learning: Objects and Dimensions.
Cognition, 54, 253-297.

Crane, T. (1988). “The waterfall illusion.” Analysis 48.3,142-147.

Dodwell, P. C., and Humphrey, G. K. (1990). A functional theory of the McCollough effect. Psychological Review, 97, 78-89.

Guy, A. (2008). Higher Order Visual Processing in Macaque Extrastriate Cortex. Physiol Rev January 1, 88:(1) 59-89;doi:10.1152/physrev.00008.2007

Howard R.J., Ffytche D.H., Barnes J., McKeefry D., Ha Y., Woodruff P.W., Bullmore E.T., Simmons A., Williams S.C.R., David A.S., Brammer M. (1998). The functional anatomy of imagining and perceiving colour. NeuroReport, 9, 1019–1025.

Humphrey G.K.  (1998). The McCollough effect: Misperception and reality. In Walsh V., Kulikowski J., (Eds.), Perceptual constancy: Why things look as they do (pp.31–68). Cambridge, England: Cambridge University Press.

Hurvich, L. M., & Jameson, D. (1957). An opponent-process theory of color vision. Psychological Review, Vol 64(6, Pt.1), 384-404. doi: 10.1037/h0041403

Loftus, Elizabeth F. (1992). “When a lie becomes memory’s truth: Memory distortion after exposure to misinformation.” Current Directions in Psychological Science. 1.4: 121-123.

Mather, Verstraten & Anstis. (1998) .The motion aftereffect: a modern perspective. Cambridge, Mass: MIT Press.


McCamy, C. S., Marcus, H., & Davidson, J. G. (1976). A color-rendition chart.J. App. Photog. Eng2(3), 95-99.

McCollough C.  (1965). Color adaptation of edge-detectors in the human visual system.
Science, 149, 1115–1116.

McCollough, C., & Webster, M. (2011). McCollough effect. Scholarpedia, 6 (2), 8175.

Riggs, L. A., White, K. D., & Eimas, P. D. (1974). Establishment and decay of orientation-contingent aftereffects of color. Perception & Psychophysics, 15, 53-56.

Siok, W., Kay, P., Wang, W., Chan, A., Chen, L., Luke, K., & Tan, L.H. (2009). Language regions of brain are operative in color perception. PNAS 2009, 106 (20), 8140-8145.

Stromeyer, C.F, (1978) Form-color aftereffects in human vision. Handbook of sensory physiology, Vol. 8: Perception. New York: Springer-Verlag.

Svaetichin,G. (1956). Spectral response curves from single cones, Actaphysiol. scand. 39, Suppl. 134, 17-46.

Vul, E., Krizay, E., & MacLeod, D. (2008) The McCollough effect reflects permanent and transient adaptation in early visual cortex Journal of Vision, 8(12):4, 1-12.

Zeki, Semir, Watson, J. D. G., & Frackowiak, R. (1993). “Going beyond the information given: the relation of illusory visual motion to brain activity.”Proceedings of the Royal Society of London. Series B: Biological Sciences 252.1335: 215-222.