What to see
Above you can see a coloured background grid and 3 disks, which are (partially, depending on slider position) covered by parts of the background rid. The question is: Which colour do the disks have?
What to do
Move the slider fully to the left (that automatically deactivates AutoRun). Looking “through the grid”, to me the left disk now has a strong cyan (aqua, blue-green, turquoise) tint, the middle one magenta, the right one yellow. Actually, all three are white!
You already saw that you can slide the covering grid. You can also leave the disks covered, but change the width of the Size stepper. At size ≈ 20, the disks are for me nearly white.
This phenomenon, first described by van Bezold, is aptly known as “assimilation”, since the disk colours become similar to the (additive mixture) of the grid lines covering it: on the right, we have red and green grid lines, which mix to yellow; etc.
The opposite of colour assimilation, or chromatic assimilation, is colour contrast. While the latter can be somewhat understood on the basis of lateral inhibition, I don't understand the neural mechanism of assimilation. As Liu & Schor put it »Color contrast occurs when one color is surrounded by a different color, the difference in color appears exaggerated. Color assimilation occurs when two color patches are perceptually grouped, and their apparent color difference is reduced. King (1988) hypothesized that contrast occurs when patches are perceived as separate “wholes” while assimilation resulted when the two patches are perceived as one integrated “whole.” «
There are close similarities to White's Illusion, Neon Colour Spreading, and Munker. There are many ways to depict colour assimilation, the particular arrangement above was inspired by Gianni Sarcone.
van Bezold WM (1876) An introduction to color. New York: Wiley.
King, DL (1988) Assimilation is due to one perceived whole and contrast is due to two perceived wholes. New Ideas in Psychology 6(3) 277–288
Liu & Schor (2005) Effects of partial occlusion on perceived slant difference. JOV 5(4)