Above you see two checkerboards; the left has a high contrast, the right a low contrast. In the center of each there is a smaller 4×4 checkerboard patch within a green box. Compare their contrasts – initially, on “auto run”, their contrasts are going up and down in counterbalance.
Use the middle slider (or use the cursor left/right keys) to adjust the contrast of the two small checkerboards until they seem similar. When you are satisfied, uncheck “Draw surround” (or press the space bar). Now: without the surrounding checkerboards, the contrasts of the small patches no longer seem identical.
A case of contrast adaptation localised in space: Because the left checkerboard has a high contrast, the local contrast gain is reduced to bring the so-called contrast transfer function (CTF)¹ into its optimal range (this is usually a good thing). On the right side the inverse happens. So, although the center patches can be physically identical, they seem different because they are viewed with different CTFs. (Also known as “contrast contrast”.)
¹The “contrast transfer function” maps local contrast on the retina to neural activity in the brain. It has a sigmoidal shape, first flat near zero, then rising, then saturating. The rising part of the CTF can shift horizontally through adaptation, or here through average neighboring contrast. See also next page.
Chubb C, Sperling G, Solomon JA (1989) Texture interactions determine perceived contrast. Proc Natl Acad Sci USA 86:9631–9635 [PDF]
Solomon JA, Sperling G, Chubb C (1993) The lateral inhibition of perceived contrast is indifferent to on-center/off-center segregation, but specific to orientation. Vision Res 33:2671–2683
Dakin (2005) Weak suppression of visual context in chronic schizophrenia. Current Biol 15(20):R822 [PDF]