Above you see a so-called luminance grating with 4 dark-bright cycles. In the middle there is a horizontal greyish bar. Is the lightness (grey tone) constant when gazing along it? You may observe that the grey in the stripe is slightly darker and brighter along its length, in counterphase to the grating.
What to do
Move the right compensation slider until the middle stripe shows about the same brightness throughout. That’s call “cancellation” or “nulling”. When you’re satisfied, click on the left “Ø→” button, thus setting the contrast of the background grating to zero. Now (probably) you will find that the stripe is not even at all – it has a grating too. Thus the background grating has induced an illusory counterphase grating in the grey stripe (which, initially, was really a homongenous grey). The size of the effect can be read from the text field on the left of the right slider, which gives the cancellation contrast in %.
You can check for yourself more of Mark E. McCourt’s original (1982) findings:
The slider on the very right lets you change stripe width. The lightness modulation induced by the grating is strongest at fairly narrow stripe widths.
Increasing the spatial frequency (number of stripes or “cycles” of the grating) decreases the induced contrast.
Comment
I initially had thought this to be a classical case of lateral inhibition in the retina. However, there is interocular transfer – it still works if you show the stripe to one eye and the background grating to the other, thus it’s cortically based. Further quantitative analyses (see sources below) also suggest that this effect has its mechanisms in the primary visual cortex. Details remain unclear.
Sources
McCourt ME (1982) A spatial frequency dependent grating-induction effect. Vision Res 22:119–134
