Contrast

Notes

Reorganize around: Brightness contrast, Color contrast, Grid illusions, and Assimilation, then Exploration. Start by referring to the example above (can then remove the gradient/bar version of the basic illusion). For grids, show Hermann and Scintillating. Then give the usual explanation with . Finally (or maybe in Exploration), link again to the generic grid and prompt changing twist, gaps, etc. Ask about the effects and whether any of them challenge the usual explanation of grid illusions. OR should Grids be done along with Enhancement? Could do them as suggested above but refer them as a class of enhancement illusion.

Where to introduce the method of adjustment and point of subjective equality? Maybe a full section on them. Pose the question of how to measure these effects. Describe the methods, then link to adjustment versions of each illusion. Could also cite some papers on methods for grid illusions (but not do them). Prompt students to consider multiple measures and why to repeat the measurement (here or in instructor notes?).

Contrast Enhancement

Before continuing, try and , following the instructions in their captions. This is the simplest type of brightness contrast illusion, in which the visual system enhances the contrast between an object and its background. One mechanism of contrast enhancement is lateral inhibition, which increases the contrast at the edge between two areas that differ in brightness. This is clearly illustrated by .

Lateral inhibition helps us see edges, but can also make us see things that are not there. is a classic example. Does it also work with colors? What if the grid is inverted (white squares and black lines)? is a striking variant of the Hermann grid, discovered in 1994 by Elke Lingelbach (Schrauf et al., 1995). This illusion has a number of interesting properties. Where do the dots go when you look directly at them?

Of course, artists have also used contrast illusions. is inspired by the 1966 painting “Arcturus II” by Victor Vasarely. The illusory star is due to brightness contrast. See for an explanation.

Color Contrast

The examples above showed that brightness contrast works when adjacent areas have different intensities of the same color, but what about contrast between different colors? Try . Once again, the visual system enhanced the contrast between object and background, but this time based on color rather than brightness. is another illustration of color contrast, based on Brou et al. (1986). In this case, identical gray tiles appear tinted.

Assimilation

The above contrast effects have one thing in common, whether gray or colored. In each case, an object is made to look more different from its surroundings than it actually is. Thus a gray area looks darker when it is surrounded by white than when it is surrounded by black, and a tan cross looks green when surrounded by red but red when surrounded by green. There are a number of counterexamples in which an object comes to look more like its surroundings. Assimilation illusions are discussed at length by Purves and Lotto (2003), while Corney and Lotto (2007) offer a novel type of explanation. For more of this perspective, see other works by Purves and Lotto in the References.

There are several illusions in which the brightness of an area seems to match its surroundings more closely than it actually does. Some are subtle, and you may not experience all of the effects. Try , (the strongest one), , and , all named after their discoverers.

Further Exploration

You can modify an illusion to test its strength under different conditions. For example:

Questions

  1. How do the assimilation illusions challenge the usual explanations for contrast illusions? Do these counterexamples necessarily mean that those explanations are wrong?
  2. Why do the illusory dots in the various grid illusions () not appear at the point that you are looking directly at?
  3. In addition to the illusions they are meant to illustrate, many of these examples also show . Which ones do and which do not?
  4. Why does the color contrast illusion vanish when the tiles in tiles are mixed randomly instead of sorted by color?

References