Mixing Memory

Here is an illusion that was discovered relatively recently. Take a look at this (from here):

i-ecf78751f49ccbf3fff9e1db25e196bd-watercolorillusiona.bmp

You should see two figures with a purple outter border and an orange inner border. What color is the interior of the figure? It probably looks like it’s orange, though a lighter shade of orange than the inner border. It’s actually white, though. Don’t believe me? Take a look at these two figures:

i-aab6966ef45b937d6f96fdb172d1fa17-watercolorillusionb.bmp

These are the same two figures as the ones above, except that I’ve removed the orange inner border from the figure on the left. Now you can clearly see that the interior is white.

This is called the “watercolor illusion,” or “watercolor effect,” and was first described by Baingio Pinna in 19871. Here is a description of the illusion from Devinck et al. (p. 625)2:

The watercolor effect (WCE) is a phenomenon of long-range color assimilation occurring when a dark chromatic contour delineating a figure is flanked on the inside by a brighter chromatic contour; the brighter color spreads into the entire enclosed area.

Here’s another demonstration. Notice that when the darker border is on the inside of the lighter border, the effect goes away (the figure on the right — from Kim & Blake, 20053):

i-21bbbdfac584a18045e896c75c05180f-watercolorillusion2.bmp

Researchers have only recently begun to study it systematically4, and the mechanisms that underlie it are not yet understood, though it probably has something to do with the center-surround structure of receptive fields on the retina5.

Here’s an interesting twist on the watercolor illusion. For individuals who experience a particular type of synesthesia, some letters produce a synesthetic effect in which colors are seen along with the letters. These synesthetes are called “projectors”6. Kim and Blake recently conducted a study designed to determine whether projectors saw the watercolor illusion when just presented with “achromatic” letters7. For example, one of the projectors who participated sees the letter R in purple and the letter B in orange. So Kim and Blake presented her figures composed of R’s and B’s that should look to her like the figures above.

i-6d4cf21b80e1b193af3e2b01098019e9-watercolorillusionsynesthetic.bmp

Only having seen the figures with letters, and not the one with actual colors, she reported seeing the watercolor illusion for the figure on the left (R’s outside, B’s inside), but not the one on the right (B’s outside, R’s inside). What this says about synesthesia or the watercolor illusion, I don’t know, but it’s pretty damn cool anyway.


1Pinna, B. (1987). Un effeto di colorazione. In V. Majer & M. Santinello (Eds.), Il laboratorio e la citta`. XXI Congresso degli Psicologi Italiani (p. 158). Edizioni SIPs, Milano: Societa` Italiana di Psicologia.
2Devinck, F., Hardy, J.L., Delahunt, P.B., Spillman, L., & Werner, J.S. (2006). Illusory spreading of watercolor. Journal of Vision, 6, 625-633.
3Kim, C.Y. & Blake, R. (2005) Watercolour illusion induced by synaesthetic colours. Perception, 34, 1501-1507.
4Pinna, B., Brelstaff, G., & Spillmann, L. (2001). Surface color from boundaries: A new Fwatercolor_ illusion. Vision Research, 41, 2669-2676.
5Devinck, et al. (2006).
6Dixon, M.J., Smilek , D., & Merikle, P.M. (2004). Not all synaesthetes are created equal: Distinguishing between projector and associator synaesthetes. Cognitive, Affective, and Behavioral Neuroscience, 4, 335-343.
7Kim & Blake (2005).

Comments

  1. #1 Blar
    June 24, 2006

    “Only having seen the figures with colors, and not the one with letters”

    These should be reversed, no?

  2. #2 chris
    June 24, 2006

    Yes, yes it should. I should probably hire you as an editor. ;)

  3. #3 Blar
    June 24, 2006

    But why? when I’m already doing it for free.