: Reverse-Colorblind Test :

Update, 13 March 2007: First, thank you for stopping by. You should read my blog entry about this test, which explains its background a little more, and also discusses some of the results people have communicated to me over email, since I first put this page online in April of 2001. It also explains that this test is only a proof-of-concept, and is not meant to be robustly diagnostic. In other words, YMMV. Enjoy! And, please feel free to email me with your own experience of the test. End Update

Color blind is the name given to people who do not have a 'normal' range of chromatic (color) contrast sensitivity. There are several different kinds of color blindness, and not all are due to genetic variation. The subjective experience of color vision is partially created by the retina measuring differences in the wavelengths of incident light. That is, 'normal' color vision is enabled by having essentially three relatively independent 'channels', each sensitive to a different distribution of wavelengths. When, say, the red channel is stimulated relatively more than another, then we subjectively experience the color red, with the details being partially controlled by the relative intensity of the stimulation. A fourth channel - your rods - is not used in this way, and is thus more sensitive to variations in lightness versus darkness (luminosity). (For a much more thorough explanation of how color vision operates, try here.)

Tests for color blindness (such as the popular Ishihara test (a fancy flash version is available here) - the one where bubbles of diferent sizes and colors are used to conceal a number from color blind people) are typically designed around 'confusion lines', which are basically colors that don't differentially stimulate different color channels, and thus appear to be basically identical.

In April of 2001, while discussing with a close friend of mine Nick Yee, who is red-green colorblind, the differences in our subjective visual experiences, we realized that color vision is not an unequivocal gain. That is, in some situations, he could perceive variations in luminosity that I could not. This difference got us to thinking whether we could design a "reverse" color blindness test - one that he could pass because he is color blind, and one that I would fail because I am not. The hypothesis we worked from is that, under certain circumstances for a color-sighted person, the information the brain receives about color contrasts can overwhelm (or otherwise discount) the information it receives from luminosity contrast. That is, we could distract color-sighted people with bright colors from noticing "hidden" information in subtle but systematic variations in luminosity. This test is the result of that conversation.

Red Spectrum Gradient

This image contains both a simple red gradient, and a pattern.

colorblind_gradient.jpg

If you can see the pattern overlayed on the gradient, this suggests that you may be red-green color blind. If you can't see it, you could try looking to either side of the image. Outside of the fovea - i.e., in your peripheral vision - there is a greater relative concentration of rods than cones, and this may increase your ability to distinguish variations in luminosity. (For instance, ever notice that when looking up at the stars, there are some you can't see when you look dead-on, but which you can if you look slightly away? Your rods are so sensitive that they can basically detect single photons, but obviously color contrast requires more light than that!)

Red Speckles

In this example, I created an image more reminiscent of the classic Ishihara plates. Again, there is a pattern (different from the previous one) overlayed on the background speckles. If the pattern is clear to you, then you may be red-green colorblind. That is, if you 'fail' this test, then you probably have the full range of color sensitivity that is attributed to 'normal' people.

colorblind_speckle.jpg

'Answers' for these images are on a separate page. To make these, I tweaked the color contrast to make the overlayed patterns significantly more obvious. If you 'failed' the test, once you've seen the answers, come back and look again. It may be easier to pick out the subtle variations in luminosity now that you know what to look for. And, I'm always interested to hear about people's subjective experiecnes with this test, so please feel free to email me directly about them! (Over the years that this page has been online, I've received about 100 personal testimonials, and while there are certainly some people who erroneously 'fail' or 'pass' the test, it seems to be otherwise pretty accurate.)