This is really a follow-up from my post about the history of seeing (or not seeing) colors like blue. It’s sort of taken on a life of its own, so that what was going to be a nice little post about the primary colors of light vs. pigment has turned into a two-parter.
Today we’re going to stick to colors in light. The reason to start there is that we only need to talk about colors because we see them, and we only see them because we have retinal cells that are sensitive to different light wavelengths. (As in my post on radio waves, let’s forget for a while that light’s also a particle. It’s just easier this way.)
If you’ve heard about primary colors of light, you probably know there are three: red, green, and blue. (“Primary” means that we can see light that appears to be any other color, just through a mixture of the big three.) This simple statement is the tip of an iceberg of possibly misleading facts. By itself, it sort of makes it sound like there’s something inherently primary about the RGB (red-green-blue, not to be confused with the notorious RBG) color scheme. What’s special about those three colors of light?
Nothing, really. What’s special is our eyes—most humans have three types of cones or color receptors in our eyes, sensitive to red, green, and blue light. Some animals share the same setup, but there’s lots of variation. (For example, dogs only have two types of receptors, which beats being totally colorblind, while butterflies can have up to fifteen types! The mantis shrimp gets a lot of attention for having twelve types of cones, but research suggests they don’t provide the mind-blowing spectrum we expected.)
Actually, even that view of color is misleading. We talk about having cone cells that respond to red, green, and blue wavelengths, but they’re better labeled as long, medium, and short wavelengths (or L, M, and S; red is longer, blue is shorter).
But wait, there’s even more! These labels, whether RGB or LMS, tend to give us the mental image of receptors that respond to a very specific color. We’ve trained ourselves to identify most colors easily by a general name, so it’s easy to think of a receptor that only handles green light and ignores everything else. In reality, though, each receptor has a zone of sensitivity that includes a whole range of wavelengths, with a peak that tails off on either end.
You’ll notice from that drawing that not only does each receptor type have a wide range, those ranges overlap somewhat. Red and green (or L and M) are especially noticeable. This is really confusing if all you have to draw on is an elementary-school memory of a color wheel with primary red, yellow, and blue, and secondary orange, green, and purple. Isn’t red across from green? How can they overlap so much? Why are red and green closer on the graph, while green and blue are neighbors on the color wheel?
(We’ll talk more later about how the traditional “rainbow” color wheel can be improved to help us actually understand light and pigment. For now, just know that it’s less useful than most people think.)
The truth is that the long-wavelength, “red” cones actually peak somewhere in yellow. We consider them red because they’re closest to that end of the spectrum and are the only cones really activated by red wavelengths.
Of course, we haven’t even talked about how these peaks vary from person to person…what’s that? Oh yes, I agree. Everything most of us know about primary colors is a lie.
…All right, a little dramatic, but obviously it’s way more complicated than the simple red-green-blue.
When we look beyond the red-green-blue model, the plot thickens even more. There’s another color vision theory, which involves comparing four colors, basically laid out on two axes. This is called the opponent process because it’s defined by the opposition between colors—in this case, between red and green on one axis, and yellow and blue on the other axis.
How did this theory come about? Well, in the nineteenth century, German physiologist Ewald Hering made an observation about our perception of color shades. As a quick illustration, try to imagine a color that you could describe as a “greenish-red” or a “bluish-yellow.” Maybe you can, but for the vast majority of us it doesn’t really compute. This suggested to Hering and others who continued his research that our brains treat the red-green and blue-yellow pairs as opposites.
Again, here’s the color wheel most of us learned, which only reflects one of these pairs:
Clearly, the wheel isn’t set up to make sense of the opponent process model. Instead, the model organizes colors along two oppositional axes. In the same way a number can’t be both positive and negative, a color can’t look both red and green, or both yellow and blue. (Octarine appears to be an exception. It would be.)
So which wins out in reality, the opponent model or the RGB model?
Both! Eventually, scientists realized there’s no conflict between the two. Our eyes are sensitive to red, green, and blue, but that input is processed by our brains according to the opponent process. Let’s see how it works!
Essentially, our visual processing is a more updated version of, say, a dog’s vision, which only has yellow- and blue-sensitive cones. Instead of just having yellow, we have red and green, which our brains add to give us a measure of yellowness that we contrast with blue. At the same time, subtracting the relative red and green responses gives us a whole other dimension of color along the red-green axis. Then we have the most basic level of color sensing, light/dark, which comes from the added responses of all three receptor types.
The result is the ability to distinguish millions of colors, based on wavelength and brightness. It’s pretty amazing what three cone types can do…just imagine having a dozen! It’s almost enough to make you jealous of butterflies, though I can pretty much guarantee they don’t understand the structures of their own eyes. You can’t have it all, I guess.
Next time, we’ll tackle primary colors in pigment. Yes, they’re different from light, and if you don’t know why, you will!
In the meantime, follow the links in the blog to learn more about our trichromatic vision, and check out the only way we can see those impossible red-green, blue-yellow colors! (It’s kind of cheating, but still.)