Why We See Colour

Here is a bit from an interview with theoretical neurobiologist Mark Changizi in a blog post by David DiSalvo in his blog Neuronarrative:

Our color vision fundamentally relies upon the cones in our retina, and I argue in my research that color vision evolved in us primates for the purpose of sensing the emotions and states of those around us. We primates have an unusual kind of color vision – our cones sample the visible spectrum in a peculiar fashion – and I have shown that one needs that kind of peculiar color sense in order to pick up the color modulations that occur on our skin when we blush, blanch, redden with anger, and so on. Our funny primate variety of color vision turns out to be optimized for seeing the physiological modulations in the blood in the skin that underlies our primate color signals.

So, we evolved special mechanisms designed for sensing the emotions and states of others around us. That sounds a lot like the evolution of a “mind-reading” mechanism, which is why I (only half in jest) describe it that way.

In an earlier post I mentioned the puzzle about whether we all see the same world. As a colour blind person I'm aware that I'm odd man out. This bit points out that our eyes are tuned to see "emotional states" and act as a mind-reading mechanism if we are fully adept and not colour blind. We see something objective, but something that is attuned to an inner state of another person. Interesting.

Here is another bit from the interview of Mark Changizi:

[Interviewer] You mention in the book that reading and writing are relatively recent advances in human development, and yet we take for granted that we “see” and understand words, as if our brains were simply meant to see and understand them. What’s really going on that allows us to make sense of symbols on a page—and why can we do this at all?

[Mark Changizi] In talks I often show a drawing of a child reading a book titled “How to Somersault.” The “joke” is that most kids are able to read very early, often even before they can do stereotypical ape behaviors like somersaults and monkey bars. Sure, they comprehend speech much earlier, but they’re getting orders of magnitude more speech thrown at them than writing. Kids learn to read very early, and very well; and as adults we are ridiculously capable readers, and spend nearly all our day reading.

Aliens might be excused for thinking we evolved to read.

But the invention of writing is only thousands of years old. In addition, for most of us, our grandparents, great grandparents or great great grandparents didn’t read at all. Writing is much too recent for our brains to have evolved to have reading mechanisms.

How does our brain do it?

Is it because our visual system can become good at reading whatever we present to it? No. Kids would surely not be capable readers by around six if they were tasked to read bar codes or fractal patterns.

The solution is that culture made writing easy on the eye, by shaping letters to be what the eye likes. The idea that culture shapes our artifacts to be good for us is not new. What’s new here is a specific hypothesis for what writing should look like in order to be good for us.

To be easy on the eye, writing needs to “look like nature,” just what our illiterate visual systems are fantastically competent at processing. The trick of that research direction was making this “writing looks like nature” idea rigorous, and coming up with ways of testing it. I show that there are certain signature visual patterns found in nearly any natural environment with opaque objects strewn about, and that these signature patterns are found in human writing. In short, writing has evolved so that written words look like visual objects.

Interesting suggestion. Without reading Mark Changizi's book "The Visual Revolution" I can't judge the above claim.

My fundamental problem with the above argument is that humans have two major different writing styles. (1) The ideograms of Chinese, MesoAmericans, and Egyptians. (2) The phonetic alphabets that came out of the traders of the Middle East (and evolved into a number of different "alphabets" with different shapes, and even within the Western world we have an amazing array of fonts with enough differences to make automated character recognition to be very, very challenging for computer science.

Here is a review of Changizi's book from the Wall Street Journal. This review makes the book sound exciting because it claims:

Why are we humans so good at seeing in color? Why do we have eyes on the front of our heads rather than on the sides, like horses? And how is it that we find it so easy to read when written language didn’t even exist until a few thousand years ago—a virtual millisecond in evolutionary time?

Most of us, understandably, have never given much thought to questions like these. What is surprising is that most cognitive scientists haven’t either. People who study the brain generally ask how it works the way it does, not why. But Mark Changizi, a professor at Rensselaer Polytechnic Institute and the author of “The Vision Revolution,” is indeed a man who asks why, and lucky for us: His ideas about the brain and mind are fascinating, and his explanations for our habits of seeing are, for the most part, persuasive.

And from this review, there is the bit explaining the "fast to read" ability:

He claims that we learn to read so easily because the symbols in our written alphabets have evolved, over many generations, to resemble the building blocks of natural scenes— exactly what previous millennia of evolution adapted the brain to perceive quickly. A “T,” for example, appears in nature when one object overlaps another, like a stone lying on top of a stick. With statistical analysis, Mr. Changizi finds that the contour patterns most common in nature are also most common in letter shapes.

I find this claim hard to believe because of my two points about written language I gave above. It sounds like Changizi's "statistical studies" were limited to a Western alphabet, and even there the bit about a "T" being a stone lying on a stone doesn't make sense because that only describes the majuscule version "T" and not the minuscule version "t" which can be a crossbar or a crossbar with the little tail running off to the right.

If you ever look at old manuscripts where there is no spacing for words or paragraphs and no majuscule/minuscule differentiation, it is very, very hard to read for a modern reader. This makes me very leery of Changizi's argument. I would need to sit down with his book and look at his evidence before I would be convinced.