How we gave colors names and it messed with our brains (part II)

I was actually talking with some friends about just this recently.

Particularly, I speak Russian as well as English. In Russian, we actually have two separate words for blue: one for light blue and one for dark blue. They are completely different colors. My idea was that this actually changes how Russians view colors as compared to English speakers. Good to see my idea has some scientific backing :).

I personally am in a particularly odd position: I learned Russian first but still learned English at a very young age (5) and have since used it more. I'm not entirely certain how this has affected my view of blue, but I think I see it more like English speakers (e.g. not differentiating between the two blues intuitively) than Russian speakers. This probably says something about my relative comfort in the two languages.

Another interesting thing is that for the longest time I did not even realize that the difference was so fundamental. I just took it in stride. When I thought about it, it was a little weird: there is actually a different set of colors (rather than just shades of color) in the two languages. The fact that I could go from one set to the other without noticing is rather interesting as well.

I have the feeling that most HN readers have their colour perceptions affected far more by the limited spectrum of an RGB LCD than anything else.

For instance - I challenge anyone reading this to tell me what colours are not represented well on their screen. The spectrum is wide and 3 channels is really 'tolerable' rather than 'good' coverage. In reality there is a vast range of violets, reds and green-yellow-orange in the world that are represented as muddy-blurs of other colours rather than themselves on screen...

Nobody seems to make the same fuss over quality of colour and the visual spectrum on computers as say, audiophiles do about miniscule differences in what they hear. I'm not quite sure why this is.

The end of the embedded video, where it shows the experiment with the African tribe and how they see colors differently, is totally amazing.

http://www.youtube.com/watch?v=4b71rT9fU-I&feature=playe...

This is interesting because in Arabic colours are presented very differently than in English. I'm not sure about other languages but in Arabic colours are a bit more "looser" for lack of a better term. There are more shades of red for instance because Arabs were used to different shades of red in their horses and deserts. The colours are less defined for instance "the colour between red and brown" (as many horses are) or "the colour of ash" are used instead of "the colour red / grey". Colours also have properties and traits which define them (some colours are strong or thick and others are weak).

Perhaps any HN Arabic grammarians might be able to discuss this further.

Perhaps I've been warped by machine learning but it seems obvious to me that labeled color differences would be easier to distinguish. Since each time we used the word we would be training our brain to discriminate between color input patterns whose differences are not stark. Colors with no labels would not get many opportunities to be learned, but for the called out colors, on each use it would strengthen the connections between which ever set of associated neurons.

Language is not the only thing to train this difference. Hobbies and jobs should be able to as well. I suspect artists and photographers perceive subtle color gradations better than non artists; audiophiles, dancers, and musicians can pick out tempo and pitch much more readily, blind people can pick out echo patterns, engineers perceive catastrophic corner cases where normal people see an idea and a botanist sees a whole world where I just see grass and trees.

The problem with building massive hypothesis on human cognition based on findings like http://www.empiricalzeal.com/wp-content/uploads/2012/06/reac... is that there's no mention of just how many folks actually participated in the study.

Much of 'evolutionary psychology' is based on really complicated mathematical models fit to very small samples, or no sample at all, expanded upon by lengthy blocks of meandering prose.

Just-so stories are a literary genre, not a science.

"Having a word for blue seems to make the color ‘pop’ a little more in our minds. "

A lot to digest in this article. The quote above reminded me of 'categorical perception' of musical notes among musicians (without perfect pitch) and the rest of us. Professional musicians seem to 'bin' frequencies near a tone centre into notes with sharp divisions between them.

This is very interesting stuff. I’m a native English speaker but also grew up learning French and other languages; to me, colour categories seem to be pretty much an arbitrary matter of convenience. There is no inherent distinction between light blue and dark blue in English, just as there is no distinction between pure and impure code in C++. You can introduce it artificially, it’s just that the language puts them in the same category.

Also, English may have only 11 “basic” colour categories, but there are loads more colour words than that, and you can get basically as specific as you like.

I also have a suspicion that I’m a tetrachromat, because I distinguish various red-orange colours that no one else seems to. Since I’m male, though, that would only be possible if I had two X chromosomes and the right mutations on both sides of my family—possible, but unlikely.

So if this is really true, I wonder how we should name colors for maximum usefulness? For instance, which set of names will help us tell colors apart better, "red orange yellow green blue violet" (http://en.wikipedia.org/wiki/File:Color_star-en.svg) from the RYB color model, or "red yellow green cyan blue magenta" (the six labeled angles on http://www.had2know.com/images/hsv-color-model.png) from the RGB model?

I suppose, for one thing, that it depends on whether we want to describe colors we often see in life (lots of blue sky, not many purple things) or the spectrum of colors evenly for looking at art.

RadioLab had a show recently on this subject, it's really well put. a must listen episode.

This makes me that we really are but neural networks running on massive CPUs called brain. We need training data to accomplish much of anything. And the training data is the result of the analysis from previous generations, especially our parents.

The difficult one with the green squares challenges us to find the difference between 4FBB0E and 5FC004. It's a pretty substantial change in the red value (though totally invisible to me). They cover themselves in ochre, right? So they're probably more sensitive to red.

Also, they screwed up producing the video. The first time they show the green squares, the different one is in the top left. The second time, it's in the top right.

For those interested in learning more about the subject, I would recommend Through the Language Glass by Guy Deutscher. It deals heavily with the relationship between language and color as part of exploring the overall relationship between language and culture. (This particular experiment is one of the topics he discusses.)

I'm a bit confused by this article as it claims:

Koreans are familiar with the colors yeondu and chorok. An English speaker would call them both green (yeondu perhaps being a more yellowish green). But in Korean it’s not a matter of shade, they are both basic colors. There is no word for green that includes both yeondu and chorok.

Wouldn't that be chartreuse?

Naming things changes our perception of the world. This is know from centuries ago, zen master, buddist, they know this.

In the video, the Tribe need to see the difference between greens because they must know the difference between a poison leaf and a one good. Each culture evolves to name the important things to them

Wow, interesting. Hard evidence that language affects our low level perception of the world. There was always speculation, that different languages influences the way we "think", but I haven't seen such a black-and-white proof yet.

This is fascinating. By teaching our kids to use our culture's color words, we are training them in a specific classification skill, so that when they see a color, the name of its class comes to mind.

I had no trouble distinguishing the second circle either, but I'm speculating that's because I call one of those colors turquoise.

Well, I actually do not think that the tiles in the "circle of colored tiles" are green. For me, they are as much blue as green.