Why Red, Blue, Yellow and... Green?

[willpell]

I think the thing of subtle color variations being a woman thing is just socialization. I can't be sure since I've never actually competed with a woman to see who can best disambiguate paint swatches or anything (that might make a neat concept for a game), but I easily identify at least hundreds of colors as being distinct, so I don't think the distinction can be biological.

[Erloas]

Originally Posted by Surfing HalfOrc

I can see the "cultural" color differences when it comes to my wife. She is Korean, and when the Stop Light changes, she always says the light is Blue.

Do they use a different set of colors for their stop lights in Korea? (South I would assume) Or... depending how long she has been speaking whatever your natural language is (since it could very well not be English) is she just getting the word usage wrong?
I'm not aware of any colorblindness that is common between red and blue distinction.

Originally Posted by Willpell

I think the thing of subtle color variations being a woman thing is just socialization. I can't be sure since I've never actually competed with a woman to see who can best disambiguate paint swatches or anything (that might make a neat concept for a game), but I easily identify at least hundreds of colors as being distinct, so I don't think the distinction can be biological.

There is some socialization in that a lot of people will see "white" and say it is "white" when they could clearly put 10 swatches together and see that they are all slightly different "whites". And when walking into a room they aren't going to care that this particular "white" is slightly more tan or slightly more blue or slightly more green then some other white because it is just white.

However there is also some biological aspects to color distinction. For instance color blindness much more common among men then women. A quick check on the internet shows about 20% of men have some form of colorblindness but only about 0.5% of women do. *Though I could be reading the numbers wrong as it is broken down kind of weird, either way it is clear that it is much more prevalent for men. There are also many different forms of colorblindness and many degrees of it.

And looking at the color blind tests on Wikipedia it seems I've got issues with the protanope type... I almost can't see the number, though it is very faintly visible.

[Serpentine]

In some places they do mix a bit of blue into the green traffic light to help colourblind people...
According to wikipedia, there's only 2 possible cases of human tetrachromats. I found a site some time ago that supposedly had a tetrachromacy test (similar to a colour blindness test), including an "if you can see the number here, please contact..." note on it. Unfortunately it seems the original site is gone, but I think this might be the test ('course, it could also be a fake):

Spoiler

My ex is colourblind. He can't tell the difference between green and orange, and between pink and blue. He insists he's "not that colourblind", that he "just can't tell between colours of a really close shade". I found it out when I was eating my dinner of carrots, peas, broccoli and brussels sprouts (or something like that), and he said "there's far too much orange on that plate". He thought the phrase "eat your greens" was meant ironically.

[Winter_Wolf]

I apparently have a minor blue-purple color deficiency. Minor, because I can still tell the difference most of the time, but I had a car which I am to this day convinced was purple, but everyone else says was dark blue.

That said I get a definite sense of a large 7 or small 26 in both the yellow and green circles, but the red just gives me eye strain. I'm probably just seeing dot patterns instead of actual numbers, though. I've always disliked color blindness tests and magic eye pictures.

[GrlumpTheElder]

That test is most likely a fake. The main reason I say that is because is because the colours on your computer screen are only being presented in three channels (RGB). Tetrachromatic birds usually extends out into the ultraviolet range of the EMS. Computer monitors are unable to render Electromagnetic radiation outside of the range of Visible light.

[Flickerdart]

Originally Posted by GrlumpTheElder

That test is most likely a fake. The main reason I say that is because is because the colours on your computer screen are only being presented in three channels (RGB). Tetrachromatic birds usually extends out into the ultraviolet range of the EMS. Computer monitors are unable to render Electromagnetic radiation outside of the range of Visible light.

Tetrachromatic birds, sure, but human tetrachromats have an extra orange cone rather than a UV cone.

[Yora]

Originally Posted by Winter_Wolf

I apparently have a minor blue-purple color deficiency. Minor, because I can still tell the difference most of the time, but I had a car which I am to this day convinced was purple, but everyone else says was dark blue.

There is no purple!

Or at least, there are no purple wave lengths. What looks like purple is seeing both red and blue light at the same time. Telling the difference between blue and purple might actually be a problem with red. But in that case, you would be seeing red where there is none...

Or maybe, it's just a matter of your personal definition which shades are blue and which shades are purple, and you just happen to be an outlier among your group of friends.
I constantly get into short debated with my parents every time we see something turquoise about it being a bluish turquoise or a greenish turquoise.

What color is this anyway? This is almost perfecly in balance.

Originally Posted by Serpentine

In some places they do mix a bit of blue into the green traffic light to help colourblind people...
According to wikipedia, there's only 2 possible cases of human tetrachromats. I found a site some time ago that supposedly had a tetrachromacy test (similar to a colour blindness test), including an "if you can see the number here, please contact..." note on it. Unfortunately it seems the original site is gone, but I think this might be the test ('course, it could also be a fake):

Spoiler

I'd tried to put this through some filters and contrast changes in photoshop, but I never got it to detect any visible or nummerical differences. Though I don't really know how these things work and was just moving sliders at random.

This is the invisible pink unicorn, which as data is both defined as colored pink, but also as 100% transparent.

><

[Surfing HalfOrc]

Originally Posted by Erloas

Do they use a different set of colors for their stop lights in Korea? (South I would assume) Or... depending how long she has been speaking whatever your natural language is (since it could very well not be English) is she just getting the word usage wrong?
I'm not aware of any colorblindness that is common between red and blue distinction.

No, Korea uses (to me) the exact same shade of red/yellow/green as we do in the U.S. My wife emigrated to the U.S. when she was 11, so she speaks fluent, colloquial English. And she definitely knows the difference between a blue and green poker chip!
Blue=10
Green=25

She has lived in the U.S. (mostly Hawaii) for 30+ years, and we have been married for 21 years. She just says "The light is blue" if I'm not paying attention when stopped.

[Yora]

Japanese introduced a clearly seperate term for green centuries ago, but some things are are still called "aoi", even though in the color spectrum, they are clearly in the region of "midori". Like apples and traffic lights.

[Siosilvar]

Originally Posted by Erloas

Do they use a different set of colors for their stop lights in Korea? (South I would assume) Or... depending how long she has been speaking whatever your natural language is (since it could very well not be English) is she just getting the word usage wrong?
I'm not aware of any colorblindness that is common between red and blue distinction.

IIRC, the word for "blue" and "green" in Korean is the same word, so she probably doesn't notice the distinction unless she makes a conscious effort to.

I watched a video where they tested this with an African tribe. They displayed a circle of 12 colored squares on a screen, with one different from the rest. In the first case, there was a very obviously (to me) blue square in a ring of green. The second case had a slightly yellow green in a ring of lime green. The tribe members, with a word to distinguish between yellow-green and lime green but just one for lime green and blue, spent much longer on the first test and very quickly pointed out the second one, which was the opposite the results of English speakers who took the test.

[Heliomance]

The video was linked earlier :P

[Winter_Wolf]

Originally Posted by Yora

There is no purple!

Or at least, there are no purple wave lengths. What looks like purple is seeing both red and blue light at the same time. Telling the difference between blue and purple might actually be a problem with red. But in that case, you would be seeing red where there is none...

Or it could be that the car was coated in pearl-coat paint. It did look strikingly different in direct sunlight and overcast days.

I also have a weird thing where one eye sees reds more, and the other eye sees blue more. No one else seems to know what I'm talking about, though. I used to test this repeatedly during fishing season while waiting pick the nets. Yeah, mostly drift-netting is NOT a high adrenaline pursuit, just a lot of hurry up and wait. Anyway! The radar cover was blue, and I'd sit up on the flying bridge and close one eye, then switch to the other, and it was consistent.

I was later diagnosed with macular degeneration in one eye, so it's also entirely possible that I've been looking at the world through a layer of blood (in one eye).

Also, purple is as real as spoons.

[Ravens_cry]

I have a theory we are so sensitive to different greens because we are omnivores.
A slight shade difference in the colour of leaves could mean the difference between dandy dinner and doomful death.

[Elemental]

Originally Posted by Ravens_cry

I have a theory we are so sensitive to different greens because we are omnivores.
A slight shade difference in the colour of leaves could mean the difference between dandy dinner and doomful death.

That would explain why some cultures have more words for various greens and almost none for blue.
After all... Blue isn't a very commonly found colour in nature...
Except for the sky, but when you're looking for stuff to eat, unless you're a good shot with a bow, there's not much point looking up.

[Ravens_cry]

Originally Posted by Elemental

That would explain why some cultures have more words for various greens and almost none for blue.
After all... Blue isn't a very commonly found colour in nature...
Except for the sky, but when you're looking for stuff to eat, unless you're a good shot with a bow, there's not much point looking up.

And bows are a fairly recent invention, evolutionarily speaking.

[Musashi]

I feel this is relevant to the discussion.
(WARNING: swears and general immaturity behind the link.)

[Erloas]

Originally Posted by Ravens_cry

I have a theory we are so sensitive to different greens because we are omnivores.
A slight shade difference in the colour of leaves could mean the difference between dandy dinner and doomful death.

I don't think there are any correlation between color and eatability in plants. As it is there is a huge range of greens in plants that we do eat. I think the majority of plant based problems is also the berry rather then the greens and most berries are not green.

One big reason for green being easier to distinguish is the fact that it simply takes up more space on the visual spectrum.

Color	Frequency	Wavelength	range
violet	668–789 THz	380–450 nm	70
blue	606–668 THz	450–495 nm	45
green	526–606 THz	495–570 nm	75
yellow	508–526 THz	570–590 nm	20
orange	484–508 THz	590–620 nm	30
red	400–484 THz	620–750 nm	130

So green is the second longest section, but both red and violet are at the ends of the visible spectrum so some extent of that range it outside of what people can see.

[Ravens_cry]

Originally Posted by Erloas

I don't think there are any correlation between color and eatability in plants. As it is there is a huge range of greens in plants that we do eat. I think the majority of plant based problems is also the berry rather then the greens and most berries are not green.

No, but it can help you identify one plant from another, especially in the non-fruit seasons.

[Elemental]

Originally Posted by Ravens_cry

No, but it can help you identify one plant from another, especially in the non-fruit seasons.

Indeed. And very few plants are constantly in flower or in fruit.
If they were, then identifying them would be easy.

Otherwise, you need to go by leaf shape, which is often very similar from plant to plant.
And echoing what Ravens_Cry said earlier, a slight variation in shade means the difference between salad and food poisoning. Foxglove, for instance, can be easily mistaken for other plants.

[Erloas]

Originally Posted by Elemental

Otherwise, you need to go by leaf shape, which is often very similar from plant to plant.
And echoing what Ravens_Cry said earlier, a slight variation in shade means the difference between salad and food poisoning. Foxglove, for instance, can be easily mistaken for other plants.

But there are a lot of plants that have different coloring depending on the time of year or how much water or sun they have. You can get one plant that has different coloration on different leaves at the same time. Some plants are very consistently colored, others are not.

And if we take a plant like a maple tree, they have very distinctive leaf shapes but they come in all sorts of different colors, with greens, reds, and silvers being very common.

[Elemental]

Originally Posted by Erloas

But there are a lot of plants that have different coloring depending on the time of year or how much water or sun they have. You can get one plant that has different coloration on different leaves at the same time. Some plants are very consistently colored, others are not.

And if we take a plant like a maple tree, they have very distinctive leaf shapes but they come in all sorts of different colors, with greens, reds, and silvers being very common.

The main difference being that humans don't eat maple leaves, but they eat the leaves of numerous smaller plants.
But, there are a lot of examples of plants that are highly toxic that bear striking similarities to other plants that are rather good to eat, and leaf shape can't always be relied upon.
Foxglove, for instance, looks rather similar to comfrey and sage and poison hemlock resembles wild carrot and wild parsnip.

[Devils_Advocate]

There's a Wikipedia article on the Opponent Process Color Theory, which describes what's meant by "opponent colors" in that context.

Language can definitely mess with your perception of color. I wasn't even aware until recently that brown is dark orange... but I imagine that I would have been if English didn't have "brown" as a basic color term!

As another example of this: Do a Google image search for "traffic light". See how the "go" light is typically a lot bluer in the photos than in the drawings? That's because those drawings were made by people who were taught that traffic lights are "green"! O_O The "go" color on most traffic lights is actually more of a cyan that it seems pretty reasonable to call "blue" when I consider it closely. But I never did consider it closely until it was brought to my attention... and thus the way that that color is commonly refered to by English speakers greatly influenced my perception of it.

Originally Posted by Heliomance

The case was clearly made that they can tell the greens apart far better than the green and the blue, but I'm unconvinced that that's caused by the language. I saw no evidence for that at all. As I'm pretty certain spectral analysis would show the two greens are far more similar than the blue and the green, there's something else going on there.

... That doesn't make any sense. If someone can more readily distinguish two colors that she has different terms for from each other than she can distinguish two colors that she uses the same term for from each other, despite the differently-termed colors being more similar to each other than the same-termed colors, surely that's evidence in support of the theory that having different terms for colors makes them easier to distinguish! It's the exact opposite of an indication that "there's something else going on there".

Originally Posted by Flickerdart

Magenta isn't even a spectral colour but a fraud

What about being non-spectral makes magenta fraudulent?

Originally Posted by Heliomance

What makes CMY any more valid than RYB?

The former can be combined to produce more colors than the latter. You can make red by mixing magenta and yellow; and you can make blue by mixing cyan and magenta; but you can't get cyan nor magenta by mixing red, yellow, and/or blue. So RYB allows you to produce only a subset of the colors that you can produce with CMY. I think.

Originally Posted by willpell

On the topic of color geekery, I've always kind of rolled my eyes at the idea that the rainbow includes Indigo

"A careful reading of Newton's work indicates that the color he called indigo, we would normally call blue; his blue is then what we would name blue-green or cyan."
- Gary Waldman

The color termed "blue" in the RGB model looks a bit violet to me. And judging by the xkcd color survey results, people more commonly think of blue as being less violet than that. I've taken to thinking of RGB blue as "indigo blue"; it's helpful to have a more specific term for it, since "blue" is really a much broader basic color term.

I've semi-recently come to the conclusion that I consider Gold to be a distinct color between Orange and Yellow. The rainbow doesn't agree with me, but I'm not one for taking my cue from nature anyway. So for my purposes, there are eight primary and secondary colors.

Which colors are they?

Originally Posted by Yora

What color is this anyway? This is almost perfecly in balance.

"Teal" is the word for dark cyan.

Originally Posted by Erloas

Do they use a different set of colors for their stop lights in Korea? (South I would assume) Or... depending how long she has been speaking whatever your natural language is (since it could very well not be English) is she just getting the word usage wrong?
I'm not aware of any colorblindness that is common between red and blue distinction.

I'm pretty sure that Surfing HalfOrc meant "when the light changes from 'stop' to 'go'", not "when the light changes to 'stop'".

Originally Posted by Siosilvar

IIRC, the word for "blue" and "green" in Korean is the same word, so she probably doesn't notice the distinction unless she makes a conscious effort to.

What you mean, I think, is that Korean has a basic color term that covers colors that English-speakers call "blue" and colors that English-speakers call "green". Describing this state of affairs by saying that Korean uses the same word for green and blue is misleading at best, like saying that the English word for azure is "blue". It's not English lacks the narrower term, it's that the narrower term isn't commonly used and the distinction between different hues of "blue" isn't commonly drawn -- despite the fact that they're readily distinguishable from each other even by someone whose language doesn't include different basic color terms for them!

[Heliomance]

Originally Posted by Devils_Advocate

... That doesn't make any sense. If someone can more readily distinguish two colors that she has different terms for from each other than she can distinguish two colors that she uses the same term for from each other, despite the differently-termed colors being more similar to each other than the same-termed colors, surely that's evidence in support of the theory that having different terms for colors makes them easier to distinguish! It's the exact opposite of an indication that "there's something else going on there".

Correlation does not imply causation. And even when causation is there, we must establish which is cause and which is effect. Personally, I think it far more likely that they have different words for the shades of green because they appear as different colours to them, rather than the other way around.

[Rising Phoenix]

Originally Posted by Asta Kask

I wonder how cichlids would see this. They have five different genes for color-discrimination.

Mantis shrimp have 14+ colour receptors in their eyes... their view of the world must be... spectacular...

[ThiagoMartell]

Very interesting thread.

[Iruka]

Originally Posted by Heliomance

Correlation does not imply causation. And even when causation is there, we must establish which is cause and which is effect. Personally, I think it far more likely that they have different words for the shades of green because they appear as different colours to them, rather than the other way around.

Clearly we should kidnap their children before they start to speak, teach them English and have them identify colours for us. For science!

I find it more likely that their colour perception is influenced by language then them having a random genetic mutation that changes their perception of green. But that's mostly a gut feeling on my side. I'm sure I've read about other studies about that topic, but I can't remember where.

Here's a study that claims the number of color-sensitive cones doesn't affect how you perceive colours.

There's even a long wikipedia article about this debate. Fascinating.

[Herpestidae]

This thread is making me question my literal worldview.

[Heliomance]

I'm pretty sure there have been studies done on whether language influences culture (in the popular "they have no word for peace" fashion), and found little evidence to support the idea. I can't remember what to search for to find the results again though, I'm afraid.

[Surfing HalfOrc]

Originally Posted by ThiagoMartell

Very interesting thread.

It is! I was actually only wondering why game makers chose those four colors most commonly for pawns, especially for "children's" games. Especially since the first games I remember tended to have either red and black pieces (checkers) or white and black pieces (chess).

There is a certain "cheerfulness" to RBYG, which seemed absent from R/B or W/B game sets.

Sorry! had white pawns (I think? It's been over 40 years since I last played) in addition to blue, green and red, while Monopoly had silver pawns, and multi-colored money. (I saw an image showing how modern U.S. currency matches the Monopoly color scheme).

[Devils_Advocate]

Originally Posted by Devils_Advocate

The "go" color on most traffic lights is actually more of a cyan that it seems pretty reasonable to call "blue" when I consider it closely.

For that matter, the "yellow" light is really orange or at least orange-yellow in the vast majority of cases. But it's usually drawn that way too, so I guess that people are more aware of that. (Pure yellow is pretty distinct, at least to me.)

you can make blue by mixing cyan and yellow

For the record, I never said this. Why would I have said this? Especially in the same post in which I mention that "indigo blue" (as I termed it) looks rather violet to me, as if it weren't already plain enough that green is what you get by mixing cyan and yellow. So obviously I never said it.

Originally Posted by Heliomance

Correlation does not imply causation. And even when causation is there, we must establish which is cause and which is effect.

Hmm. OK, true. But that video doesn't appear to support any particular causal explanation over any other. Which I guess means we were both wrong?

Personally, I think it far more likely that they have different words for the shades of green because they appear as different colours to them, rather than the other way around.

Those possibilities are hardly mutually exclusive!

But if I understand you right, you're saying that you already had your mind largely made up on this issue, and that you were interpreting what you saw under the unspoken assumption that the influence of language on color perception is minor.

If so, that's why I had a hard time understanding you. It hadn't really occurred to me that someone might assume that. ... Why would you assume that?

For that matter, why are you "pretty certain spectral analysis would show the two greens are far more similar than the blue and the green"? Obviously, it doesn't make sense to say that just because the two greens look more similar to each other to you; you have no reason to assume that your own subjective perception of color more closely matches objective reality. Furthermore, I'm not sure that there's an unbiased measure of "similarity" in this context.

My understanding is that humans start out with perceptual superpowers in infancy, able to distinguish far more colors, faces, sounds, etc. than adults. But then synaptic pruning or something happens to bias us towards making only "relevant" distinctions -- to the point that we lose much of our ability to make other distinctions!

Originally Posted by Iruka

Clearly we should kidnap their children before they start to speak, teach them English and have them identify colours for us. For science!

But in order to control for everything but language, you'd have to keep every other aspect of their environment the same, and language may be interrelated with other aspects of their environment in ways that make this impossible even in principle, never mind in practice. It would probably be necessary to run several different tests on several large groups in order to control for different variables separately.

Teaching the Himba language to children in English-speaking countries would almost certainly be the easier approach.

Originally Posted by Heliomance

I'm pretty sure there have been studies done on whether language influences culture (in the popular "they have no word for peace" fashion), and found little evidence to support the idea. I can't remember what to search for to find the results again though, I'm afraid.

Wikipedia article!

Originally Posted by Surfing HalfOrc

I was actually only wondering why game makers chose those four colors most commonly for pawns, especially for "children's" games.

It seems that those are the four most "basic" colors other than white and black. They're probably chosen as both primary colors and game piece colors for that reason (though the designation of a color as "primary" undoubtedly reinforces its status as "basic").

Were I picking out four colors meant to be as distinct from each other as possible, I'd probably go with white, black, red, and green. But in many cases you're working with a black and white background and want colors to stand out well against that background. In that context, red, yellow, green, and blue make the most sense. And of course non-neutral colors are more vibrant -- more "cheerful" as you put it.

Sorry! had white pawns (I think? It's been over 40 years since I last played) in addition to blue, green and red

A Google image search shows that Sorry! uses the RYGB model. Google seems to be pretty good for quickly checking what something looks like.

Also, it apparently lets you filter images by color, which I never really noticed before. Specifically, you can choose Full Color, Black & White, or... any of the 12 "basic" colors, including the one that English lacks a basic color term for ("teal", here).

Pretty nifty!