I see trees of green, red roses, too...

August 12, 2011 7:09 PM

Do you see what I see? Do people always see the same thing when they look at colours?
posted by crossoverman (68 comments total)

Slightly different, but related phenomenon: if I had a dollar for every customer who's tried to tell me "I want you to print the shirts in the same color blue as [name of obscure team which, like every other team, gets merchandise printed by 1,327,029 different suppliers on 3,129,072 different fabrics which oscillate around as many shades of blue]"...

Pantone for the win...

And finally, this seems to vindicate that weird kid I knew in junior high who liked to ask "what if what I see as red looks like what you see as blue; how would we ever know?"
posted by randomkeystrike at 7:23 PM on August 12, 2011

I was that kid!
posted by nev at 7:28 PM on August 12, 2011 [27 favorites]

Shouldn't you have posted this on the Orange?
posted by bondcliff at 7:28 PM on August 12, 2011 [14 favorites]

Even with Pantone it's really difficult to accurately reproduce or describe color. Pantone books age and fade over time. Really expensive projects often demand or stipulate the use of fresh, newly opened and recently produced Pantone color swatches or chits.

Then there's the ongoing (and often mind-bending) challenges trying to keep those colors consistent over different substrates or materials. The pigment or dye formulas to get the same color from a piece of injection molded plastic as you do from a dyed woven garment or piece of paper are drastically different, and it's amazing they even get close, much less spot on and nearly undetectably different.

When you see something at - say, Ikea - where it has different materials but nearly exactly the same colors matching across the different materials they're dropping some pretty heavy science and money into the product to make that happen and maintain that consistency across huge production runs.
posted by loquacious at 7:31 PM on August 12, 2011 [20 favorites]

Trivially, no, everyone doesn't see colors the same: Colorblind people, for example.
posted by smcameron at 7:32 PM on August 12, 2011

You say that if a language has a word for the color blue, it will almost certainly have a word for red, but not vice versa. What does this sort of asymmetry tell us about the perception of color?

First: not “almost certainly" but “certainly"—we don’t know of any exceptions to this rule. The initial conclusion of scholars in the nineteenth century was that this asymmetry reflects very recent biological improvements in color vision. It took a long time and a good deal of pain to come to terms with the realization that the development of color names reflects profound cultural changes, not anatomical ones. A large part of the book is spent trying to get to terms with the counterintuitive fact that the color distinctions we make are heavily influenced by cultural conventions and are not merely given to us by nature. The bottom line, very crudely, is this: People find a name for red before they develop a name for blue not because they can see the former before the latter, but because we find names for what we think is important to talk about, and red is more important in the life of people in all simpler cultures than blue.

--Paris Review interview with linguist Guy Deutscher. Deutscher's recent book, Through the Language Glass: Why the World Looks Different in Other Languages , about how language shapes culture, and how culture shapes language.
posted by jng at 7:32 PM on August 12, 2011 [13 favorites]

what if what I see as red looks like what you see as blue

Then how did the two of you end up calling that color, which you hypothetically see so differently, by the same word?

I mean, think of how children learn the names of colors. They see a schoolbus, and are taught that that color (in the U.S., anyway) is yellow. Let's say one child can't see the difference between yellow and red. He's asked to sort pictures by color, and puts a cherry in the same pile as the schoolbus. In that case, it's easy to see what's meant by "see differently".

Your childhood friend was stumbling toward the idea of qualia - a property of perceptions which has no bearing on behavior. I, and many others, think this is absurd. If you and I treat schoolbuses, big bird, and sunny delight as yellow - well, that's all there is to yellow. There is no room for ineffable properties to creep in.
posted by elektrotechnicus at 7:36 PM on August 12, 2011 [6 favorites]

What a stunning video. I, too, was quick to point out the difference on those color rings, but then again, red and green are the same to me. o.O
posted by cavalier at 7:38 PM on August 12, 2011

Distinguishing blue from green in language

Also, the Deutscher book is really fun in the discussion of how the color blue is seemingly never used by Homer. He discusses a lot of the outdated theories and some of the latest thinking.
posted by jng at 7:41 PM on August 12, 2011 [5 favorites]

Then how did the two of you end up calling that color, which you hypothetically see so differently, by the same word?

Because they're the same colour, and that's the word for the colour. But you each see them differently.

(You're probably just not high enough to grok it.)
posted by Sys Rq at 7:42 PM on August 12, 2011 [10 favorites]

Then there's the ongoing (and often mind-bending) challenges trying to keep those colors consistent over different substrates or materials. The pigment or dye formulas to get the same color from a piece of injection molded plastic as you do from a dyed woven garment or piece of paper are drastically different, and it's amazing they even get close, much less spot on and nearly undetectably different.

It's been a good number of years, but the first time I encountered metamerism, it blew my mind. I was trying to match a shade of purple fabric with plastisol, and the only thing that seemed to work was mostly fluorescent pink, and ultramarine blue. It was pretty close under the fluorescent bulbs near my ink area, but when I got the ink to the press which got a mix of daylight in from the open overhead door, it was significantly bluer than the fabric. I took it down the hall to an office with incandescent bulbs, and it was redder than the sample. Fluorescent pigments are especially troublesome for color matching, because they absorb ultraviolet light and re-emit it in the visible spectrum, thus putting out more visible light than they naturally reflect, which makes them appear brighter than their surroundings.

Crazy stuff, and I'm still working hard at learning how it all works.

if I had a dollar for every customer who's tried to tell me "I want you to print the shirts in the same color blue as [name of obscure team which, like every other team, gets merchandise printed by 1,327,029 different suppliers on 3,129,072 different fabrics which oscillate around as many shades of blue]"...

I've told the denim story here before.
posted by Devils Rancher at 7:48 PM on August 12, 2011 [5 favorites]

That circle on the left FREAKED ME OUT
posted by ThePinkSuperhero at 7:55 PM on August 12, 2011 [2 favorites]

What Color is the Sky in Your World?
posted by BeerFilter at 7:58 PM on August 12, 2011

See also Mary's Room.

I don't believe that Qualia is nonsense, though it almost certainly does not go so far as my red equaling your blue. There would still be a spectrum involved, presumably, with some shifts around it. The question is more that it is currently impossible to accurately share notes on the experience of seeing different colors.
posted by Navelgazer at 8:10 PM on August 12, 2011

"Dude... wouldn't it be weird if, like, my green was your orange?"
posted by Rhaomi at 8:13 PM on August 12, 2011 [2 favorites]

...I've been staring at the circle on the left for a while now, and I honestly still can't tell which one's supposed to be different! Help, please?
posted by estlin at 8:21 PM on August 12, 2011 [4 favorites]

In order to figure out which color was different on the left, I literally had to load the image into GIMP (preserving the Color Profile) and play with the eyedropper tool. When I flip back and forth between the two greens, I can see a shift, but I still can't distinguish it in the picture. Color's weird, man.
posted by jozxyqk at 8:23 PM on August 12, 2011

Yeah, a friend forwarded me this a couple of days ago, and, although I had always known that different cultures have different names for what is an ultimately mathematical way (in the Western culture) of distinguishing between blue and green, for example. But the way it is shown in this video gave me a totally new gut feeling of what it is to see colors differently. Amazing.
posted by kozad at 8:26 PM on August 12, 2011

When I was a kid, I asked myself that "dude" question, and since then I've read a lot of philosophy of mind. Before reading the link, I thought, "Of course you could never tell if someone else sees the same colors! Philosophy 101!"

After reading it: "Oh."
posted by abcde at 8:47 PM on August 12, 2011

...I've been staring at the circle on the left for a while now, and I honestly still can't tell which one's supposed to be different! Help, please?

Same position as on the right. I had to pull it into an image editor to convince myself.
posted by meinvt at 8:57 PM on August 12, 2011

I too was a kid somewhat obsessed with this question, and who explored it in my Philosophy of Mind program. Methodologically, I have some questions about these experiments as shown in the video (I don't assume that these questions were ignored by the studies, but they aren't addressed in what we have here.)

1. The Namibian tribe appears to be a tiny culture, with likely very little outside influence into the gene pool. (I might be wrong about this, obviously.) Aside from language, isn't there a very real possibility that the cones and rods in the eyes of the tribespeople have developed to be quite different from westerners?

2. While I can't see any difference in the colors in the left ring in the article, or when shown straight-out in the video, when the video showed footage of the monitor in the tent, I could see a difference instantly (one box had a distinguishingly more mustard hue than the others) and that was the one pointed out a moment later. Perhaps there are some presentation issues here?

3. Why wouldn't the man pick a more neutral shirt when conducting an experiment of having people place colors together? Probably made no difference but that shirt could cause grand mal seizures.

Still, all of this fascinates me. Just yesterday my friend described my car as "maroon," and I objected, having never thought of my car as being maroon. I would have described it as red, or perhaps burgandy. "Maroon," to me, having grown up in Texas, was only the specific color of the Texas A&M Aggie uniforms. Considering it further, perhaps Dr Pepper cans could fit the bill as well, but I have a very limited conception of maroon.

More broadly, the easiest example of this difference as a thought experiment is Pink. Pink is the only color which Westerners distinguish as fundamentally different based purely on value rather than hue. It is also a color with distinct connotations from birth. Language absolutely makes a difference here.
posted by Navelgazer at 9:08 PM on August 12, 2011 [2 favorites]

The left circle's square is really hard to detect on an LCD screen. Both the left-right and the up-down color shifts make slight color differences very difficult to distinguish from angle-based artifacts. I'm always wanting to turn my MacBook sideways to read comics, but even on this nice screen the left-right shift between my two eyes drives me crazy.

In addition, though my first instinct was the same location as the blue one, I thought that might be due to an afterimage effect from looking at the circle on the right. Color certainly is a noisy channel...
posted by chortly at 9:11 PM on August 12, 2011 [1 favorite]

However, when scientists visited a tribe in northern Namibia that has a completely different way of grouping and naming colors, they found that the exact opposite was true — the tribe members picked out the slightly different green square easily, while struggling to see the blue one.

Ah, sounds like bullshit, as far as I can perceive it.
posted by ovvl at 9:21 PM on August 12, 2011 [6 favorites]

"How do I know the color blue to you is the same as the color blue to me?"

"Check the crayon box, asshole! I'm fuckin' high over here!" -Dennis Miller
posted by ShutterBun at 9:44 PM on August 12, 2011 [2 favorites]

In the English language (and most languages), there are distinct words for “green” and “blue. Therefore, it’s very easy for most people to detect the blue square in the right ring, but difficult to detect the slightly different green square in the left one.

This has gotta be one of the worst utilizations of the word "therefore" I've ever encountered.
posted by ShutterBun at 9:49 PM on August 12, 2011 [7 favorites]

My eyes have a different white balance settings when looking at colors. QED, people can see colors differently.
posted by zippy at 10:13 PM on August 12, 2011 [1 favorite]

I don't care what color you see, get out of my room. I'm a super-scientist and I need to work.
posted by maryr at 10:18 PM on August 12, 2011 [2 favorites]

As Zippy says. My left and right eyes see slightly different colours. One eye sees bluer; the other, redder.
posted by five fresh fish at 10:44 PM on August 12, 2011 [2 favorites]

Well, that shirt is definitely the thing to wear if you want people to remember you as That Guy Who Is Maybe A Little Too Into Color.


Ah, sounds like bullshit, as far as I can perceive it.

I haven't seen anything this dramatic before, but there is evidence that language can affect color differentiation. The classic example is with Russian speakers - Russian has no single word for a generic "blue",there are two words for what English speakers would consider blue - siniy, light blue, and goluboy, dark blue. Russian speakers can differentiate between colors across those categories more easily than English speakers, but if they do something like count to 100 out loud while doing the color discrimination task, disrupting their ability to use language for color discrimination, they have no significant advantage. (Here's a video of Lera Boroditsky explaining that better than I can, starting at around the 14:00 mark.)

This is pretty crazy, though - I'd be interested to see if a language interference task would change the results in this case - at least from what's presented in that link, I don't see any proof that it's necessarily language that's causing the difference here. It's not totally unfounded, but from what's presented in that link I don't see why it has to be language rather than any other factor that differentiates them from people who see color the way most of us do.

Interesting stuff, in any case!
posted by Condroidulations! at 10:57 PM on August 12, 2011 [3 favorites]

And just for fun, try the Online Color Challenge where you need to rearrange the colors in each row by hue.
posted by MelanieL at 11:03 PM on August 12, 2011 [2 favorites]

It confused the heck out of me when I learned in Russian 101 that Russian has two words for blue.

Though, most of what I was taught in Russian 101 confused the heck out of me.
posted by BungaDunga at 11:06 PM on August 12, 2011 [1 favorite]

Of course, there is the matter of putting the cart before the horse. Wouldn't the assignment of distinct labels follow the distinction in perception rather than precede it?
posted by Gyan at 11:20 PM on August 12, 2011 [1 favorite]

What I'd like to know is what exactly the researchers were asking of the Himba tribe members when they were looking at the circle of green boxes with one blue box. Because if nothing else, the blue square is obviously much lighter in tone. If you desaturate both colors so they are only shades of gray, the formerly blue square is still much lighter. So, were they asking something other than "which square is different"? Because if you have the same word for blue and green, and that word is "glug," then asking which one isn't glug isn't going to get a response, obviously.

If they are trying to say that having a different (smaller) vocabulary for color names means you can't tell the difference between lighter and darker things, then I'm not feeling very convinced.

(Also, like Navelgazer, I was easily able to see the different green one on the circle of green boxes in the video, as opposed to the the jpg on the article page. I don't think the two are the same.)
posted by taz at 11:30 PM on August 12, 2011 [5 favorites]

English speakers only have one word for blue. (besides Metafilter)

Likewise, we only have one word for aqua, one word for cornflower, one word for periwinkle, one word for teal, one word for turquoise, one word for navy, one word for cobalt, etc.
posted by ShutterBun at 11:31 PM on August 12, 2011 [9 favorites]

as opposed to the the jpg on the article page.

Indeed. The percentage of difference for the color balance is nearly within the normal variation range of the jpeg artifacts, according to the eyedropper tool.
posted by ShutterBun at 11:38 PM on August 12, 2011 [1 favorite]

This has always been really interesting to me, and is a great find, crossoverman.
Dovetailing with the colour perception is something I've always wondered: eye colour, and the impressions people get when talking to someone whose eye colour is not what you're used to as most people, in my experience look at your eyes when you're talking to them. If you're from a culture that has predominantly brown eyes, and you're not really used to talking to someone with blue or green eyes, how does that effect the way that you'd interact with them? An example could be from the clip linked in the post - would the people of the Namibian tribe behaved differently towards a researcher who had brown eyes because that's what you'd be used to seeing, and not the blue eyes of the researcher.
posted by Zack_Replica at 11:53 PM on August 12, 2011 [1 favorite]

It's been a good number of years, but the first time I encountered metamerism, it blew my mind.

Oh, awesome, there's a word for "what happens to the tomato soup when I take it from the kitchen to the dining room"!
posted by NMcCoy at 12:11 AM on August 13, 2011

Always bring a grey card to every meal.
posted by ShutterBun at 12:24 AM on August 13, 2011 [1 favorite]

All I know is...

When it comes to mixtures of blue and green, I will class something as across the "more greenish than blueish" line, and everyone else classes it as on the "more blueish than greenish" line. I do not know why. I feel I have a good grasp on what blue and green are.

Not sure what is going on with this.
posted by marble at 1:15 AM on August 13, 2011

My eyes have a different white balance settings when looking at colors.

Wow me too - how interesting! Never noticed that ...
posted by iotic at 1:51 AM on August 13, 2011

Me & gf caught that episode of Horizon by accident the other day and it was pretty fascinating. We both picked the odd green out of the circle in a couple of seconds on TV, though it is difficult.

Just did the 'online color challenge' mentioned above, to see if my colour vision is better than average - I got a score of 3 which I think is nice and high, but the results page disappointingly didn't tell me what average is.
posted by dickasso at 2:02 AM on August 13, 2011

Condroidulations!: "siniy, light blue, and goluboy, dark blue"

Vice versa.

What fascinates me is that light and sound can both be described in terms of frequencies and spectra. But the way we perceive them is radically different.

With light, we have three sensors which are stimulated to various degrees by different frequencies, and that's it. You could have two spectra, radically different on a spectrometer, which look identical to the eye. The monitor you're looking at uses this fact to fake colors: it only produces three colors, to stimulate your three receptors. But by carefully controlling the intensity of each, it can simulate the effect of any frequency on your eye!

It's as if you could make a piano with only three keys, but which would make you believe you heard any note, just by combining the three notes in a variety of intensities. And all within one octave.

Which leads you to realize that your perception of color is ridiculously limited (all the colors you see can be described as one three dimensional space, like the one you find in Photoshop, while accurately describing all potential colors would require an almost infinite-dimensional space). Colorblind people with only two types of cones cannot see the differences between certain colors-- they're missing out, yeah? Well, compare yourself to a hypothetical person with four types of cones-- or five, or ten. He would be able to distinguish a host of new colors, which all look alike to you.

So if we ever meet aliens, one thing is almost certain-- color will not be something that we'll find common ground over.
posted by alexei at 4:18 AM on August 13, 2011 [3 favorites]

The study of Himba color naming mentioned in the program is presumably this one [pdf] (Color categories: Evidence for the cultural relativity hypothesis, Debi Roberson, Jules Davidoff, Ian R.L. Davies, Laura R. Shapiro, Cognitive Psychology 50 (2005) 378–411).

The paper includes color charts for Himba, English and a third language, Berinmo, that's closer to Himba than English. The charts are in black and white with Munsell hue and value designations. You can kind of make sense of them even if you don't know the Munsell system, by comparing the other two charts to English.

The Munsell hue designations are: R red, YR yellow-red (orange), Y yellow, YG yellow-green, G green, BG blue-green, B (greenish) blue (cyan, teal), BP blue, P purple, RP red-purple (pink, pinkish or purplish red).
posted by nangar at 4:19 AM on August 13, 2011 [2 favorites]

alexei: compare yourself to a hypothetical person with four types of cones-- or five, or ten. He would be able to distinguish a host of new colors, which all look alike to you.

What the cones are sensitive to are frequency bands, so IOW the color spectrum one sees is a collinear mapping to a frequency spectrum. Even with five cones, it's not "almost certain" that the alien brain would generate novel chromatic apparitions to represent the same frequency range for visible light. They may experience more emphatic discrimination between two nearby shades of a hue, though.
posted by Gyan at 4:55 AM on August 13, 2011

So if we ever meet aliens, one thing is almost certain-- color will not be something that we'll find common ground over.

Good question. If there was a use for more cones, or a much wider bandwidth in vision for terrestrial-like environments, one might expect to find it provided by evolution, here on earth. I don't know how much cone number and frequency varies, though it seems insects have developed to perceive ultraviolet light, and given it's use in night-vision applications, one can foresee infra-red being handy too.

This wiki article contains an interesting thought: In general, the optical spectrum encompasses the most common electronic transitions in matter and is therefore the most useful for collecting information about the environment.
posted by iotic at 5:08 AM on August 13, 2011

Looks like there are, at least, some pentachromats, including pigeons and butterflies.
posted by iotic at 5:20 AM on August 13, 2011

As a person who does design work, and is also slightly red-green colorblind, I find color to be an incredibly fascinating/frustrating topic. I work with people who don't have the same issues I do and sometimes it's almost like they speak a language I've never learned.

Very,very interesting topic. Thanks for the link.
posted by Benny Andajetz at 5:25 AM on August 13, 2011

Crazy stuff, and I'm still working hard at learning how it all works... the first time I encountered metamerism, it blew my mind.
posted by Devils Rancher

If you decide to crack open some of the technical literature, say The Reproduction of Color (Hunt,) or Color Engineering (Green, MacDonald,) you're in for a serious slog and perhaps some brilliantly colored migraines. You should set aside a decade or so, if you want to have a working knowledge of all the things that figure into matching and reproducing color experience.

Just one eye-opener, regrading metamerism: there's a little parlor trick developed by people who sell controlled color-viewing boxes. You're looking at a small hooded tabletop stage with built in lighting. On the stage are small arrangements, small still-lifes of models of food and beverages, with a left and right grouping. In the first light setting both arrangements look edible, both show seemingly natural coloration, but they are slightly different in color.

Then a switch is flipped and the illumination spectrum is altered. One grouping changes as you might expect, and looks about the same, just differently lit. The other grouping changes in a bizarre way, into something completely unnatural and unappetizing.

The pigmentation of each group is very carefully constructed. For starters, in real life, neither pigments nor illumination are comprised of primary color components, that's a reductive shorthand. Both are comprised of continuous weightings of the electromagnetic spectrum, that could only be described by a uniquely shaped curve for the entire visible range.

The point by point multiplication of the two curves, to determine the result of illumination, is not constant, and varies at each position crossed with every position in the other spectrum. By exploiting the ways that this complex result does not match the color component approximation, very unexpected results can be demonstrated.

Every issue like this that is examined, opens up ten more scientific inquiries.
posted by StickyCarpet at 5:39 AM on August 13, 2011 [2 favorites]

In general, the optical spectrum encompasses the most common electronic transitions in matter and is therefore the most useful for collecting information about the environment.

It's been speculated that the most prominent human color experience, being the reg-green discrimination, is centered where it is primarily to see who's bleeding.
posted by StickyCarpet at 5:45 AM on August 13, 2011

There is no room for ineffable properties to creep in.

Sure there are. There are lots of things that exist but are ineffable. A world where language is the only reality is a dull and dead one.
posted by empath at 6:05 AM on August 13, 2011

This question was highly topical (among linguists) in the late 1960s and 1970s. Post Berlin and Kay, linguists sort of figured out that people see the same things, even though they divide the spectrum up using different categories and terms. Maps, territories, etc.
posted by spitbull at 6:57 AM on August 13, 2011

Then how did the two of you end up calling that color, which you hypothetically see so differently, by the same word?

Because it wasn't just two people, it's the whole damn society. Just because the majority see things one way and develop a consensus around that has no effect whatsoever on the immutable fact that I have three towels, two of which are blue and one is green and I can't tell the difference between them. Nor can I read color-coded maps or distinguish between red and green LEDs. Most computer displays use monochromatic phosphors that exacerbate the problem.

Next step: some total fucking asshole starts pointing at things and asking me what color they are.

Piss off: color coding doesn't work for some people and you can't fix it.
posted by warbaby at 7:16 AM on August 13, 2011 [1 favorite]

Okay, maybe this is something that somebody can help me with. I'm suddenly vaguely remembering a novel I read a few years ago. In it, in a series of asides, there were offered variations on the "Mary's Room" thought experiment, including one in which Mary uses violence to break out, and one in which biology invalidates the experiment. Can anybody help me to remember which book it was? Was it a Richard Powers book?
posted by .kobayashi. at 7:28 AM on August 13, 2011

Even people with seemingly normal color vision can have their color perception altered by, for instance, medications. There was a warning issued to pilots about flying with Viagra in their systems, because it can change the way colors are perceived enough to mess with judgments about color-coded landing-strip lights.

Myself, I had a medication-induced color anomaly. I was walking along a busy section of Broadway, and my eye was drawn to a T-shirt on someone, way off in the distance. It was a certain variation on cranberry, and it caught my eye as being distinctive. At the time I even thought to myself, "that's odd, why am I noticing a T-shirt three blocks away?" I'd recently been prescribed a certain medication, and I eventually realized that there was a not unpleasant side effect, that was dosage dependent. This medication increased my discrimination in a certain range of the bluish-red spectrum.
posted by StickyCarpet at 7:34 AM on August 13, 2011

Oh hey, someone else has discovered qualia.
posted by Eideteker at 8:21 AM on August 13, 2011

My eyedropper tool claims more than one square in the first circle of that jpeg differs from the others. The correct one more than the others, but still. It's definitely not something you can show very well with jpegs and LCD monitors.
posted by straight at 8:25 AM on August 13, 2011

I've been thinking about concepts a lot and this is all you need to solve these problems:
Color concepts are just bus stops on the color space, color perception is a load of ants dropped on a certain point on color space, ants can walk around and take buses and the first ants who get to home can tell their stories.
posted by Free word order! at 8:32 AM on August 13, 2011 [1 favorite]

It confused the heck out of me when I learned in Russian 101 that Russian has two words for blue.

Hey, remember ROYGBIV? There's two words for blue in there.
posted by Sys Rq at 8:53 AM on August 13, 2011

one word for turquoise,


Turquoise is another word like "maroon" which seems to encompass different colors to different people. I've discovered I am far more rigid in my interpretation of Turquoise than other people. I've seen design blogs were colors from Peacock Blue to Sage Green were labeled as "Turquoise." Above all, pastels. I don't think Turquoise is a pastel, ever, but apparently I am mistaken.
posted by Secret Life of Gravy at 9:18 AM on August 13, 2011

I don't think Turquoise is a pastel, ever, but apparently I am mistaken.

You are, certainly. Turquoise is the colour of turquoise. You're probably thinking of teal.
posted by Sys Rq at 9:20 AM on August 13, 2011

Hey, remember ROYGBIV? There's two words for blue in there.

Last time I looked, the French Wikipedia article on Indigo was mostly a rant about the "myth of the seventh color" and how it was "invented" by Isaac Newton to "make them coincide with ... other culturally important heptads."

Somebody needs to go tell the people on Chinese Wikipedia that. They made an entire category page for it.
posted by nangar at 9:26 AM on August 13, 2011 [2 favorites]

Nangar's link shows pretty clearly that the example on the OP's link is a massive oversimplification of the original research. Read that Davidoff paper and you'll also see they used differently shaded pictures of cattle. I'd like to see that, actually.
posted by yellowcandy at 9:41 AM on August 13, 2011

No reason I can think of there couldn't be developmental issues with color perception such as there are with phoneme perception.

You lose the ability to distinguish phonemes that aren't distinguished in languages you aren't exposed to early enough. You lose ability to hear the difference even though the fact that other human beings can hear it demonstrates that it exists.

Color could be the same way; you could lose the ability to perceive colors which aren't distinguished in any way by your culture.

Language would certainly be one of the important ways your culture would distinguish colors, and that could lead to these surprising results where what your language talks about appears to affect the most basic properties of what you see.

In the case of phonemes, later training seems to be able to allow previously indistinguishable phonemes to be discriminated, but only to a limited extent, suggesting that apoptosis, or selective neuron death, may play an important role.

I'd guess the same thing is true of color, and that the range of colors a person an perceive is, as with phonemes, partly determined by what happens during a critical period: if someone is teaching you who makes a color distinction which has some objective physical basis, you probably will be able to as well, if you have physical equipment as good as theirs.

It seems to me that in such a situation, the number of colors an average person could distinguish would tend to increase over time, with complex feedbacks to and from language.

I wonder if groups of babies and very young children might not be able to teach themselves color distinctions that don't exist in their larger culture by handling objects with very similar, but not identical reflectance profiles and putting them in piles, etcetera, and giving others the chance to think those distinctions are important.
posted by jamjam at 11:22 AM on August 13, 2011 [3 favorites]

I generally have thought that I possess a very good ability to distinguish colours, what clashes, what complements, etc. I could not pick out the slightly different green square for the life of me. Even once I knew which one it was supposed to be, I thought it was dubious! This is extremely intriguing.
posted by vaevictus at 12:48 PM on August 13, 2011

Interesting video, though I wish they showed us the patterns of "colors that go together" that Mr. Colorshirt is able to predict so accurately. Also he claims that different sensitivities to changes in color varied with sex, age and "level of status" -- wish I knew what he was referring to with that last one. (Actually I think I do know; previous clips are of a man's face and then a woman's face, young people and then old people, and... a bunch of close-ups of a young black kid clicking a mouse. Anyway. It'd be nice to know what exactly the variance in question was.)

I have a rather odd color/language problem when describing colors: I often blurt out the completely wrong color. I'll be pointing out that guy over there in the yellow shirt and I'll say "that guy over there in the purple shirt." I know it's yellow, I know I know the word for yellow, I don't conceptually confuse yellowness and purpleness as concepts, but at the last minute "purple" comes out. It feels similar to when you're speaking one non-native language and in the middle of the sentence your brain hands you the right word, but from a different non-native language. This happens for all colors. There's no consistent pattern I can discern (though I haven't discerned too hard). This only happens when I'm describing colors I'm actually looking at as I'm speaking.
posted by DLWM at 2:19 PM on August 13, 2011 [1 favorite]

For those of you unable to distinguish the two shades of green in the left circle, it's entirely possible that it's your monitor that's unable to display them as two different shades. I can see that they're different, but only barely, despite what the RGB values say, as numbers. And I've recently calibrated my monitor with a Spyder 3 thingy, and believe me, after calibrating about 6 monitors with it that I thought were decently calibrated beforehand, your monitor is probably not calibrated all that well.
posted by Devils Rancher at 5:03 PM on August 13, 2011

How utterly amazing, DLWM!

I almost don't want to say this to you before asking you for other examples, but consider the following from the Wikipedia article on complementary colors:

The complement of each primary color (red, blue, or yellow) is roughly the color made by mixing the other two in a subtractive system: ...


Mixing red and blue gives purple, of course-- so you are blurting out the complementary color to the color you actually see!

But you actually are seeing that complementary color in any portion of your retina that was looking at the yellow, because that portion of your retina is fatigued in yellow, and therefore sees the complement of yellow when it looks at something white.

Moreover, as you eyes shift around minutely as you fixate on something-- which is how fixation operates-- the retina will experience a halo of the complementary color around a colored object, but there must be a program to suppress perception of that halo.

I've thought that program must operate at the level of the retina, but if you can see it well enough to blurt it, the program must be in the brain.

Color Tourettes! Who could have guessed?
posted by jamjam at 5:40 PM on August 13, 2011

Interesting note I saw in The Straight Dope some years ago: a study of current and ancient languages showed that, in general, when a language only had two words for colors, the colors were black and white. When a language had three words for colors, they tended to be black, white, and red. Note that a tonne of national flags have white and red as two of their colors.
posted by JimDe at 1:23 PM on August 14, 2011