Wiring the brain

This story is some kind of awesome:

For those who don’t want to watch the whole thing, the observation in brief is that color perception is affected by color language. The investigators compare Westerners with our familiar language categories for color (red, blue, green, yellow, etc.) to the people of the Himba tribe in Africa who have very different categories: they use “zoozu”, for instance, for dark colors, which includes reds, greens, blues, and purples, “vapa” for white and some yellows, “borou” for specific shades of green and blue. Linguistically, they lump together some colors for which we have distinct names, and they also discriminate other colors that we lump together as one.

The cool thing about it all is that when they give adults a color discrimination test, there are differences in how readily we process and recognize different colors that corresponds well to our language categories. Perception in the brain is colored (see what I did there?) by our experiences while growing up.

The study is still missing one part, though. It’s presented as an example of plasticity in wiring the brain, where language modulates color perception…but we don’t know whether people of the Himba tribe might also have subtle genetic differences that effect color processing. The next cool experiment would be to raise a European/American child in a Himba home, or a Himba child in a Western home (this latter experiment is more likely to occur than the former, admittedly) and see if the differences are due entirely to language, or whether there are some actual inherited differences. It would also be interesting to see if adults who learned to be bilingual late experience any shifts in color perception.

(Also on Sb)


  1. says

    That’s what color discrimination tests are for. If I show you a set of disks, all of which reflect in the red, but one of which also reflects at an infrared wavelength which is invisible to most people’s eyes, would you be able to pick out the bleen-colored disk every time?

  2. peterh says

    I see lots of cultural conditioning in the examples given above; is there actual evidence of physio-neurological differences? The color discrimination test mentioned seems bound primarily by linguistic not neurological factors.

  3. AussieMike says

    That is just freaking incredible. How amazing this world, it animals and its people are. How sad that some try to resolve it down to being gods ‘way’! Rational reasoned thinking is to the rainbow as religion is to black and white.

  4. jan says

    concur with #3

    the difference in perception from culture/language are well known in linguistics and anthropology since the early sixties.

    The new stuff (baby brain imaging of color perception) is not really related to the second part (ethnolinguistic fieldwork on said color perception).

  5. ChasCPeterson says

    The new stuff (baby brain imaging of color perception) is not really related to the second part (ethnolinguistic fieldwork on said color perception).

    Did you watch the video? Did you read the post?
    a) there is no brain imaging discussed
    b) of course they’re related. The infant studies are elucidating the process and mechanism by which language and culture influence perception.

    You seem to think that ethnolinguistic fieldwork is all that’s required to understand these differences in perception among populations. That’s a pretty myopic view. (Apologies if it’s inaccurately attributed; this is what I infer from your comment.)

  6. ChasCPeterson says

    and btw, drawing a bright line between ‘linguistic’ and ‘neurological’ factors is just stupid.

  7. madtom1999 says

    Do they suffer from that red/blue red/green margin wobble thingy too – and what is that called?

  8. machintelligence says

    Are they certain that the members of this tribe actually have color receptors that are identical to those of typical Westerners? There is the possibility that the language follows the perception rather than drives it.

    In the case of RG colorblind people who also have synesthesia (perceiving numbers as having colors), they say that some numbers have “martian” colors — they have never seen them in nature. This would seem to indicate that the “wiring” to perceive the whole spectrum is functional, even though the “hardware” (photoreceptors) are defective. See the work of V.S. Ramachandran.

  9. Dubs says

    I seem to recall reading something similar about how men & women see colors differently. It’s usually seen most in the context of stand-up comics, or epitomized in the line from Steel Magnolias… “My wedding colors are blush & bashful.” “Your wedding colors are pink and pink.”

    If you develop a vocabulary distinguishing a difference between ecru & eggshell, you can see the difference, but to someone who hasn’t developed the same vocabulary, they’re all just white.

    What I find most interesting about the Himbu tribe was the bleed over between categories. It’s easy to see how blues and greens can be in a single color group, but for reds & blues to co-exist as a single color-group is counter-intuitive, at least to me.

  10. says

    With these things I always wonder how you could isolate differences caused by language as opposed to all the differences of growing up in a different culture. The language differences might just be a reflection of this.

  11. ICMike says

    Way back in 1986 Cecil Adams, author of the Straight Dope column in the Chicago Reader, addressed a similar question.

    Cecil relied on a book by Berlin and Kay for much of his answer (Basic Color Terms: Their Universality and Evolution, 1969). The interesting point with regard to the current discussion is that Berlin and Kay were also relating color and language; they looked at 98 languages, and were able to develop a set of rules relating the number of different colors described by a language and the particular colors the languages described. The entire column, which includes Berlin and Kay’s rules and some history of the subject, is worth a read.

  12. peterh says

    V. S. Ramachandran was mentioned above; some might try his Phantoms in the Brain, Morrow & Co., 1998, ISBN 0-688-12547-3. It’s a fascinating tour of some of the brain’s neurological wonders.

  13. says

    I don’t see how this differs from, say, how language shapes our perceptions of sound. It’s clear that children notice differences pre-language that they don’t notice after language, as the latter either lumps sounds together, or treats them as unimportant, or rather, you cease even to “hear” the difference that are “unimportant.”

    Abstractions play a part of reception/interpretation, where our knowledge is input into the neural “data stream.” That’s how we manage to see what’s literally “not there” to any camera yet likely is there in very fact. Such as, we can see the form of a simple leaf that is covered and not visible, so long as we see enough of the rest of the leaf.

    This is stuff that makes it so hard to get computers to “see” the world as we do. Knowledge/information is heavily involved in how we see the world, and we found that out to a significant degree because computers did not “see” the world much like we did at all.

    It is highly unlikely that anyone would bother checking to see if there is anything genetic behind it. Odds are very good that there isn’t, that it’s a matter of experience modifying perception. Someone mentioned culture, but then I think that’s more or less bound up with linguistic differences, so that even though clearly it has something to do with cultural attention, these cultural differences in attention can be considered to be covered by linguistic learning and usage for the purposes of these observations.

    Glen Davidson

  14. says

    I agree with Dubs (13). The color categories seem baffling. I could understand if all blues and greens were in one category, for example, but they seem to be spread in some incomprehensible (to me) manner through several Himbu color categories. Same for red. And it didn’t seem to be just about dark or light either.

    I’d love to see a full Himbu color chart.

  15. says

    Knowledge/information is heavily involved in how we see the world

    Plus, of course, experience, that is, practice in observing (what does a trained baseball player see in a pitch that one who never even saw the game does?). Likely we could learn to discriminate greens much more finely than we do, if we could get our brains to suppose that it was important to do so. Children would learn a good deal better how to perceive what is “important,” language being one of the more powerful factors in determining “importance.”

    Glen Davidson

  16. Carlie says

    seem to recall reading something similar about how men & women see colors differently. It’s usually seen most in the context of stand-up comics, or epitomized in the line from Steel Magnolias… “My wedding colors are blush & bashful.” “Your wedding colors are pink and pink.”

    Both of those quotes were spoken by female characters.

  17. Tim says

    So, what are these eleven words for colors that English is supposed to have? I figure red, yellow, blue, orange, green, and purple are in there. Probably also white, black, and brown. So, what are other two? I suspect that only people who work professionally with color printing are likely to think of cyan and magenta as basic colors. There’s pink, tan, and grey, but those are just light versions of red, brown, and black (plus, they would bring us up to twelve words). And, the works of Mr. Roy G. Biv notwithstanding, I don’t think indigo counts.

  18. Dubs says

    Carlie – I apologize; I didn’t mean to conflate the gender differences with the actual lines spoken, just that those lines are a great example of the distinctions between one person seeing shades where another doesn’t.

    Tim – tan’s your only outlier. A quick google gives me the below as the 11 categories:
    red, orange, yellow, green, blue, purple, pink, brown, grey, black and white.

  19. Dean Buchanan says

    There is a difference between looking at color on a computer monitor where the light is projecting onto our eyes and looking at other objects where the wavelengths that are not absorbed by the material are reflected onto our eyes. I have ‘seen’ this phenomena regularly over the past 13 years working in the design industry.

    I am not sure whether this would affect the research, but it might.

    In addition, the perception of color is greatly affected by the surrounding colors and texture of the surface. In short, our perception of color is very, very contextual.

    It would also be interesting to see if adults who learned to be bilingual late experience any shifts in color perception

    While not exactly becoming bi-lingual, as a customer or professional learns the language that designers use in describing small differences in color, over time their perception of color grows tremendously and they can spot differences between various hues and shades that before appeared to them to be the same.

  20. stan says

    I would be very surprised to hear that cones function differently in the Himba versus ‘Westerners,’ but I instead suspect that language has a profound effect on visual processing. That is, the eyes work the same, but the information provided to the brain is handled differently depending on the subtleties of language.

    Hue distinction is not a project undertaken by the eye itself, but by the brain’s processing of raw visual data. It seems quite clear that linguistic nuance might have a measurable effect on that process.

    There is, however, a eugenic or culture-centric worry here: if language can so affect one’s visual processing, then might it also affect one’s processing in other areas (i.e. problem-solving, spatial relations, application of logic), and might that not imply a hierarchy of language — insofar as the affected areas are themselves granted some ordered importance?

    Simply put, this might be seen to give a sense of legitimacy to claims such as, ‘Asians are better at math.’

    I’m certainly not trying to support culture-centrism, and I’m very definitely interested in what the science says here, but nonetheless it seems clear that a possible implication of this sort of finding (given no meaningful genetic differences and the sort of follow-up experiment PZ recommends) is that if we want to be good at X, then we should learn language Y at as early an age as possible. Whether or not this is an actual worry is up to you.


  21. RFW says

    Color vision is a fascinating field of research. I was lucky enough to hear Feynman’s lecture on the subject when I was an undergrad at Caltech in the 1960’s, complete with a light show projected on the screen.

    Two highlights:

    1. Brown light. Feynman showed the effect by projecting a disk of dim pink light surrounded by an annulus of brighter white light. The sensation “brown” is dependent on brightness contrast.

    2. Individual differences in color perception. By using three projectors with filters (RGB) and rheostats, any mix could be projected on the screen. Feynman would project a spot of color, then use the other three projectors to match it, asking for audience response to signal matching. He then asked if anyone saw the projected patches as not matching, and of course there were a few who felt a slightly different mix was necessary to match. They were invited to adjust the rheostats to obtain, in their view, a match.

    Oliver Sacks’ book, The Island of the Colorblind, is also well worth reading by anyone interested in the general topic.

  22. kijibaji says

    The particular use of reaction time shown in the video seems like a poor methodological choice for this experiment. The Himba participants presumably have very little experience doing such tasks (as opposed to the university students that complete these studies in Western countries) and that is bound to effect the results. Yes, the researchers would probably do some normalisation of the results, but still. And if you want to make claims that colour perception is affected by language, then a better reaction time methodology would be a speeded discrimination task so that participants don’t have time to activate linguistic categories and you can thus avoid introducing a big fat confound.

  23. Dean Buchanan says

    @Glen #20

    language being one of the more powerful factors in determining “importance.”

    Exactly. In this case, what we talk about focuses our perceptions on certain features of the environment, colors, thus changing our brains. As a designer and client discuss whether things ‘go’ together, they are exploring a broad range of factors like texture, motif, durability, and etc. In this larger context, colors literally look different to us depending on what is ‘important’.

    I can imagine being able to perceive a lot of different greens would be very valuable to a desert dwelling group.

  24. amphiox says

    There’s pink, tan, and grey, but those are just light versions of red, brown, and black (plus, they would bring us up to twelve words).

    Think instead of pink, tan, and grey being the half-way mix point between white and red, brown, and black, just as yellow is the half-way mix between red and green, and purple is the half-way mix between red and blue.

    And to expand on #23, pink and grey are very prominent concepts in the English language, and the definitive distinction between red and pink, pink and white, white and grey, and black and grey are commonly used in a variety of expressions and metaphors.

    That’s all consistent with their status as major color categories.

    Tan, though, is used much more rarely, and is more a specialist’s distinction.

  25. amphiox says

    I can imagine being able to perceive a lot of different greens would be very valuable to a desert dwelling group.

    For me with my western-based language imprinting, it boggles the mind that it’s even possible for there to be any circumstance wherein the ability to distinguish reds from blues is so unimportant that there need only be one language category for both colors.

    But I think that’s part of the point of choosing to use the Himba as an example, to demonstrate just how distinct cultural differences can really be, how inconceivably “alien” it can appear to a naive native-english speaker.

    The red-blue distinction is probably one of the most important in English, almost as prominent as the white-black distinction. The Himba example hammers home that something we may automatically think is fundamental need not be so.

  26. blbt5 says

    One of the more interesting posts, especially the ethnic difference between green and blue. The video could do with a few comments about color physics, however. For example, although green and blue seem quite different, green is blue with a tiny bit of yellow, and the amount of yellow needed to turn blue into green is far less than the amount of yellow needed to turn red to orange or the amount of blue to turn red to violet. The human eye can discriminate a nanometer of wavelength in the yellow to orange monochromatic range, but far less on either side and this differential sensitivity makes discrimination complex with colors that are reflected combinations of several colors, such as brown, for which coincidentally there are far fewer descriptive European words than say a primary color such as red, which lists in a thesaurus a great many words for various intensities and shades. It’s also odd that the video doesn’t note that water is clear and that most in the Western world are aware that water takes on a combination of the color of the sky and its dissolved contents, so water can be blue, green or brown (or clear) depending on its context. I wonder if the Himbi have an awareness of this.

  27. Jacques says

    This kind of research is not exactly new, and frankly, I find the results to be completely underwhelming.Take the Russian-English case that was mentioned here earlier (http://www.pnas.org/content/104/19/7780.full): Both English and Russian had nearly identical color boundaries, the difference was that Russian had two different words (goluboy and siniy), whereas English speakers had… two different labels that shared a common word (light blue and dark blue). So it is clear that whatever difference language makes is not in how these colors are actually perceived, but in how the linguistic labels might affect performance under some testing conditions. And these effects are really tiny (ignoring all other possible objections, around 100 ms). How can people sell these findings as “language makes you *see* the world in an entirely different way!” is sincerely beyond me.

  28. Davric says

    I noticed this phenomenon in Turkey once when I was teaching a group of Turkish students the names for hair colour, complexion, etc.

    In a room of people who, basically, all looked the same to me, they saw about four or five distinctly separate hair colours and complexions. I had about the same colouring as PZ at the time, which they saw as ‘fair’ or ‘blond’ (and they kept to this judgement when shown pictures of faces and hair colours I saw as fair and blond!).

    A year earlier I was in Angola (teaching marine biologists, as it happened) and one break they asked me what I did back in Sweden.

    “I teach them English.”

    “But don’t all Europeans already speak English?”

    At this point I started trying to explain the difference between blond blue-eyed Swedes, dark-eyed Italians, etc. They stopped me apologetically and one of them said, “We’re really sorry, but all you Europeans look the same to us.”

  29. NelC says

    As a graphic designer, I’ve gotten used to defining colours by their HSL (Hue, Saturation and Luminosity) values, as this is most useful for matching colours, or finding complementary colours or any number of colour manipulations. Not a solid visualisation of the whole 100 levels of each that Photoshop allows you to define, you understand; just able to spot the difference between two colours and understand it in terms of whether one or the other needs to be “darker” or “grayer” or “warmer” reasonably reliably.

    I guess the HSL system counts as a vocabulary, though it’s a very technical one, and as I said I don’t have a solid visualisation of each of 1,000,000 colours or even a majority of them, but then, I came to this vocab when I was already an adult. I guess my conscious comprehension of colour is quite rich because of it, though I don’t know that my perceptions are sharpened.

  30. rusma says

    There’s a nice book about this subject. “Through The Language Glass” from Guy Deutscher. He compares the definition of colors in different languages and cultures. For example the old greeks used a similar color definitionn like the Himba.

  31. Rebekka says

    I remember a similar experiment with an Australian language called Guugu Yimithirr. It uses cardinal terms of direction instead of egocentric ones, so they don’t say “left” “right” “in front of” etc., but instead use the equivalents of West, East, North and South. The speakers of this language had a very different perception of mirrored images than English speakers, for example. Better explained in this link here: http://www.nytimes.com/2010/08/29/magazine/29language-t.html?_r=2&pagewanted=4

  32. says

    I have a grandson who, when he was small, taught me to see reds differently. I was teaching him the colours, as one does with small children, and ran into difficulties; green was fine, blue ditto, but red, no. Each “red” (pure red, scarlet, cherry, burgundy, mamey, etc.) was a different colour to him, and he insisted on different words for each.

    Next time I see him, I’ll ask whether he still sees a whole spectrum of colours other people call “red”.

  33. Tim says

    I really should have done a Google search before I posted my earlier comment. I guess it just seemed like such a strange number, I must have assumed they made it up. It is the BBC and science, after all.

    But, I guess I can see why pink and grey are included, given their linguistic importance. I’ve occasionally wondered in the past why we have a special word for “light red”, when we don’t have ones for “light blue”, “light green”, etc.

  34. SirBedevere says

    Fascinating stuff. And I now have a way to introduce the concept of RGB and CMYK color systems in my Photoshop class next week!

    Like PZ, I’d be interested to find out if there are any inherited physiological differences in these people. I also wonder what people with tetrachromatic vision would make of the Himba color system.

  35. amphiox says

    It would be pretty big news if there was actually a genetic difference in the cone opsins. As far as I know, excepting known color-blindness mutations, the cone opsins in humans are supposed to be fixed.

    Neural plasticity in color perception circuits in the brain probably develops concurrent with the acquisition of the language for naming colors, so the two occur together and reinforce one another by feedback.

    So it’s probably a cultural factor at the root of the difference in perception, and such cultural considerations could possibly predate the development of true modern language – ie ancient human groups may well have been perceiving colors differently before they fully acquired the ability to communicate such perceptions with language.

    There’s certainly the possibility that such differences of perception are adaptive, reflecting various selection pressures of different native environments in terms of what kinds of color distinction are most advantageous. But it’s also possible that some aspects are stochastic/random/accidental, an early trend that just happened to become culturally fixed.

  36. says

    Perhaps the next experiment they could do would be one where the Himba have to learn tasks where the differentiation between blue and green is vital. It might be like learning a new language, as you are more exposed to it you start to find the breaks between words that you couldn’t hear before. I think the role language plays in this is only about focusing of attention and thereby handing down the differences to children who learn the language. I don’t think it can explain the origin of the differences, these could be due to genetics or the particular niches people grow up in or perhaps random accidents. If their particular niche changed to the extent that telling the difference between green and blue mattered I think their ability would change and their language would change with it.

  37. ChasCPeterson says

    I sincerely doubt that PZ’s reference to possible subtle genetic effects included anything as basic as cone opsins (btw, a difference there would be said to affect sensation, not perception). I think he’s talking about patterns of retinal/brain wiring, and of course we have little to no idea how it is that genes establish those (though there is no doubt that they do, or can, do so somehow).

  38. F says

    What color does sniny have?

    I haven’t found a reference yet, but I recall a similar study involving shape and object identification between cultures, which has implications for what Stan considers to be a worry.

    The one case involved some outdoorsy tribal sort of culture in a warm clime. The people had a harder time distinguishing between, or recalling if they had previously seen, things like scissors or pencils, but they could easily distinguish between two very similar-looking stones and laugh at the researchers for their complete inability to do so.

    As for Stan’s comment, it doesn’t mean they are “not as smart” or whatever, regardless as to how a handful of idiots would like to spin such a fact. It just means that their mental faculties are very well adapted to their environment and culture, and are not adapted to things which are not a part of their world. If, however, individuals of one society thought they would like to participate in some other society, it would obviously behoove them to learn the things that would be necessary to operate in the adopted culture. For those who would like to make derogatory points of how people of one culture have a “lower IQ” in terms of their own culture, I would point as an example to USAnians, many of whom will flat out refuse to understand anything in terms of another culture in which they demand to participate or visit (frequently in terms of economic exploitation), even in the simplest case of understanding history, let alone how perception or processing works in another society.

  39. cyberCMDR says

    I wonder how extensible this concept is. Instead of just colors, how does the language environment (even with the same language) affect how a person views the world? Can this be extended to whether a person “sees” a world with a maker, versus someone who analytically “sees” the world in terms of science? In other words, is the debate between science and religion as much a matter of wiring laid down during childhood as it is a debate about legitimate ideas? This could explain the ingrained resistance of some to abandoning their worldview, in spite of evidence to the contrary.

  40. bric says

    NB this is a small extract from an hour-long Horizon programme [http://youtu.be/5nSDJHAInpo], the point of which was to show that colours as ordinarily understood are mental constructs; it is hardly surprising that different human traditions construct them on different lines. Languages are pragmatic but flexible and deal with concepts their users need; Guy Deutscher (mentioned above) makes similar points about number words, he has an example from a tribe that has no complex numbers: a hunter has 50 arrows in his quiver, his language doesn’t provide a way of expressing ’50’, but he knows if any are missing because each arrow is an individual to him. If the tribe ever needs to count that high the language will undoubtedly provide a way.

  41. Teshi says

    This is a fascinating clip. I will have to watch the full program! I first heard about this on QI when they were talking about the colour linguistics of the Ancient Greeks who used to be thought to have different perception of colours but are more recently thought, like these Himba people, to simply have different concepts of colours. Sadly, I could only fine writing about this with the former argument rather than the latter (followed by tons of posts of people refuting it): http://serendip.brynmawr.edu/exchange/node/61

    Thinking about how much perception changes how we categorise colours in something like looking at a painting or a photograph, I don’t think this is much of a leap. Imagine you are painting from a photograph. It is advisable to isolate the colour of the sea and sky by putting a sheet of paper over the rest of the painting before you decide what colour paint you will use to most accurately represent it. This is partly due to comparison (comparing, say, one type of green that makes another look brown) as well as expected colours– we expect clear sky to be blue, grass to be green etc.

    I have often been surprised at how green the sky can appear, or how purple water can be, or how orange or red something green can appear.

    This counter-intuitive red/green issue makes me think of adjusting colours in film. When adjusting the colours of Hobbiton in LOTR: FOTR, the designers wanted Hobbiton to look really green and luscious. However, I remember hearing them describe how eventually they went for adding more red/brown to the green in order to make it look very, very green. Counter-intuitive, yes. But I have a pair of goldeny sunglasses that make everything look like Hobbiton in the movie– totally green (and yet I know how much more orange everything thing is than if I take them off). I can see how certain greens and reds could be associated, just as certain blues and greens are.

    When it comes to including tones as unifying factors rather than shades, all bets are off. That’s how the sea according to Ancient Greeks can be wine-dark and the sky bronze. These are levels of zingyness, rather than tones.

  42. bric says

    Teshi – yes the Horizon programme neatly demonstrates how our own perceptions of colour can be manipulated, even when we know what they ‘should’ be. I have trouble distinguishing green from blue especially when remembering an object, so I wasn’t surprised to learn that Classical Chinese used the same term for both (although modern Mandarin enables one to distinguish them the old combined word is still in use). The divisions of the spectrum are just a way of arbitrarily dividing up a continuous set of frequencies; perhaps it’s a bit like the way different cultures perceive musical scales.
    Wikipedia has a useful article on blue-green terms

  43. Samantha Vimes, Chalkboard Monitor says

    Does that tribe live in one of the arid areas, where light colors help reflect sunlight away from you during the day, but dark colors absorb the heat so you have something warm to lay on at night? And certain shades of greens and blues are seen only during rainy season or if one has traveled far enough to see big waters? Because if so, I can totally see the logic behind the color groupings.

  44. The Dancing Monk says

    While we’re on the subject of the brain’s perception of reality, this weeks Horizon deals with how our genetic & environmental conditions govern our response to good & evil. Apparently some of us are born with a predisposition towards anti social behaviour

  45. says

    My suspicion is that what is really happening here is a result of the difference between incoming data and perception, as flavored by expectations and training, cultural and otherwise.

    As anyone who has gone to art school knows (even those of us who did so back in the days before computers), most people don’t really pay much attention to color. We get trained as tots, in school. We are issued our crayons and given sheets from workbooks to color. THE APPLE IS RED. THE TREE IS GREEN. THE SKY IS BLUE. We color between the lines. We trace the letters: A-P-P-L-E, R-E-D. There is a “right” way and a “wrong” way. Never you mind that the apple on the teacher’s desk is a half a dozen different shades of red and brown and maroon. Never you mind how it reflects the piss yellow of the light from the ceiling fixtures and here, on the side, the blue of the sky from the window.

    Freshmen art students are often given an exercise in which they have to draw or paint something white, something like an egg or a sheet of crumpled paper. You stare at it until your eyes pop. Then you begin to SEE. Have you tweaked, somehow, the rods in your eyes so that they work better? It is unlikely. Seems to me what happens is you are finally paying careful attention. And after you do a couple of these kinds of exercises, you start noticing the delicacy of colors EVERYWHERE.

    I’ve been thinking about this a lot. Several years ago I was diagnosed with a rare form of retinitis pigmentosa, sector r.p., and patches of my retina have died. I don’t see so well in low light.

    I’ve become obsessed with photography since my diagnosis. I so treasure the light. (I have about 25 thousand photos up on flickr: http://www.flickr.com/photos/50728681@N06/sets/72157626254350181/ )

    My photography has really improved over the last two years. I used to take pictures of things. Then I took pictures of things in good light. Now, mostly, if I can, I take pictures of good light and if there are THINGS in the shot, well swell.

    I SEE so much more than I ever did before. I see more colors. But I don’t think it is really because of the loss of retina has substantially changed my vision. It kicks up the contrast is all. I think I see more nuances of light and color because I am constantly paying attention to light and color. These eyes are what I got, and so help me I’m going to make the most of them.

    I strongly suspect that the differences in the colors that are seen by people in different cultures have everything to do with culture, with early childhood training and a lifetime of expecting to see a certain way. I could be wrong. In any event, it is all terribly interesting.

  46. Ben says

    @49 — my partner and I are both graphic designers but even we have different ideas of what is classed as blue and what is classed as green. Our car is a particular hue that I call green, but my partner insists that it is blue. I have had clients who have asked for “green” and when I’ve shown them a design they said “no, i asked for green, not blue”. My partner tells me he’s had the opposite happen.

  47. MadScientist says

    I agree the conclusions are lacking. If people are simply describing a different region of the CIE color chart using a different name, then so what? That would simply demonstrate that different groups decided to group their colors differently.

  48. pseudonymoniae says

    Cool study. I would guess there must be some top-down circuit which allows linguistic cues to categorize inputs to the cells which make up our color representations in visual cortex. On the other hand, a more parsimonious explanation is just that the Himba haven’t been exposed to the same range of colors as us, which means that their brain circuitry never developed the ability to differentiate between two similar shades of blue, for example. Either way, it is interesting to consider that we all could have slight variations in the colors that we see, as small differences in how our visual systems develop might cause the cells in one person’s brain which correspond to a specific color like “red” to receive inputs from cells corresponding to a slightly different range of wavelengths than those of the next person. If this were the case, then presumably these differences would be so small as to be hardly noticable, a few um difference here or there probably would just come out as noise.

    On this other hand, I think this is a good point here made by Davric @33:

    In a room of people who, basically, all looked the same to me, they saw about four or five distinctly separate hair colours and complexions. … They stopped me apologetically and one of them said, “We’re really sorry, but all you Europeans look the same to us.”

    If color perception can be so easily modified by early life experience, then isn’t it possible that higher level representations of things like facial structure might be affected as well? Why shouldn’t people of different ethnic groups, at least those who were largely raised apart, have less ability to discriminate between facial features that are peculiar to another ethnic group? I don’t know whether or not this has been studied, but it would be a great social science topic. Too bad I’m in the wrong field :P

    Something else I noticed which was kind of weird. When they showed the green squares, the BBC version looked like it was all the same color. But in the version that the Himba man was looking at it was immediately apparent to me which square was a lighter shade of green. Did anyone else notice this? I suspect the BBC people just put a bunch of squares up that were all the same colour, whereas the researchers used one square which was off by 20 or 30 um.

  49. says

    My spouse and I often argue about what is green and blue in the turquoise region… I’ll say it’s green and she says it’s blue. And then in the darker regions we’re reversed.

    I’m pretty sure it’s not just language, but personal experience as people learn how to label and interact with the world.

  50. bric says

    @53 – hmmm yes, my partner is Chinese (and formerly a fashion designer) and never tires of pointing out that I get colours ‘wrong’. The mention of Eleanor Irwin’s work above reminded me that I had noticed that for him, when working with fabrics there was some connection between colour and texture that always eluded me.

    More on Greek colour perception http://serendip.brynmawr.edu/exchange/node/61
    the comments are more perceptive than the article imho

  51. Teshi says

    [quote]Something else I noticed which was kind of weird. When they showed the green squares, the BBC version looked like it was all the same color. But in the version that the Himba man was looking at it was immediately apparent to me which square was a lighter shade of green. Did anyone else notice this? I suspect the BBC people just put a bunch of squares up that were all the same colour, whereas the researchers used one square which was off by 20 or 30 nm.[/quote]

    I noticed this too, but I think it was to do with the screen the people were looking at. I found it easy to tell where the different one was in the shot that included both the subject (which I think was the woman at that point) as the screen because of the angle I was looking at the screen. I think straight on it would have been more difficult.

    However, my skepticism about this was a little bit mollified when they showed the blue/green test. It was extremely obvious to me which was blue. Even if I had been able to pick out the different colour facing straight on the screen it would have undoubtedly taken me longer, even if a second longer, than the blue/green test.

  52. Svlad Cjelli says

    The obvious example with japanese blues and greens is averted with younger generations, who do mind the difference.

    Northern Europe also had a lot of this before, especially with blues and greys I think.

    The rainbow’s colours have also been grouped differently in the past. Even Newton noted five colours.

    On a barely related note, old people in Sweden may say fireyellow instead of orange.

  53. Inflection says

    Now I would like to be able to see the world like the Himba do.

    There was a guy who wrote an app that would adjust color perception for the colorblind, and I think it could do the reverse so that a colorsighted person could get a sense of how an image would look to someone with various particular forms of colorblindness. I wonder if that app could adjust hues in a camera image — maybe just by measuring experimental reaction times for Westerners and Himba — to show us the world as the Himba see it.

  54. Tsu Dho Nimh says

    @33 … When you have to use colors to break down a large group that is important to you, you make distinctions.

    Getting my driver’s license in Mexico was similar to your Turkish experience. There were several hair shades of what the USA would call “brunette”, ranging from crow-black, which is shiny blue-black, through dark brown, selections for reddish brown shades

    Me? I was “guero/guera” which is all shades of blond, reddish blond, strawberry blond, diswater, ash, etc, etc. And the eyes? They became “azul” which covers all shades of blue and grey. There are enough green eyed Mexicans that they got “zarco”.

    Skin color was much the same. Seven shades of brown and then “library paste white”.

  55. Anthony Bradley says

    @Neil Rickert: you beat me to it: The SWH implies that language itself affects thought (or at least cognition). No wonder it’s controversial.

  56. tonyfleming says

    If the Himba see blues and greens as the same color, would a green page with blue text appear blank to them? Could you write “secret” messages to another non-Himba by using blue ink on green paper?

    How about subliminal messages, a la “They Live”?

  57. Svlad Cjelli says

    @68: Not any more than you yourself can’t tell the difference between two shades of blue.

  58. Svlad Cjelli says

    Ever been to a BBS where the only spoiler function is to change the font colour to match the background? It will usually end up a similar shade because nobody cares enough to get it just right.

    It can be tough to read without highlighting the text, but in most cases you’ll instantly spot that there’s text.

    Now, if you wanted to examine something interesting, that should be tried with a group of humans who haven’t based their whole lives around spotting texts.

  59. Arazin says

    Wouldn’t this be expected? Growing up in the african savannah I would expect survival would depend on being able to distinguish between shades of the same colour than the stark contrast of different colours. It would be more beneficial to be able tell the difference between the yellow hair of lions and the yellow grass of the savannah.

    The mechanics of how the brain achieves this is very interesting though……….

  60. Neilp says

    This also works with things like direction. We use left/right, but many languages don’t. Some use NSEW, for example. Take an Aboriginal man from Australia, put him in the middle of central park, and he will instantly know which way is which.

  61. says

    “The divisions of the spectrum are just a way of “arbitrarily dividing up a continuous set of frequencies;”

    Not exactly, we have overlapping but not contiguous sensitivity in our cones. They do have peak wavelengths, the blue and green have the least overlap, the red and green the most.
    The RP inflicted person above: I’m so happy you’ve made the most of your disease. ‘Training’ ones eyes and ears (the brain, of course) to adapt is really cool.

    I was a sound engineer for many years and I have a bit of hearing loss, but I can still hear details in sounds that most people can’t.
    I was also wondering about the differences in UV exposure from Europeans v. equatorial peoples. The cornea and the lens are both affected. The cornea is primarily affected by UV-B and the lens by UV-A.

  62. rrhain says

    They already did do the experiment and found that people who speak different languages see color in the same way despite the differences in color terms, at least to an extent.

    First, what is a “color term”? It’s a word that describes a color that isn’t related to another object. For example, “blue” is a pure color term in English but “turquoise” is not as it is a reference to the stone.

    Various languages have different numbers of color terms. Some have only two: What we would call “black” and “white.” This doesn’t mean people who speak such a language don’t see other colors. Instead, they use terms that refer to objects of the color they mean, not a pure color term. They might say something is the color of unripe bananas as there is no pure color term, “green.”

    Interestingly, there is a pattern to the acquisition of color terms. When a language has three, the third term is always “red.” Next comes either “yellow” or either what we would call “blue”or “green.” Next, if you had “yellow,” you then get “blue/green,” then “blue” and “green” split.

    Now, how do I know that we would call it “black” as opposed to “dark”? Because they tested speakers of such languages to find out what they meant. They gave the speaker a color-chip board with a rainbow of colors and asked them to pick the most iconic example of the color term and in all cases of languages with only two color terms, they chose the chips that speakers of English choose to represent “black” and “white.” Similarly, three-color speakers always choose a chip that we would call “red,” even though that term would be used to describe other colors we wouldn’t call “red.”

    So these five-term speakers have terms that cover colors we separate but if you were to show them the color-chip board, they would choose, black, white, red, yellow, and either blue or green.

    This doesn’t mean that what they’re finding isn’t real. It’s just that the Sapir-Whorf Hypothesis isn’t strongly confirmed by this. Color perception might have some influence from language, but bilingual speakers don’t see the sky as different colors based on what language they speak.

  63. Daniel says

    The evidence seems shaky at best to me: The first test took no account for genes also wiring for language acquisition that kicks in and that will shape perception as well. Nor did they show brain activity with fMRI for instance. It would be a lot more convincing if they’d look at how people learnig their first and different language would look like in the wiring of the brain, so look at when a baby in germany starts to learn german and look at his/her brain and do the same for different languages and see if there are any interesting developments in differences. The second I immediately thought that they were colourblind– that seemed to explain the categories, and the ability to spot contrast differences (remember that during the second world war they hired colourblind people to spot camouflage from planes).

    And we’re talking about colour! You could sit for eternity slicing up the spectrum to their own categories and it’s completely arbitrary, a human construct, built in through the millenias. Putting colours to categories like darker ones etc seems like a perfectly sensible thing to do. But did they control for genes? That is, were they colourblind? (We’re talking about a small tribe with probably not much genes coming from the outside)

    Linguistic Determinism/relativism suffers from deepities: The analogue to them and people abusing the observer effect and other concepts from quantum physics is amusing.

  64. John Kingston says

    There is a vast literature on this issue, which shows that there are both universal and language-specific aspects of color perception. The quickest way to find it is to go to Paul Kay’s website:


    He’s done or participated in most of the reliable work.

  65. John Kingston says

    P.S. Also check out Lera Boroditsky’s paper:

    Winawer, J., Witthoft, N., Frank, M., Wu, L., Wade, A., and Boroditsky, L. (2007). Russian blues reveal effects of language on color discrimination. Proceedings of the National Academy of Sciences doi:10.1073/pnas.0701644104

    for evidence of quite subtle language-specific effects on color discrimination.

  66. says

    This put up (Wiring the brain | Pharyngula) was a very good read so I posted it on my Fb to hopefully provide you with more readers. Perhaps you may review The Lion King in 3D on one of your subsequent posts. I hear they have worked over 12 months on the 3D aspect of it. Simply an thought for you