Scientists have found a woman whose eyes have a new type of color receptor | Hacker News Reader

Scientists have found a woman whose eyes have a new type of color receptor

sciencealert.com

48 pointswolfgke53 comments

I was just reading the Wikipedia article on Tetrachromacy and saw this interesting fact.

"Humans cannot see ultraviolet light directly because the lens of the eye blocks most light in the wavelength range of 300–400 nm; shorter wavelengths are blocked by the cornea.[27] The photoreceptor cells of the retina are sensitive to near ultraviolet light and people lacking a lens (a condition known as aphakia) see near ultraviolet light (down to 300 nm) as whitish blue, or for some wavelengths, whitish violet, probably because all three types of cones are roughly equally sensitive to ultraviolet light, but blue cones a bit more."

https://en.wikipedia.org/wiki/Tetrachromacy

Hmm, I see those "blacklight" CFL ultraviolet lamps as violet-white. I see the sameish color on ultraviolet lasers. Am I the only one who sees it that way? Can anyone describe how they perceive those lamps?

Communicating color is extremely difficult!

> Communicating color is extremely difficult!

Odd personal anecdote: In first grade art class the teacher held up a piece of art paper and said "This is red". Being the curious student I was, I raised my hand and asked "how do you know?" I was looking for something that would answer both "how do you know I perceive red the same way you do" and also "what defines it to be red". Something about wavelengths and color spectrum etc. would have been a great answer.

I was sent to the principal's office as a troublemaker. Never did like art class after that.

Did anybody happen to see anything in the article about what wavelength the extra cone type was actually (most) sensitive to???

The article hyped the "millions of colors" aspect, but nothing about whether it was deeper reds, higher blues, or just more precise yellows (or whatever).

cDa29 is a cone that is a deuteranomaly, which is a kind of altered sensitivity to green light. So, presumably she has better discrimination of yellows/greens/reds, and not blues.

See this 2010 paper for details on this person: http://vision.psychol.cam.ac.uk/jdmollon/papers/JordanDeebBo...

Ah, thanks. So further improved primate detection of ripe fruit, then.

There's also the story of Concetta Antico, who is also supposedly a tetrachromat and a painter.She claims to add hues to her paintings that she sees but we do not. Researchers are looking into her abilities.

https://en.wikipedia.org/wiki/Concetta_Antico

http://www.huffingtonpost.com/maureen-seaberg/worlds-first-t...

https://concettaantico.com/oil-paintings/

(watching her paintings through what is likely an LCD screen that isn't even wide gamut is a bit ironic, of course)

Why would only women be capable of tetrachromacy?

I am a man, but if the test is made public, I would be interested in testing myself. I am not color blind according to color blindness tests, but there have been instances where I disagreed on the color something was with others and while I could consistently use the name that they used for it, I felt it really ought to have been named something else. One involves "green" status LEDs on electronics, which seem more in common with yellow than green to me. They look different from yellow, but they seem closer to yellow than green to me and for lack of a better word, I would rather call them yellow. If I could use a different name for the color, mapping it to green when interacting with others would seem less weird. Another involved the paint of a house when I was young, whose color seemed to poorly resemble what my parents said it was when it was dark. The house has since been repainted. Having three primary colors never made much sense to me either, but that could just be my own ignorance.

Anyway, I am sure that there are many more variations in human vision than we realize. The dress incident brought that to light:

https://en.m.wikipedia.org/wiki/The_dress_(viral_phenomenon)

It would be nice to have ways of categorizing them.

Only women are capable of tetrachromacy because the gene that controls which wavelengths of light are detected by cones are on the X chromosome. Tetrachromat women have a mutation on one of their X chromosomes, and are heterozygous for "normal" cones (detect one wavelength) and "abnormal" ones (detect another wavelength). As it turns out, this means that some of their cones detect one wavelength of light, and others detect the other.

As men can not be heterozygous for traits on their X chromosomes, they can't express two cones simultaneously. They CAN have the mutation, but that means they see three colors that are somewhat different from the three colors that other men see, not that they see four colors.

All of the above as I understand it from a popular article on tetrachromacy. I am not a biologist.

The alleles being sex linked makes sense, although there are a couple of theoretical exceptions. One is the very rare male that has two X chromosomes because the SR-Y gene was transposed onto the X chromosome. The other would be having the reverse where the second gene is transposed onto the Y chromosome. The former are sterile while I would guess that the latter are not sterile, but unless they really do exist, are found and one that is found has the same combination as a tetrachromatic woman, there is no way to confirm that.

Genetics was one of my strong points when I was in school. I would have gone into bioinformatics if it were not for my poor performance in organic chemistry, which I felt crippled my ability to do the sort of work that I had wanted to do.

You don't need to be good at orgo to do informatics; I've worked with a number of very capable analysts who hadn't been anywhere near a lab since their undergrad days. (If you want to do lab work, things are different, but my experience is that it's kind of rare for informatic analysis and lab work to overlap.)

Another possibility for male tetrachromats would be people with the 47,XXY karyotype, who are phenotypically male but have the paired X chromosomes necessary to make tetrachromacy possible. Given the rarity of both variations, I can't imagine it being likely that any such person has been identified, but based on my (admittedly meager) knowledge of the subject, I don't see a theoretical barrier to such a person possibly existing.

While I call it bioinformatics because that is a handy umbrella term for biology and computers, what I really wanted to do was to figure out how to do in-Vivo gene splicing to correct genetic illnesses. Specifically, I wanted to build a compiler for generating gene sequences that could be synthesized and injected into bacteria to produce biological machines tailored to alter specific regions of a person's genome. You could call it a compiler for biological instances of sed commands on DNA. It would need to construct things that bypass with the immune system and two membranes with varying methods of entry and varying sized things permitted entry.

Others are making progress in that field, so I am hopeful that some day it will be a reality. The idea that molecular dynamics simulations are NP-complete is going to be a pain for anyone attempting to do that unless they are really clever in figuring out how to do construction like people were when they made the perfect ping. That should require Linus Pauling level or greater knowledge of chemistry, which I do not have. Building it would also produce an easy construction technique for a incredible number of biological weapons, so if it is never made, maybe it is for the best. Not to mention it would probably sow the seeds of the eugenics wars.

You are right about XXY. I had forgotten about it. I had no idea how to theoretically delete an entire chromosome without deleting all copies of it when I was studying with the goal of producing that tool, so it was easy for me to forget. Now that I think about it, it might be possible. Just append the entire chromosome in pieces to the end of another with the genes disabled and some trivial difference to enable final slicing into discrete chromsomes, delete all of the copies of the old trisomal/monosomal chromosomes and do final slicing and splicing to produce two new functioning chromosomes in what would hopefully be a large percentage of the cells while managing not to kill the organism in the process. It would need to be done when the organism is young to minimize negative impacts to its development and would be somewhat different when dealing with sex chromosomes. You would also need to be careful that genes that are needed for the procedure are not disabled by it (like not unlinking libc when trying to replace RHEL with Gentoo in in-place when using a script that calls commands that link to libc). If you manage to rewrite a sufficiently large enough percentage of the organism without killing it, that procedure should cure trisomy and monosomy conditions if done early enough in development. Doing it requires tools that we do not have though.

Anyway, that should make it clear why being bad at organic chemistry kept me from thinking that I could pursue this.

> the very rare male that has two X chromosomes because the SR-Y gene was transposed

Isn't Klinefelter's Syndrome (XXY) more common than that?

I had forgotten about klinefelter's syndrome. You are correct.

> Having three primary colors never made much sense to me either, but that could just be my own ignorance.

Tree primary colors thing seems to be mostly ignorance of the educators teaching that stuff to children. There are no absolute "primary colors" - basically, you can pick any colors at random and use them as a base for linear combinations; the set of colors you can express as such linear combinations form a color space defined by those base colors you picked. Red, green and yellow are a common historical picks, RGB is more common now because that's what is put on the computer screen, but CMY form a color space too.

C.f.: https://en.wikipedia.org/wiki/Primary_color#/media/File:CIE1... - the triangle inside is the color (sub)space of RGB overlayed on CIE 1931 color space (https://en.wikipedia.org/wiki/CIE_1931_color_space).

Source: I educated myself when I got asked by a local drunk to resolve a bet between him and (presumably) another local drunk while I was shopping in a local grocery store. Those two got into a bet over whether RGB or RGY are primary colors.

I always understood it as RGB being additive -- each color directly stimulates a given color cone in the eye. Whereas process colors (CMYK) are subtractive, because they are a tint. White light hits the paper, but each color (Cyan, Magenta, or Yellow) takes out a certain RGB color leaving the remaining two colors to reflect off the paper. For example, Cyan is a filter for Red, leaving Green and Blue to reflect.

I'd say it's a property of the medium, not of the colors. Computer screens blend additively because you emit a mix of light waves; pigments blend substractively because they are filtering out wavelengths from incoming white light. Color spaces and "primary colors" are orthogonal to that.

RBY is what is commonly taught as primary bases in school, due to color compositions and the ease of expressing it through mixing pigments.

R + B = Purple R + Y = Orange B + Y = Green

Right, I probably meant RBY not RGY; I wrote from memory, and meant the one most commonly taught in primary schools in Europe.

So who won, based on your exposition? Or was it a draw?

I left that for the two to settle between themselves. I just explained the thing to the guy that asked me first. He was on the RGB side, having shown me a picture similar to this one[0] on his mobile phone. My guess is that he ended up claiming victory, because when I tried to explain the details of the problem, he was going all "so this means RGB can be considered primary colors, right?". I never met the other guy.

[0] - https://en.wikipedia.org/wiki/RGB_color_model#/media/File:Ad...

http://news.nationalgeographic.com/news/2012/09/120907-men-w...

Men and women seeing slightly differently is in line with our biology. Just how it is, really. Men are more prone to color blindness too.

https://nei.nih.gov/health/color_blindness/facts_about

That explains the LED thing. Thanks.

>One involves "green" status LEDs on electronics, which seem more in common with yellow than green to me.

Almost all "green" LEDs emit at a wavelength of 565nm, which is almost equidistant [1] between green and yellow. Most "green" LEDs are thus green solely by convention, and seeing them as greenish yellow (or yellowish green) isn't unusual at all, because that's the color they actually emit. You can find green LEDs that actually are green, but they're relatively uncommon, and tend to be costly unless you order in bulk direct from a manufacturer.

[1] http://ledmuseum.candlepower.us/35/m-565re2.gif

Simplified version:

X chromosome can express 1 or 2 color receptors, Y only expresses 1.

Women (XX) can get up to 4, but the X occasionally only produces 1, since women have two X they mostly have 3, sometimes 4 and very very rarely only get 2 (both X only express only one)

Men (XY) can get 3 but when the X chromosome only gives 1 they only have 2 and are color blind. Depending on which one they didn't get, that determines their color blindness.

[1] http://www.radiolab.org/story/211193-perfect-yellow/

https://en.wikipedia.org/wiki/XY_sex-determination_system is a useful start for anyone not familiar with the very strong link between some genes and sex determination.

The next step from there is that in humans the X chromosome carries more genes than the Y chromosome. Since women have two X chromosomes, they can express multiple variations of genes located there. Men will only possess one variation.

This is the reason that color blindness is so much more common in men, some of the genes related to color vision are carried on the X chromosome so a single error in one of the pigment genes is sufficient. Women are only color blind if both copies are defective.

I see some things that people identify as orange as more yellow.

I am certain is has to do with all cones not being equal. You may have a cone that is more sensitive to a wider or narrow range; or one that has a peak sensitivity to a slightly different wavelength than the norm; or both.

it could be a recessive trait carried on the X chromosome, and women have two copies of X, whereas men have an X and a Y, so could never satisfy the two copies criterion for recessive traits.

I thought the discovery of individual tetrachromats had been reported several times in recent years. This article seems to suggest that women who were anatomically capable of tetrachromatic vision were found before, but that none of them were successfully confirmed to be tetrachromats in practice until this new study. Is this really the case?

yes. http://www.digitaljournal.com/article/326976 http://jov.arvojournals.org/article.aspx?articleid=2191517

Hmmm, that shows that this research finding was published in 2010, which might be the timeframe I was thinking of. I first thought from the link that this article was newly published.

yeah, me too.

Reminds me of this bit from BSG (a conversation between two "artificial" humans):

    Cavil: In all your travels, have you ever seen a star supernova?

    Ellen: No.

    Cavil: No. Well, I have. I saw a star explode and send out the building blocks
    of the universe, other stars, other planets, and eventually other life, a
    supernova, creation itself. I was there. I wanted to see it, and be part of the
    moment. And you know how I perceived one of the most glorious events in the
    universe? With these ridiculous gelatinous orbs in my skull. With eyes designed
    to perceive only a tiny fraction of the EM spectrum, with ears designed only to
    hear vibrations in the air.

    Ellen: The five of us designed you to be as human as possible.

    Cavil: I don’t want to be human. I want to see gamma rays, I want to hear
    X-rays, and I want to smell dark matter. Do you see the absurdity of what I am?
    I can’t even express these things properly, because I have to — I have to
    conceptualize complex ideas in this stupid, limiting spoken language, but I
    know I want to reach out with something other than these prehensile paws, and
    feel the solar wind of a supernova flowing over me. I’m a machine, and I can
    know much more, I could experience so much more, but I’m trapped in this absurd
    body. And why? Because my five creators thought that God wanted it that way.

It's very exciting for me to dream of a future where we manage to tap into the visual cortex and maybe then we'll be able to see as if we had 4 or maybe more cones.

I really hope this result is verified and confirmed. It's very interesting to think what would it mean to see like this. Could we liken it to jumping from 8 bit color to 32 bit color for example?

Nope, the trichromatic vision is the reason we use color spaces like RGB -- it means that combinations of three colors can produce nearly all of the subjective hues that we experience. The idea of "three primary colors" comes from trichromatic vision; there's nothing about physics that implies that there should only be three primary colors, and machines can be built to distinguish an unlimited range of hue combinations that are not regarded as built up from any finite number of primary colors, whereas human vision simply can't make these distinctions.

Tetrachromatic vision means a four-dimensional color space which will have to use four, not three, primary colors in order to represent all subjective hues through combinations. So you would need to add another channel on top of RGB.

The bit depth in a color space is effectively about how many different intensities of each primary color can be distinguished (although you can have a system that measures or distinguishes colors in terms of other features instead). So if you have a greater color depth, you can make more fine-grained distinctions about the intensive of the primary colors, but having a new primary color (that reaches otherwise-unrepresentable colors!) is something else entirely.

Forgive my ignorance, please... I don't understand this color space thing, much less a 4D color space.

Here's what I do know:

All light has some wavelength. That wavelength determines the color. Red light has a longer wavelength than violet light. Infrared is longer still. Ultraviolet has a more narrow wavelength than violet.

    ROYGBIV

So, I think of color as a number line... But it wraps around. So:

     O Y  
    R   G 
     VIB  

    R + Y = 0
    Y + B = G
    ...etc

Why would a tetrachromatic person see a color sphere instead of a larger color wheel (more than 7 letters on it)?

Color isn't a physical property. Normal color vision is the sum of three signals. Tetrachromacy is equivalent to having a fourth signal, so you need an extra axis. Color spaces are weird in that cones have overlapping ranges (so the signals are not independent), and there's an added linguistic/perceptual step in the brain.

I don't buy it.

The different types of cones are all sensing the same physical property: wavelength.

Each type of clone is more sensitive to some specific range of wavelengths (detail: the graphs of sensitivity vs wavelength for each type are approximately but not exactly bell curves).[0]

    TRICHROMATIC
    ------------
    
        L  M  S    <--long, medium, short cone cells
    R O Y G B I V 
    6 5 5 5 4 4 4   <--wavelength (read down)
    5 9 7 1 7 4 0
    0 0 0 0 5 5 0

So your visual system does the math to determine the scalar value of the color that is being perceived. If M cones are firing a lot and L and S cones only a little, it's a blue green. If L fires a little and M less so and S hardly at all, it's a red. Etc.

Adding another type doesn't change the nature (dimensionality) of color, it merely changes two things: 1. the math you use to determine that scalar. and 2. your ability to distinguish between 650nm light and 651nm light (perhaps).

Perhaps this will make it clear: imagine one million different types of cone cells, each with some distinct sensitivity range. That doesn't give you a million-dimension space. It would certainly give you the ability to tell the difference between 510nm light and 510.05nm light, though!

0: https://en.wikipedia.org/wiki/Cone_cell

I think you're mistaken there simply because most colors are not a single wavelength. If colors were a single wavelength, then indeed something around 2 cones should already be sufficient to resolve any color in the range of their joint sensitivity, because there is enough information to deduce which single wavelength was responsible for their activation levels. But because we see things that are combinations of various wavelengths, adding more cones with different sensitivity always provides useful information for resolving which wavelengths are present at the same time at which intensity.

I know that color spaces are a thing (I just don't yet understand WHY), and surely tetrachromatic people can distinguish more colors. So I'm sorta playing devil's advocate... I'm eager to have an 'aha!' moment over here.

Does not this:

    Blue + Yellow = Green

simply mean this?

    (475nm + 570nm) / 2 = 523nm (the average)

Supposing that wavelength was the only property of light, yes, two ideal cones could determine exactly where on the spectrum some light belongs. However, there's intensity to be measured, too. Variable intensity means that if you just got two scalar values from two sensors... you haven't narrowed it down to a specific wavelength, because those same two values would arise for states. ..Right?

The blue + yellow = green is in subtractive mixing so it doesn't directly apply here, but taking red + green = yellow from additive mixing, there are two different things that could happen:

* you can have a red light and a green light both shining together (like on your computer screen)

* you can have a pure yellow light (like from a yellow laser, or the yellow part of a rainbow)

These are physically different, as could be detected by a spectroscope. But for a normal trichromat human, perceiving either produces the same color sensation and we can't distinguish them using our eyes (because the excitation of the cones produced by either is roughly equivalent). (I should add that there can be still other combinations of wavelengths which, when they occur together at the same time, we would perceive as the same yellow.)

If you add more cones, you could hope to distinguish these physically different phenomena, and have them somehow produce different sensations. However, we also apparently can't describe what those sensations would be like...

https://en.wikipedia.org/wiki/Knowledge_argument#Thought_exp...

Aha!

> Nope, the trichromatic vision is the reason we use color spaces like RGB -- it means that combinations of three colors can produce nearly all of the subjective hues that we experience.

There are lots of RGB color spaces though, and the widespread sRGB colorspace only supports a fairly small section of the human subjective color space, it's especially poor in the greens and blues

What about CMYK for tetrachromatic?

I think the primary difference between CMYK and RGB is that CMYK is a subtractive color space, while RGB is cumulative (I don't know if these are the correct terms in English). The first is used for reflective surfaces, while the other is used for light sources.

On reflexive surfaces, you make colors by subtracting wave lengths (the paint has pigments that absorb specific wavelengths). On light sources, like our computer screens, we make color by combining primary wavelengths.

CMYK is three colors. K is just black.

You could analogously call the RGB color space RGBW, but it wouldn't mean you were suddenly working with four colors.

It would still be RGB with an extra. Like RGB+X

More like 8 to 16 bit. Tetrachromats see 100x more color. 16 bit has 256x more colors than 8 bit.

This is not news. This article is based on another article from 2012 (http://discovermagazine.com/2012/jul-aug/06-humans-with-supe...).

And actually the whole story is based on a paper from 2010: http://vision.psychol.cam.ac.uk/jdmollon/papers/JordanDeebBo...

Reply on news.ycombinator.com