29

Our eyes contain 3 photoreceptor cells (cones) to perceive three wavelength ranges of light. Here is a visual representation of the wavelengths by these receptors (S, M and L).

Wavelength per photoreceptor

So if we have light of 440 nm, it results in the color blue. If we have light of 540 nm, it results in the color green. If we see light of 650 nm, it results in the color red.

I think I understand our ability of the brain to mix the results of the signals of these receptors, producing colors like yellow. However, what I do not understand is how the color spectrum is displayed like this:

Visible spectrum representation by wavelength

Given that spectrum, I would suggest that the color 'blue' is actually a mix between the receptors S and M. And the pure result of activating the S-receptor would result in the color 'purple' (I would describe the color in the left of the image as purple, right). Therefore, the receptor colors should instead be RGP (red, green, purple) instead of RGB (red, green, blue).

However, there is one problem with this which I cannot explain. How come that mixing red light with blue light also results in purple light? How is it possible that purple light can be achieved through mixing (additively) blue and red light, just as going to the shortest wavelength boundary of what we can see (from blue to ultraviolet via purple)?

So the actual problem here is:

  1. Purple is the color at the very shortest wavelength we can see.
  2. Purple is an additive mix between what we see as red light and blue light.

That just doesn't make any sense. I don't see how our brain can possibly perceive this as being the same color. Shouldn't both purple colors actually be different colors (thus we would have a new different color for that)?

UPDATE: Pure activation of the S-cone (blue) may actually give violet instead of blue. That means that the color blue contains also a bit activation from the M- and L-cones (green and red). So then the "blue-est blue" is actually violet, not blue. This confusion originates from the RGB-system (and other sources) that assume blue as being a primary color, whereas violet would perhaps be more accurate! Further details can be found in the accepted answer.

The line of purples may further help to understand what purple/magenta/violet really is.

Yeti
  • 581
  • Way back when I had elementary school art class, all the teachers made a big fuss about how purple was not the same as violet. In any event, I'm not sure physics is the right place to ask about color perception as such. This is probably better for Biology or Cognitive Science. –  Jul 01 '14 at 10:33
  • It does not have much to do with cognitive science, but perhaps biology would fit better. Do I need to copy&paste the question, or can I transfer it another way? – Yeti Jul 01 '14 at 10:35
  • If you want to move it, use a moderator flag (a custom one probably works best) - they can migrate the question. That said, I make no guarantees about how questions are received on sites I don't frequent. There's also the possibility others here will disagree with my judgment about physics suitability - I don't want to be accused of single-handedly bullying new users :) –  Jul 01 '14 at 10:40
  • 1
    Related: http://physics.stackexchange.com/q/40763/2451 , http://physics.stackexchange.com/q/103448/2451 and links therein. – Qmechanic Jul 01 '14 at 11:19

7 Answers7

17

I have a strong reason to believe I have found the correct answer to my own question, you may correct me if I'm wrong. But this image seems to explain everything about my question in one single hit:

Graph derived by Bowmaker & Dartnall (1980)

These are results from Bowmaker & Dartnall (1980). Relevant reference: Bowmaker, J.K., & Dartnall, H.J.A. Visual pigments of rods and cones in a human retina. Journal of Physiology, 298, 1980, 501-511.

It seems that the L-receptor is actually more active at the very shortest end of wavelengths than it is for just longer than what we can see as visible light. You can see curve of red going up towards the short end of the wavelength axis. The L-receptor (associated with red) activation is not a bell-curve over the linear wavelength axis (as one would expect). That would explain the little bit purple-ish blue we see at 400 nm!

So luckily the brain is not freaking out, but the receptors are just a bit strange, probably with the goal to distinguish blue from more blue (from a functional view of 'evolution').

Note that it is logical that this is not the case on the right (longer wavelength) side of the graph, because there red is accompanied by green closely. Thus we can distinguish red from redder by the mixture of green.

Ruslan
  • 28,862
Yeti
  • 581
  • 2
    That's part of it, but as other comments have said, photopic perception is not the same as wavelength. There are colors we perceive (magenta is the classic example) which cannot be produced with any single wavelength. – Carl Witthoft Jul 01 '14 at 11:33
  • 1
    Of course I do understand the concept of color, naturally color only exists in our mind which is the reason we cannot imagine a new color. But still, color is based on the activation of the cones, which react to certain ranges of wavelengths in different strengths.

    Lets consider purple, magenta and violet as the same thing. We can see purple as one wavelength: 400 nm. And we can see purple as a mixture of e.g. 420 nm and 640 nm. And this is what I thought was strange, however, the graph derived by Bowmaker & Dartnall (1980) gives a perfect explanation of this phenomenon.

     – Yeti Jul 01 '14 at 12:03
  • 2
    I think the issue is you are approaching it backwards. We cannot think of purple as one wavelength. The function going from spectra (one or more color frequencies) to perceived color is a one way function that loses information. If we limit ourselves to single wavelength sources, we can make a claim that there is a mapping between wavelength and color, but that mapping gets thrown right out the window when we consider sources with multiple wavelengths. Thus we can think of 420nm+640nm as purple, but we cannot think of purple as 420nm+640nm – Cort Ammon Oct 03 '15 at 21:31
  • 1
    Also worth noting is that there is believed to be another color scheme in use in the mind. It is believed that the rod and cone innervation are eventually converted into three antagonistic color pairs: black-white, red-green, and blue-yellow. This is why a bunch of red and green horizontal lines shimmer when you look at them... the antagonism is hard to resolve. Its also why we have "red-yellow" as a color, but no "yellow-blue." Yellow-blue is a color we cannot see. (some argue there's some clever tricks with exhaustion you can use to see it, but its not well agreed upon) – Cort Ammon Oct 03 '15 at 21:33
  • @CortAmmon: but, we can see red-green (it's just brown or yellow) and black-white (it's gray) so why is there no "yellow-blue"? – sumelic Jan 10 '16 at 04:03
  • @sumelic We cannot see red-green. As you say, we perceive such a mixture of photons as a brown or a yellow, rather than perceiving them as red-green. Same for grey. We have been taught in science that you can mix the colors in that way, but intuitively we do not find people think of grey as black-white. They may think of it as "something inbetween black and white," but never black-white. Scientific studies have looked deep into this, which is why we have the current theory of perceived color which is built from 3 antagonistic pairs. – Cort Ammon Jan 10 '16 at 04:30
  • The technical name is "Opponent processes," and it's pretty well explained in the Wikipedia article https://en.wikipedia.org/wiki/Opponent_process . The article also points out some reasons why the theory is popular, such as it does a better job of predicting the results of after images, such as why staring at a red square produces a green square when you look away. – Cort Ammon Jan 10 '16 at 04:31
  • The graphs show (presumably) the (S,M,L) receptor response to a monochromatic input. How can one compute the (S,M,L) response to a mixture of frequencies at different powers? Is the response roughly linear? – copper.hat Jul 30 '16 at 19:50
  • No. It's nothing to do with the little upwards tail at the left end of the red curve on that graph. You'd still be able to get "purple light" if that tail didn't exist. The point is that there are colours between "red" and "violet" that don't correspond to any kind of monochromatic light; that is, you can only get these colours by mixing. – Dawood ibn Kareem Jul 31 '16 at 09:27
  • If your theory is correct, we should see purple when we mix blue and green as well, because the M-receptor has the same behaviour as the L-receptor at the left end of the graph. – timur Aug 07 '17 at 03:28
16

So the actual problem here is:

Purple is the color at the very shortest wavelength we can see.

Purple is an additive mix between what we see as red light and blue light.

That just doesn't make any sense. I don't see how our brain can possibly perceive this as being the same color. Shouldn't both purple colors actually be different colors (thus we would have a new different color for that)?

Your problem arises because you are mixing two different frameworks of reference for the word color, the biological, and the physical.

The physical colors of the spectrum have one to one correspondence with the specific wavelength given in your picture. If you have a beam of pure 450 nanometers and another of 700 nanometers and throw them on the same screen, your eye will see purple, because your eye is in the biological framework. A spectrum analyzer in the place of the screen will see the correct percentage of 700 and 450 wavelengths, because a spectrum analyzer sees the physical quantities. Adding light with two wavelengths does not create a new wavelength, the way that adding apples and oranges does not make pineapples:).

In short the color of the spectrum comes from the underlying physical framework, but human perception generates colors that are combinations of primary spectrum colors. The correspondence is not one to one. One wavelength gives a fixed color perception, one color can be a combination of wavelengths.

Community
  • 1
anna v
  • 233,453
  • 1
    I think you did not understand my question, and I do understand how color works. However, it turned out that the receptors are not really bell-curved for activation in certain ranges of light, as I thought they were (once again most books and pictures are simplified too much so that people who think about it more deeply get really confused about certain things). See my answer for more clarity on this. – Yeti Jul 01 '14 at 10:55
  • 1
    I think then your question is badly phrased. It is about the physiology of the cones and should be correctly asked at a biophysics site. – anna v Jul 01 '14 at 12:53
  • reference for perception of color http://hyperphysics.phy-astr.gsu.edu/hbase/vision/colper.html – anna v Dec 17 '23 at 11:40
13

The answer to this is to do with how the brain responds to the stimulation of the S, M and L cones in different ratios. This is best seen in the standard CIE XY chromaticity chart.

standard CIE XY chromaticity chart

The points on this chart represent different ratios of the stimulation of the three sets of cones. Increase the proportion to which the L cones are stimulated (compared to the other two), and you're heading towards the bottom right of this diagram. Increase the proportion to which the M cones are stimulated and you're heading towards then top. Increase the proportion to which the S cones are stimulated, and you're heading towards the bottom left.

The various wavelengths of visible light (in nanometres) are found around the curved edge of this shape; and the colours on the curved edge are the "rainbow colours" - that is, the colours of monochromatic light, from violet at about 400nm to red at about 700nm.

Most of the light we would perceive is not monochromatic, so we have a mixture of a whole lot of different wavelengths. The colours that we see when exposed to a mixture of wavelengths are found on the inside of this curve. For example, if we mix bluish-green light (500nm) with yellowish-green light (560nm), we'll see a colour that lies on the line segment joining "500" on the left side of this diagram with "560" on the right side. There's a variety of shades of green along this line segment - and you can get to all of them by changing the proportions of 500nm light and 560nm light.

Now, colours such as purple and rosey-pink are not found in a rainbow, so they can never arise as the response to a single wavelength of light. But they are found in this diagram, along the straight edge at the bottom. These colours are the brain's response to a mixture of wavelengths that don't match any single wavelength of light.

So the way to make light that appears purple is to mix the 400nm violet light (or even blueish light with a slightly longer wavelength) with the 700nm red light (or even orange light with a slightly shorter wavelength). By changing the ratio of these two, you can get to any of the colours near the straight edge of this diagram.

Dawood ibn Kareem
  • 689
  • 6
  • 16
  • "Now, colours such as purple and rosey-pink are not found in a rainbow, so they can never arise as the response to a single wavelength of light." Wouldn't you agree that there is a color that is both purple, violet, and magenta? Yet one might be of one single wavelength, and the other might be a mix of two wavelengths (blue and red)? – Yeti Jul 31 '16 at 17:48
  • I don't understand what you could possibly mean by "a color that is both purple, violet and magenta". If you look along the straight edge of the coloured region in the diagram, you see a whole range of different colours, from monochromatic violet to monochromatic red. It doesn't really matter which of these colours you refer to as "purple" and which as "magenta" - there are infinitely many colours here. And other than the violet and red at the two corners of the diagram, none of those colours are found in a rainbow. (Of course, this ignores the fact that you're looking at that diagram ... – Dawood ibn Kareem Jul 31 '16 at 20:35
  • ... on an RGB screen, and not actually seeing true monochromatic light). – Dawood ibn Kareem Jul 31 '16 at 20:36
  • Let me rephrase, because using names of colors is maybe not the right way to go about this. Let's agree that a single rainbow only shows colors all visible colors separated by their wavelength, from 380 to 700. Wouldn't you then agree, that the violet color of 380nm, is (in our eyes) equal to some specific mixture of 470nm (blue) + 700nm (red)? – Yeti Aug 03 '16 at 08:06
  • 1
    No, @Yeti, I wouldn't agree such a thing at all. Every point on the diagram in my answer is a different colour; and a human who's good at telling colours apart will see every one of them differently. Again, ignoring the fact that this diagram has been rendered on an RGB screen. The violet at 380nm is NOT on the line segment joining 470nm blue and 700nm red. So it's NOT a mixture of those two; and a human with normal colour vision will NOT perceive it as such. – Dawood ibn Kareem Aug 03 '16 at 08:27
  • So you really think that there is no line from the edge of the graph (single wavelength color) towards the center of the graph, without a change in hue? Because it seems to me, that from every point of the center of your chart you can go in a (curved) line towards the outside without a change in hue. Okay, the color is different because the lightness/brightness changes, but with color I'm of course talking about hue because brightness was never relevant to the question. – Yeti Aug 03 '16 at 08:29
  • 1
    OK, now you're asking something different. A colour has a hue and a saturation. Colours with full saturation are around the edges of the diagram. If you move on a straight line from the edge of the diagram to the white point in the centre, the saturation decreases, but the hue remains the same. So the "same colour" is not the same as the "same hue". – Dawood ibn Kareem Aug 03 '16 at 08:33
  • Correction, I meant to say saturation instead of lightness, of course. Well, maybe I should update the question with using hue instead of color then, to avoid confusion. However, I thought it would be obvious that if I talk about blue and red, I don't care what their saturation or lightness is (as long as it is not black nor fully desaturated, but then it wouldn't be blue or red anymore ;) ). In any case, that chart in your answer does not answer the question, because the chart is simply made by observation. The question is not about how color theory works, but rather about human perception. – Yeti Aug 03 '16 at 08:43
  • (1) I think it's a bad idea to change the question after you've got a bunch of answers; it may invalidate the efforts that everyone's already put in. By all means, ask a new question, using "hue" instead of "colour" - that would be OK, since that's what you meant. (2) Colour theory is all about human perception. The chart looks the way it does because of the way a human brain interprets signals from the different groups of cones in a human retina. Other species see colours differently - if we were to draw the CIE XY chromaticity chart for another species, it'd look entirely different. – Dawood ibn Kareem Aug 03 '16 at 08:47
  • You don't say anything incorrect, so far, everything you've said is true. However, you just seem to miss the point. To further clarify the problem, if you mix any hue, say: red and yellow, we perceive it as orange, which is also a single wavelength hue. This works for any hue, except for purple/violet. When blue is mixed with red, then we get the purple hue. However, the exact same hue can be achieved by a single wavelength that is at the far end of the higher frequencies of visible light (violet). The explanation for this is already given in my own answer to this question. But you disagreed.. – Yeti Aug 03 '16 at 08:50
  • 1
    But there are loads of hues of purple that aren't found in the spectrum. If you draw line segments in the chart, from both corners to the white spot in the centre, you'll inscribe a triangular area which contains purple hues that are not in the spectrum. You can't get to those purples by going far enough into the low wavelength end of the spectrum. – Dawood ibn Kareem Aug 03 '16 at 08:52
  • Let us continue this discussion in chat. – Yeti Aug 03 '16 at 08:52
0

"Mix" could be the confounder

when you "mix" red and blue paint you get purple, that's subtractive, when you add it is magenta

Red light + Blue light = Magenta

Red + Green = Yellow

Blue + Green = Cyan

Red + Blue + Green = White = Magenta + Cyan + Yellow

But is there any morphing or mixing of the constituent wavelengths?

there is no single wavelength we perceive as magenta

Dawood ibn Kareem
  • 689
  • 6
  • 16
Andrew Church
  • 11
  • 2
  • Dear Andrew Chruch, please try to be more professional in your answers. Let's not bother users by typing capitalized responses for keywords when it is not needed logically. Even if you are right, you need to be careful other users are not offended by your writing styles. Sincerely, – Benjamin Mar 21 '16 at 05:25
  • I never mentioned paint. If you add red and blue light, you get purple/magenta, this is an additive mix. Magenta and purple are both the same colors. Look at wikipedia, purple is defined as #800080, and magenta as #ff00ff, the hue and saturation is the same. Any color from a real world paint can be reproduced with light, but the primary colors for paint are different than for light due to the special properties of paint. (btw, English is not my motherlanguage, there is no magenta equivalent in my language, so the translation of purple is the only word used for that) – Yeti Mar 21 '16 at 15:26
  • Ok, well the miss understanding is purple/magenta. To my eye about 30 to 40 degrees apart on a color wheel. No specific frequency defines either color. If you shine a light of X Ray's and red light onto a screen, do either change? No? Would that be due to temporal incoherence, hence impossible for any interference pattern? If you "mix" two colours of light on a screen, none of the constituent frequencies /wavelengths change. – Andrew Church Mar 23 '16 at 19:47
  • Wavelength or Frequency, what determines the color we see?
    Our lecturer said both, it's a quantum energy level explanation. But in our problem sets, colors people see underwater (same frequencies, smaller wave length) we assume color equivalence of the underwater new wavelength as it would be perceived in air. How do linebreaks work?     – Andrew Church Mar 23 '16 at 20:13
0

The colour of light is defined by how it looks. The light at the short wavelength end of the spectrum is violet, not purple and violet does not look the same as purple. We actually have red, green, and violet cone cells. Stimulation of just red cone cells makes us see red. Stimulation of just green cone cells makes us see green. Stimulation of just violet cone cells makes us see violet. We see blue when red cone cells aren't activated at all and violet cone cells are stimulated the right number of times more than green cone cells. If you combine red light and blue light in the right proportions, you get purple light. That means violet is actually a deeper version of a bluish purple. Sometimes when somebody sees violet light, they don't notice that it looks different than purple because they're not very observant; it's close to purple; and they can't imagine seeing a deeper version of a colour than that which can be made by combining red, green, and blue light.

Source: http://www.sjsu.edu/faculty/watkins/violet.htm

Timothy
  • 1,648
0

This is how I understand things. I will not assume, that our eye cell cones are more sensitive to violet rather than blue. Violet and blue are just labels, that people use. I think I see Your point, but we need to better define the colors.

I think I understand our ability of the brain to mix the results of the signals of these receptors, producing colors like yellow.

The eye of average human only has three distinct types of cone cells. Let us call them red, green, and blue. Thus, we actually can only see three colors. What You call yellow is not yellow per se, but the combination of red and green that You can see. If You had yellow cone cells, I suppose, You would see a color You have never seen before. And You may not be able to imagine a color, that You have never perceived in Your life before. The same goes for colors like, orange, purple, violet, cyan, etc.

However, there is one problem with this which I cannot explain. How come that mixing red light with blue light also results in purple light? How is it possible that purple light can be achieved through mixing (additively) blue and red light, just as going to the shortest wavelength boundary of what we can see (from blue to ultraviolet via purple)?

In the figure with the visible spectrum, that You posted, it does look like "yellow" is between red and green, "cyan" is between green and blue, but "purple" (as well as "violet", "magenta") is not between blue and red. It is at this moment, that we must understand, that no average human being has actually ever witnessed true "purple" (as well as "violet", "magenta"). We simply call a mix of red and blue that way, but what we see is actually red and blue at the same time.

What is color, really? Color is a type of information. Just like information, it can travel at the speed of light. Furthermore, color propagates in form of light, where light consists of photons, and thus photons carry colors. We can associate a color a photon carries with the energy or frequency of that photon. Photons interact with matter in different ways (absorption, reflection, transmission). Eye cone cells are made of matter and can absorb photons. Different types of cone cells absorb particular photons better.

When red photons interact with Your cone cells, their energy is best absorbed by the red cone cells, stimulating further signals within the neural network of Your brain, that Your brain then associates with the qualia of red color. In reality though, it is information of red color, that gets read by Your eye and further processed and stored by Your brain.

Abyr Valg
  • 141
-3

I may be completely wrong about this, but I feel like the chart is backwards. Perhaps we should not be measuring light in wavelengths since the length of the wave is the inverse(?) of the amount of energy it is carrying.

So, blue light has a short wavelength - a high energy. Adding anything to that amount of energy should result in a higher amount - a shorter wavelength i.e. violet.

Experimenting with this concept proves to be inconsistent :( But I do wonder how the amount of energy carried along the wavelength plays it's part.

Tommie
  • 1
  • 2
    This doesn't seem to answer the question posed; seems to be more of a new question than anything. – Kyle Kanos Dec 31 '15 at 22:23
  • The question is not directly related to physics. Because color is not directly linked to wavelength of light, but it is a higher order word to describe our perception.

    The term you are looking for is frequency. A frequency of zero, is infinite wavelength, means that there is no energy.

    I do not understand what point you are trying to make.

     – Yeti Jan 02 '16 at 00:01