Question about the order of colors in the visible spectrum (NOT philosophy)

[sophiecentaur]

@Dave
My winge is about the words red green and blue when applied to the receptors. All three receptors respond to all wavelengths and not just to your implied red green and blue. Yes,you can mix r,g and b to be matched to any colour but that is not a function of any particular sensor but our brain's treatment of all three - whether we're looking at spectral (sodium) yellow or yellow as displayed on a TV. There is an important difference which people do not seem to appreciate.

 

[granpa]

Is it a silly question to ask *why* we see the orders of colors the way we do in the visible spectrum with respect to wavelength?

For instance, I know that comparisons to dogs are often made with the following visible spectrum comparisons:

[PLAIN]http://www.psychologytoday.com/files/u162/dog_color_vision.png
Notice how we still have yellows at longer wavelengths than the blues.

Or a bee's vision (ability to see UV light):

[URL]http://img.dailymail.co.uk/i/pix/2007/08_01/WoodDM0708_468x137.jpg[/URL]

Or snake infrared sensors:
http://en.wikipedia.org/wiki/Infrared_sensing_in_snakes

At any rate, what seems important to me is that wavelengths don't have any particular color encoding in them by definition -- it's just that our brains are sensitive to a particular threshold of wavelengths and then they chop up that threshold into visible colors, ranging from red colors to violet ones (and in that order, the interesting part to me).

My question is, though, is there any reason our brains do this? Why are the colors of the spectrum the way they are? As in, why are longer wavelengths encoded to be shifted towards red/orange/yellow colors whereas shorter wavelengths are encoded to be shifted towards green/blue/violet colors? Why couldn't the spectrum be inverted altogether?

I am not asking that question to invite a "Well, if it were the other way around, you'd be asking the same question" response -- to me, such a response is no different from asking why the moon is round ("no matter its shape, you'd be asking that question"). I am still after the underlying explanation, which we can offer for the moon -- but can we offer it to color, as well, or is it still a huge unknown?

are you asking how color vision evolved?

It evolved from black and white vision.

there are 3 photoreceptors

maybe one was originally for night vision and one was for day vision and one was for in-between.

once we evolved an iris we no longer needed all 3 so they evolved into color receptors.

 

[sophiecentaur]

Granpa.
You are my hero. You never mentioned red green blue. Well done.

 

[DaveC426913]

Is that the issue you had with my explanation sophiecentaur? That a red receptor does not exclusively pick up red, likewise green and blue?

That's certainly true, but does that level of detail illuminate or obfuscate the OP's question?

Perhaps it would be clearer if I referred to them as primarily red, primarily green and primarily blue receptors.

 

[SeventhSigma]

Okay, I'll simplify my question.

Let's assume we only have red receptors. Everything's in shades of red based on the wavelengths are are able to interpret.

For what reason does our brain interpret that to be "red" as opposed to some other arbitrary color? What I call "red" you might call "blue" and so forth -- why red? Physically speaking?

 

[Hootenanny]

Okay, I'll simplify my question.

Let's assume we only have red receptors. Everything's in shades of red based on the wavelengths are are able to interpret.

For what reason does our brain interpret that to be "red" as opposed to some other arbitrary color? What I call "red" you might call "blue" and so forth -- why red? Physically speaking?

This is wondering away from Physics somewhat and straying into Philosophy.

Let's ignore colours for the moment as they tend to confuse matters. Let us assume that we can only see light and dark, i.e. black and white. Note that this is equivalent to your scenario where we have only one "type" of receptor.

Does your question still apply in this case (black and white vision)?

 

[SeventhSigma]

As stated in the OP, this is not intended to be a philosophy discussion.

As a materialist, I have strong reason to suspect that everything in our universe has a natural explanation for it. Colors/consciousness is one such area where I feel I lack a materialistic understanding.

I am trying to understand how a bunch of atoms -- that we can discuss in purely mathematical relationships -- can give rise to something that is profoundly *not* mathematical.

And technically yes, brightness would be sufficient, too. If I can envision black and white, no reason I can't also envision colors, although I may be invoking an error by stating such a thing. I still think it's worth asking why certain color receptors encode things in certain ways -- why my red receptor takes in a certain wavelength and encodes it as "red" as opposed to "blue"

 

[Hootenanny]

why my red receptor takes in a certain wavelength and encodes it as "red" as opposed to "blue"

It doesn't. The receptors do not encode anything as "red" or "blue" or otherwise. The receptors simply release a chemical, the concentration of which is determined by the intensity of light (of a given frequency) striking it. Your receptors do not encode, describe, interpret or classify the light.

"Red", "blue", "green" etc. are simply words we have assigned to describe different physical stimulants. Other examples would be "hot" and "cold", "bitter" and "sweet" etc. Asking whether everyone experiences the same "red" is like asking if chocolate cake tastes the same for everybody.

 

[SeventhSigma]

Okay but that kind of explanation is sidestepping the point of my question (I understand the receptors technically don't do any encoding -- they're just receptors) -- at some point, the receptors take in a particular frequency and somewhere within my brain, that wavelength input is portrayed with a certain color. Likewise, other inputs/stimuli are converted into sensations.

I am trying to understand what the sensations are. I feel like answering it with "well, your brain does it" is no more satisfying to me than "Well, God did it." It's non-explanation. We can describe the brain in terms of physical processes, so why not perception?

 

[Hootenanny]

Okay but that kind of explanation is sidestepping the point of my question (I understand the receptors technically don't do any encoding -- they're just receptors) -- at some point, the receptors take in a particular frequency and somewhere within my brain, that wavelength input is portrayed with a certain color. Likewise, other inputs/stimuli are converted into sensations.

I am trying to understand what the sensations are. I feel like answering it with "well, your brain does it" is no more satisfying to me than "Well, God did it." It's non-explanation. We can describe the brain in terms of physical processes, so why not perception?

Brain chemistry is more a question for Biology than Physics. However, your question as to why (which really should be if) we all experience the same "red", for example, is more Philosophical in nature.

 

[BruceW]

Well, this is now getting to the more general question of why/how we have phenomenal experiences. (Which has already been briefly addressed on this thread).

In other words, we describe the brain as a connection of neurons that satisfies the materialist. But yet, I still experience seeing 'redness' or having 'consciousness' and I feel that it is something real, separate from what a materialist would hope to describe.

 

[BruceW]

I am trying to understand what the sensations are. I feel like answering it with "well, your brain does it" is no more satisfying to me than "Well, God did it." It's non-explanation. We can describe the brain in terms of physical processes, so why not perception?

As a direct answer, I would say the question is yet unanswered.

 

[Hootenanny]

I'm moving this thread to Biology in order that SeventhSigma's question on optical perception can be answered.

 

[granpa]

some question (like 'have you stopped beating your wife?") are best left alone till you are old enough to fully understand the issues involved.

'i don't know' is a perfectly acceptable answer.

 

[HallsofIvy]

I, for example, know that my left eye sees things as a slightly different color than my right eye!

Of course, the answer to why we see a "spectrum" is that we are seeing the order of wavelengths. Whatever color we might assign to a specific wavelengths, we will always see the same colors in the same order when we separate white light into its different wave lengths- colors corresponding to short wavelengths at one end, colors corresponding to long wavelengths at the other.

 

[eudo]

As was mentioned in a previous post, you are talking about "qualia": the sensation you get from perception. Many believe that a robot might be able to sense a red flower, but would not experience it the same way a human would. The way you experience it is the qualia.

A lot has been written about this, but mostly by philosophers, not physicists. It's not that it's a useless topic, it's just that physics currently has nothing to say about it. Some philosophers believe that it's a physical phenomenon that one day we might understand. Some believe it's a physical phenomenon, but we have no way of ever understanding it, and some believe that it's an illusion and no qualia exist (we're all zombies in effect). You clearly believe it's a physical phenomenon.

You can check out Consciousness Explained by Daniel Dennett for more on it, or take a look at the Mary's Room thought experiment ( which has more directly to do with color than the Chinese room experiment I mentioned).

But in any event, the only people who have written much about it are philosophers, so you're unlikely to find a satisfactory answer here. Honestly, you're unlikely to find a satisfactory answer from philosophers either, but you'll find more words at least :)

 

[sophiecentaur]

Is that the issue you had with my explanation sophiecentaur? That a red receptor does not exclusively pick up red, likewise green and blue?

That's certainly true, but does that level of detail illuminate or obfuscate the OP's question?

Perhaps it would be clearer if I referred to them as primarily red, primarily green and primarily blue receptors.

I think you must appreciate that the word "primarily" would add yet more confusion as it (wrongly) could easily be associated with "primary" . The peak of a response, which is what you are wanting to convey (and a very broad response, at that) would be much better described in wavelength (upper, middle or lower) without reference to colour at all. After all, colour is a subjective appreciation of all three sensor outputs. I am just keen that colour and wavelength should not be confused or used interchangeably. Your terminology is not helpful in that respect. Aren't we both trying to help people get things right?

 

[epenguin]

Okay, I'll simplify my question.

Let's assume we only have red receptors. Everything's in shades of red based on the wavelengths are are able to interpret.

For what reason does our brain interpret that to be "red" as opposed to some other arbitrary color? What I call "red" you might call "blue" and so forth -- why red? Physically speaking?

No. We both say blood is red.

Brain chemistry is more a question for Biology than Physics. However, your question as to why (which really should be if) we all experience the same "red", for example, is more Philosophical in nature.

It doesn't. The receptors do not encode anything as "red" or "blue" or otherwise. The receptors simply release a chemical, the concentration of which is determined by the intensity of light (of a given frequency) striking it. Your receptors do not encode, describe, interpret or classify the light.

"Red", "blue", "green" etc. are simply words we have assigned to describe different physical stimulants. Other examples would be "hot" and "cold", "bitter" and "sweet" etc. Asking whether everyone experiences the same "red" is like asking if chocolate cake tastes the same for everybody.

I don't care whether it is called philosophy - if it is then it is a necessary preliminary to biology. However biology does not seem yet able to cope with subjective qualitative sensations.

Everybody roughly has the same reaction to chocolate cake, and will make the same communicable facial expression which can be distinguished from everybody's face eating a lemon. It is an emotion. I intuit that emotions are the basis of consciousness. Until computers can have them too they cannot be like us. The emotions are the same fr different persons so I am fairly convinced that the taste of chocolate cake and the feel of red and blue, same emotions, is the same experience for everyone.

 

[DaveC426913]

I think you must appreciate that the word "primarily" would add yet more confusion as it (wrongly) could easily be associated with "primary" .

That clarification was for your benefit as you were the one who had issue with my explanation. Frankly, I think the readers would have had no problem with it.

Aren't we both trying to help people get things right?

Yes. Second-guessing the reader's level of understanding and nitpicking descriptions is something I don't think is bringing clarity to this thread.

 

[Ryan_m_b]

This thread has gone on long enough. Biology is not yet in a position to answer questions about the why or how subjective experience. We can point to areas of the brain that process colour and point to the organs responsible but we cannot really answer why blue is blue and red is red.

 

[SeventhSigma]

This thread has gone on long enough. Biology is not yet in a position to answer questions about the why or how subjective experience. We can point to areas of the brain that process colour and point to the organs responsible but we cannot really answer why blue is blue and red is red.

Thank you

As an aside, I feel like people really need to read what's being asked instead of handwaving the question away. I am really surprised at how almost everyone in this thread completely botched what I was actually arguing, here. Not saying this to be disrespectful -- but it's really frustrating to have words put in your mouth or misinterpreted/skewed/ignored.

I'm fully aware we both call blood "red." I'm fully aware the color processing is done in the brain itself and not in the receptors. I'm fully aware that we have color cones (this still ignores the fact that our brain sees colors by order of wavelength). It doesn't really matter to me if we all see the same colors for the same objects -- and it doesn't matter if my left eye sees things a bit differently than my left (personally, things have a more sepia hue to them out of my left). I think that if we *could* somehow average together the colors we see into general categories, we'd probably find that we all agree on what "red" is, "blue," interpret the same sensations for heat, cold, sourness, sweetness, etc.

But it's irrelevant to my underlying question of what, based on physical composition, brings rise to a particular set of qualia/sensations? Why does sweetness taste the way it does? Why do I perceive red as red and not blue?

 

[BruceW]

But it's irrelevant to my underlying question of what, based on physical composition, brings rise to a particular set of qualia/sensations? Why does sweetness taste the way it does? Why do I perceive red as red and not blue?

Yep, there is no answer to your question at present. As Ryan M B was saying in his last post, the 'redness' that we experience when we see a red object is not explained by any physical reason.

Similarly the experience of consciousness is a mystery. For example, we could have a world where human brains are simply a bunch of neurons and electrical signals, and the humans assert that they are conscious, but actually none of them have true consciousness or experience.
Clearly, such a world would progress in the same way as our world, so even the existence of consciousness is debatable. The reason the debate goes on is because I really do feel that I am conscious and that I see and experience the world around me.

 

[Ryan_m_b]

I forgot to mention in my last post, the answer to this question will only be found if we solve the hard problem of consciousness.

 

[sophiecentaur]

That clarification was for your benefit as you were the one who had issue with my explanation. Frankly, I think the readers would have had no problem with it.

If you read the very post that follows yours, you will see reference to RGB receptors. You will also read in a few earlier posts that "if you had only Red receptors then everything would look red". That clearly shows a huge misconception because Redness is only a quantity by which different "colours" can be recognised by using and comparing more than one analysis curve. With a single kind of receptor with a broadband response and a peak at the long wavelength end, all you would see is varying degrees of brightness. You would have no idea whether what you were looking at was producing a lot of light of short wavelength or less light of long wavelength. Moreover, you could shine 400nm wavelength light of an appropriate level and achieve a perfect 'match' to a different level of light of 600nm wavelength. They could not be distinguished. The receptors would 'see' brightness of light of all wavelengths.
With just two receptor types (short and long wavelength centred), you could probably resolve the spectral 'colours' and put them in an order of wavelength but there would be many different combinations of wavelengths ('non-spectral' colours) which would be confused much more than with our standard three sets of receptors. Moreover, even three receptors are very easily 'fooled' into seeing matches between the majority of our colour space by the appropriate mix of three primaries (Synthesis not Analysis, here). Just think how hard it would be to send colour TV pictures is this weren't the case.

Yes. Second-guessing the reader's level of understanding and nitpicking descriptions is something I don't think is bringing clarity to this thread.

I don't need to "second guess" levels of understanding when I see the same howlers about colourimetry appearing on a regular basis.

One man's "nitpicking" is another man's attention to what the other man may not appreciate as an important detail. Could it just be by chance that the CIE Colourimetric Observer functions are referred to with the letters x,y and z, rather than your favoured R, G and B? Or could it be because the CIE actually see it as an important distinction? You are justified in doubting my knowledge and expertise (and I am not at all offended and it's all good clean fun) but you could consider that the CIE may have something worthwhile to contribute to the matter.

I really believe that the reason for many of the quasi philosophical stuff that used when discussing colour is down to the fact that people don't actually understand the basic process but think they have it sussed. Interestingly, you don't come across so many similar discussions about hearing sound and music. This, I think, is because it's actually so complicated that people don't even try to over-simplify it - and, consequently, don't rush in where angels fear to tread, as they do with the colour thing.

 

[SeventhSigma]

sophiecentaur: If there has been misinformation/obfuscation in this thread, would you be so kind as to clear it up and explain things?

You seem to imply that if we only had one type of color receptor (red) we still would not see "red" at all but just light/darkness via our rods? If you were to take a human and remove all green/blue cones, they'd somehow lose red, too?

A bit confused as to what you're saying, sorry.

 

[sophiecentaur]

You only 'see red' when you can distinguish it from other 'colours'. I think you would agree with that.
So called black and white film produces an image which just varies in density. There are different types of monochrome film. Orthochrome and Panchromatic film have different sensitivity curves and give slightly different black and white prints of a particular scene because the relative sensitivities give different grey levels for different 'colours' of the same brightness. You can print onto normal b/w paper or 'tone' the paper any colour you like (e.g. sepia) but there is no colour information about the original scene.. Unlike a brain, photo paper doesn't try to second guess what it sees.


A set of cones that had only one sensitivity curve can only provide a single value of output signal which corresponds to the light level and where the wavelength of the received light happens to sit on the sensitivity curve. That, on its own, can't tell you anything about the wavelength (or combination of wavelengths) of light falling on it.

With slight differences between individuals, we mostly agree on what we call red. If we look through a 'red' filter (broad band around the long wavelength end of the spectrum) at a scene, it all looks red (named by popular consensus) but so do the green trees and the light blue sky of the original scene. But we only call it red, in fact we only have the concept of different colours, because we are aware of the different outputs of our three sensors.

Here is an analogy. If you hold your hand in front of a red hot or a white hot poker, its radiation will just give your hand the sensation of heat. There is no way you would say "that's infra red' " or "that's visible light", from the sensation in your hand. You have no mechanism for distinguishing with the heat sensors on your skin so the question of 'heat colour' doesn't arise. You couldn't describe, to someone else anything but 'a bit hot', 'very hot' or 'too hot'.

If your human subject suddenly had all but one set of sensors removed (I refuse to use the words red, green and blue in this context) then, i guess, initially, they would say that everything 'looks red'. That's because they are assuming that the other sensors are telling them that there is no spectral content at the short wavelength end. Their brain is still functioning on the assumption that it has all three sensors. Before long, as brains do, their brain would be tired of trying to differentiate between this red and that red and the other red and just treat a scene as a set of different 'grey levels'. It would lose the idea of colour because it would be irrelevant. A creature born with a single set of sensors would just not be aware of wavelength or mixtures of wavelengths so could have no concept of colour in the first place.
I have great difficulty with pictures which purport to display what other animals can 'see' in the way of colour. All we can say about the colour vision of other animals (or even other humans) is their ability to detect different em wavelengths and discriminate between the different 'colours' that we perceive and agree about. These are yet another instance of over-simplification which leads to confusion.

 

[SeventhSigma]

I'd agree that we know what red is because we can distinguish it, but there's no reason to say that if we suddenly removed the other two receptors, we suddenly wouldn't know what "red" was even though we clearly have already made the label ahead of time (via what we see). Similarly, this would be no different from someone born with only one type of receptor -- he'd still see "red" even though he wouldn't know how to differentiate it. It'd be like Terminator vision. Our brains might eventually grow tired of differentiating the red levels and might eventually push it to b/w vision, but that doesn't detract from the notion of being able to see all-red in the first place.

At the very least, the all-red example was just a way to push past the confusion that was arising as a result of talking about multicolored specta in order to just talk about one single color.

 

[bobze]

I'd agree that we know what red is because we can distinguish it, but there's no reason to say that if we suddenly removed the other two receptors, we suddenly wouldn't know what "red" was even though we clearly have already made the label ahead of time (via what we see). Similarly, this would be no different from someone born with only one type of receptor -- he'd still see "red" even though he wouldn't know how to differentiate it. It'd be like Terminator vision. Our brains might eventually grow tired of differentiating the red levels and might eventually push it to b/w vision, but that doesn't detract from the notion of being able to see all-red in the first place.

At the very least, the all-red example was just a way to push past the confusion that was arising as a result of talking about multicolored specta in order to just talk about one single color.

Sophie is onto the right thing here with sensitivity curves, I don't think they elaborated it that well though. It has to do with how ensemble coding and overlapping receptive fields (note not literally a visual field, but information field) work. I don't have time to elaborate more right now. Block exam week and I've used all my forum time :\.

I'll wax later.

 

[sophiecentaur]

I would also agree that we 'know about' red, with or without our the help of all three sensors but only because we are 'wired' to expect three inputs and have learned about it from experience. (The brain very quickly stops bothering to deal with superfluous information, however and the 'redness' of a permanently red sceenewould rapidly lose its significance, I'm sure). Give a new born baby a pair of red filtered goggles and it will develop absolutely no concept of red - or any colour value.
Many, or even all, of our subjective impressions of our world are based upon comparison and not absolute values. We can easily see which is the taller of two people when they stand next to one another. We need more extreme differences (and probably other subconscious clues and references) to do this by memory.
I still seriously believe that colour is essentially the result of three of bobze's (above) information fields. The justification is that colour reproduction, using primaries which are fed from models of our human colour sensitivity curves, works so well. Three numbers (Y,U,V: X,Y,Z : RGB: C,Y,M - take your pick) can be used to pass enough information to give a pretty good measure / match of nearly all of the visible colours (not just the spectral ones, with which so many people appear to be so totally obsessed).

I don't understand why something which can be identified in terms of three signals (to the brain, down a wire or in a computer memory) should be viewed any more philosophically than, for example, our sense of temperature. We have a very crude way of assessing the world around us in terms of the light arriving from different objects. We MAP many different combinations of spectra onto one thing that we call Colour (An amazing amount of data-compression!). That works fine, evolutionarily, and allows us to identify a useful set of information about our surroundings but it is a really poor analysis of what is actually entering our eyes. It's just (as with most of the features of our spec) good enough to get by and no better.

Will there be another thread, along the same lines, about taste and smell? Same things apply.

 

[apeiron]

I don't understand why something which can be identified in terms of three signals (to the brain, down a wire or in a computer memory) should be viewed any more philosophically than, for example, our sense of temperature...Will there be another thread, along the same lines, about taste and smell? Same things apply.

This is an interesting question. I think colour is different in that it is more "out there" and static as an experience.

So sound varies continually (though we would have a more "spectral" response to a constant tone - after a while listening to waveform, you might wonder at why a buzz is like a buzz, a rasp like a rasp).

Taste, smell and feel come through direct contact. As you say, some kinds of sensation like pressure or temperature are continuous gradients. So there is change that is quantitative but not qualitative. That seems philosophically less troubling (even if it perhaps shouldn't). But for smell and taste, as with colour, there is also quite distinct qualitative change. You have both sweet and salt - and one is not simply a greater or lesser amount of the other.

So colour has qualitative difference. It is also more out there in the world in that it seems a property of an object (rather than of wavelength information reaching our eyes). It does not change constantly, or make its appearance intermittently, or arrive as the result of some deliberate interaction, like most other sensations do.

No one is so bothered by light and dark visual experience. That is a simple continuous gradient of luminance. The object does not own the phenomenon in the same apparent way because brightness is so clearly the result of a light source.

But red seems to be something right "out there" as a definite objective qualitative property. And we expect the world to only have quantitative ones. That is, we can see the reasons why an object might have a quantity of brightness, or edge, or motion, or other such properties. The causes are also visible to us in a sense. But colour appears to be a quality of the world - lacking in the kind of cause that would make red look red, rather than some alternative hue.

So subjectively, colour is more of a mystery. We can't see its reasons as also part of the world.

Yet study the visual system and we can see that colour experience is the result of brain processes of very much the same kind as all our other experiencing. So yes, in principle, it should create no more (and no less) philosophical issues than the rest.