Applied Philosophy: Representation, Metarepresentations, and Qualia

V.S. Ramachandran (2004), A Brief Tour of Human Consciousness (New York: Pi Press: 0131872788):

p. 108 ff: In chapter 2 I mentioned the "blindsight" syndrome, in which a patient with visual cortex damage cannot consciously see a spot of light shown to him, but is able to use an alternative spared brain pathway to guide his hand unerringly to reach out and touch the spot. I would argue that this patient has a representation of the light spot in his spared pathway, but without his visual cortex he has no representation of the representation--and hence no qualia "to speak of."

Conversely, in bizarre syndrome called Anton's syndrome, a patient is blind owing to cortical damage but denies that he is blind. What he has, perhaps, is a spurious metarepresentation but no primary representation. Such curious uncoupling or dissociations between sensation and conscious awareness of sensations are only possible because representations and metarepreentations occupy different brain loci an can therefore be damaged (or survive) independently of each other, at least in humans....

The flip side of this is, just as we have metarepresentations of sensory representations and percepts, we also have metarepresentations of motor skills and commands such as "waving goodbye," "hammering a nail in the wall" or "combing," which are mainly mediated by the supramarginal gyrus of the left hemisphere.... Damage to this structure causes a disorder called ideomotor apraxia.... [I]f asked to "pretend" to hammer a nail into a table, they will make a fist and flail at the tabletop.... Or when asked to mime combing her hair a patient will make a fist and bang it on her head, even though she understands the instruction and is perfectly intelligent in other respects. The left supramarginal gyrus is required for conjuring up an internal image--an explicit metarepresentation--of the intention and the complex motor-visual-proprioceptive "loop" required to carr it out. That the representation of the movement itself is not in the supramarginal gyrus is shown by the fact that if you give the patient a hammer and a nail he will often execute the task effortlesly...

---

p. 147 ff: The second problem is why the sensations take the particular form that they do.... [Q]ualia--consider the manner in which we experience... wavelength... and pitch.... Even though wavelength is a continuous dimension we experience colors as four qualitatively distinct sensations--red, yellow, green, and blue.... Adjacent colors... are "miscible"... we can see orange as a blend of red and yellow and purple as comprising red and blue. But non-adjacent ones are immiscible... it is hard to even imagine a bluish yellow or a reddish green. Thus color sensations seem chopped up into four immiscible bits. But this isn't true of soundwaves... we hear the full range... as a continuum, with no breaks in qualia. All this is obvious, but the question is why....

The fact that different modes of experience apply to wavelength and pitch suggest that qualia cannot be epiphenomenal; they must have an evolutionary function--such as serving as a mnemonic aid....

Non-spectral colors are perhaps the strangest thing of all. There is no wavelength that is purple--purple is a mixture of short (blue) and long (red) wavelengths. As you go around the color wheel starting at red, you can shorten and shorten the wavelength and go smoothly through orange and yellow and green to blue and then, as you continue around the color wheel, you have to add long (red) wavelengths while keeping the blue wavelengths to get purple, and then take away the short (blue) wavelengths to get red. And there is nothing in your consciousness or its qualia to tell you what your retina knows very well: that going from red to blue through orange, yellow, and green is a very different process on the photon level than going from blue to red through purple. Not only do we have qualia, but they are very false to the way the universe really is. Purple is not a wavelength halfway between red and blue. Green is a wavelength halfway between blue and yellow.

Comments

[jfaberuiuc]

I can't say much about the former excerpt, but as to the latter, discrete color vision comes from having three discrete types of cone cells in the eye, sensitive to different frequencies. Input from these is differentiated into red-green (a genetic flaw in this one leads to the familiar form of colorblindness), blue-yellow, and black-white. The discretization isn't physical (light is coninuous), but rather physiological (the detector is discrete). Our ears are continuous detectors. As to why one is continuous but the other discrete physiologically: because we are a contingent result of billions of years of evolution, and that is the way it happened to work out. Either system could have ended up the other way, but didn't.

[christofay]

We are reaching for curious intelligence to try to understand why the other cave dwellers are not interpreting the shadows the way we are. Thanks for sharing more of everything that you know, Professor, but I prefer that we skip ahead to the post on how and where to contact seers or better yet how to find the people that can actually change the near term future. Are they nearby in the cave?

[Rich]

The discrete nature of visual qualia and the continuous nature of auditory qualia may have something to do do with vision being a "solute sense" and hearing being a "solvent sense":

http://scienceblogs.com/pharyngula/2006/10/evolution_of_sensory_signaling.php

[Rich]

I don't see the end of the URL that I just posted. It's

evolution_of_sensory_signaling.php

[Jacob]

"Blindsight" is also a pretty good SF novel that's free online: http://www.rifters.com/real/Blindsight.htm

[SJ]

"Even though wavelength is a continuous dimension we experience colors as four qualitatively distinct sensations--red, yellow, green, and blue.... Adjacent colors... are "miscible"... we can see orange as a blend of red and yellow and purple as comprising red and blue. But non-adjacent ones are immiscible... it is hard to even imagine a bluish yellow or a reddish green. Thus color sensations seem chopped up into four immiscible bits."

This sounds like made up nonsense. Look at a rainbow. Where are the four qualitatively distinct sensations? Nowhere, that's where.

Light is a continuous spectrum, but our eyes have only three different types of cone cells which are sensitive to red, green and blue light respectively (contra jfaberuiuc above).

Neverthless, our brains are set up to see the entire visible spectrum according to which cone cells are stimulated.

Red cells and blue cells stimulated? That's magenta or purple, depending on which cells are stimulated more. Red cells and green stimulated stimulated? That's yellow or orange. Etc.

Adjacent colors are miscible because we only have three different color receptors, and they "wrap around" into a circle. There simply aren't any non-adjacent colors with respect to the three types of sensors.

See the Wikipedia article on cone cells for a basic intro:

http://en.wikipedia.org/wiki/Cone_cell

[Jonathan Goldberg]

Note: the title link for "A Brief Tour of Human Consciousness" (http://exec/obidos/asin/0131872788/braddelong00) doesn't work.

[David Lloyd-Jones]

One of the interesting things about humaan consciousness, or at least about human behvious as it proceeds allegedly on the basis of consciousness, is that large changes to it can be made "in software," that is without actually taking a screwdriver to the hippocamus, nor subjecting the brain to years of pickling in alcohol, cocaine and upper-class education.

The programming of the human brain is data in the classical Turing/Church/Von Neuman sense.

Thus the simple uttering of a verbal formula, e.g. "I have been saved/reborn..." etc., can bring about many of the changes that Ramachandran has observed as caused by gross pyhsical insult to the wetware brain.

[Njorl]

"There is no wavelength that is purple--purple is a mixture of short (blue) and long (red) wavelengths. "

Simply false. 400 nm is violet, which is purple.

Now when we see purple light generated, it is usually as a combination of red and blue lights, but it can be done directly, say with a Gallium Nitride laser.

[shah8]

okay, first of all...

njorl, indigo is not purple.

second...auditory sense is largely unique in the solvent nature of it's reception. None of the other senses are like that in data, and there are alot of parts to it, not just the osteoliths moving around, stimulating nerve cells. Now, the whole "purple" thing...well, that's human conciousness for you. Most of what is constructed by your brain for whatever the "you" that you is, is false, very much in the sense that electron microscope images use false colors. Everything from your sense of touch, to your awareness of other people, even to your awareness of what you are doing, are represented by symbols, and most of the time are blurred to preserve continuity of experience.

Third...this is the wrong question. The real question is why do we percieve purple as significant at all? Not why do we percieve purple. When I thought on it, my attention was definitly focused towards color reception as an emergent phenomenon in nature. There are a bunch of critical signals/chemicals that are purple, and magenta is much more common in nature than blue. And color is one of the primative languages. It makes me wonder what butterfly "decided" to be purple, and flap purply for all the other birds/other animals to see what a lovely color purple is...

[ogmb]

Also from Wikipedia:

Violet is a spectral color of a (approximately 420–380nm) shorter wavelength than blue, while purple is a combination of red and blue light and when regarded as electric purple (see below), it is the only color on the color wheel besides magenta and rose that is not a spectral color--purple is an extraspectral color, along with magenta and rose. In fact, purple was not present on Newton's color wheel (which went directly from violet to red), though it is present on modern ones. There is no such thing as the "wavelength of purple light"; it only exists as a combination. Also, violet light varies solely by wavelength, while purple varies by the proportion of red to blue light.

http://en.wikipedia.org/wiki/Purple

[Njorl]

[This is absolutely, totally, highly cool! How come I never knew this before?]

About purple...
I noticed an interesting thing while looking at the sensitivities of the receptors in the eye. The blue (beta) and green (gamma) receptors tail off to insignificant sensitivity much more sharply than the red (rho). The red sensors are actually more sensitive to very short wavelengths than the green, and begin to rival the blue ones.

As you approach the limits of human vision near the ultraviolet, the ratio of sensitivity of the red sensors to blue sensors is increasing, not decreasing.

So, photons of purple light (and yes, there is indeed a wavelength that is purple) stimulate the receptors in exactly the same way that combinations of red and blue light would. Had it not been for the quirk of the long tail in the red receptor sensitivity, we may never have had the concept of purple. The color wheel might have been a color bar. Blue would have just covered a larger wavelength range like red does. This might need explaining. While most texts say human sight goes to about 700 nm, it really goes much further (I have no problem seeing 790 nm). It's just that everything looks the same shade of red because the green and blue receptors are of negligible sensitivity. If there were not a long tail of sensitivity in the red receptors at the blue end of the spectrum, everything we see shorter than 420 nm would look exactly the same shade of blue.

[CJColucci]

If we could see a wider swath of the electromagnetic spectrum, what would it look like? What colors would we see? Would we see new colors we cannot now imagine, or would the ROY G. BIV spectrum expand and would we see finer distinctions within that band? Imagine seeing radio waves all over the damn place.

["Q" the Enchanter]

"Green is a wavelength halfway between blue and yellow."

But does my *retina* really "know" that?

To put the question a little differently, does a swatch of uniformly distributed blue and yellow pixels merely *appear* green, or is it *really* green? It seems to me the metaphysically neutral way of describing the matter is to say that both green pixel swatches and blended blue-yellow pixel swatches are a way for the world to elicit identical green-experience. (Or is there a privileged vantage point from which to judge that the one (or the other) is "true" (or "false") with respect to how the universe "really is"?)

[SJ]

"To put the question a little differently, does a swatch of uniformly distributed blue and yellow pixels merely *appear* green, or is it *really* green?"

You're asking whether a mix of yellow and blue light is really the same as green light. There's an objective answer to this question, and the answer is "no".

From a distance, your eyes can't tell the difference. Close up, when you can resolve individual pixels, they can tell the difference.

But what if the pixels are too small, you ask? A spectrometer can still tell the difference, as can a lot of people.

For a concrete example, suppose you mixed yellow and blue paint to make something that looked green, then painted your truck with it. Would your truck *really* be green and blend in with the trees in a forest?

No. About 7-10% of human males would be able see your truck instantly.

via Wikipedia: Morgan MJ, Adam A, Mollon JD. "Dichromats detect colour-camouflaged objects that are not detected by trichromats." Proc Biol Sci. 1992 Jun 22;248(1323):291-5