Friday, July 23, 2010

Mind of matter

I propose a tentative model of qualia and consciousness.

A rich thread of comments over at another blog motivated me to tackle the problem of consciousness with as much specificity as evidence currently allows, including the voluminous evidence for its material basis. So here I try to lay out a plausible model of fully brain-based consciousness, to address the dualist, who typically says that he or she can not imagine any material basis to consciousness, and even claims that there is some kind of philosophical necessity that the "hard problem" of consciousness is beyond scientific analysis entirely. (I have previously written about the neurobiology of consciousness, which remains unresolved.)

As a philosophical naturalist, I would predict that no new physics will be required to resolve the consciousness problem (such as 5th dimensions, quantum consciousness in microtubules, etc.). It is pretty clear that the brain is a messy product of evolution and works by electrochemical / molecular mechanisms. It is encoded by DNA, bounded by physical space (the head), and is affected at all levels by known lesions, chemicals and other effects. Nothing more esoteric is likely to show up in its study, other than incredibly intricate organization, which we still have great difficulty analyzing as a purely technical matter.

That leaves us with either an identity position that dualists don't seem to understand, (that there exists some brain-based processes that are objective physical events and also constitute subjective consciousness at the same time), or an eliminativist position, which we all dislike (that consciousness doesn't really exist). The problem may go away, but surely not by claiming that we have no consciousness after all. Indeed, the fact that there are unconscious processes that include forms of perception, intuition, and many other computations means that there is some real distinction between things that are unconscious and those that enter consciousness, and thus that consciousness is not an epiphenomenon or semantic issue.

So I should explain the identity position a bit better, despite the admitted lack of a real scientific solution in place.

Everyone has heard about continuity in movies- the need to keep hairstyles constant, props in the same places, etc. even when a shoot of one scene goes over several days, so that the illusion of continuity is preserved. I think this is a big clue to consciousness, which could be thought of as consisting, in part, of a brief memory loop that keeps our experience continuous. The ability to associate a split-second ago with now gives us not only a sense of time, but of consciousness itself, since otherwise we would be bombarded by what seem random stimuli, uncontrolled, uncorrelated, and meaningless.

So that is part of the answer, which can be easily imagined to be embodied by the gamma wave system in the brain that is the humming in-phase communication of ever-changing coalitions of neurons from all over. This would be the perfect system to glue together percepts and mental contents while giving them limited continuity of a fraction of a second, some of which is then stored to short-term memory, which is so notoriously poor, (hippocampus), some of which is stored in turn to longer-term memory (frontal cortex). Recalling something from memory weakly re-creates the engram (gamma pattern) that was originally driven by the perceptual system, and engages the same consciousness system, though with much lower intensity and different feel. We need a prompt like a smell or madelaine to assemble the engram again more fully, though it is never again quite the same as the first time in any case.

But we still have not gotten to the heart of the issue- the redness of red. We know that lesions like strokes can cancel out perceptions very selectively. There was a recent New Yorker article on the inability to read words while having unimpaired vision- an amazingly specific stroke effect. But the funny thing is that the sufferers routinely have to deduce their deficits. If something is missing, even something like half the visual field, we are not notified. The consciousness system soldiers on as though nothing has happened, forming an impression of the world that feels complete out of all the remaining bits and pieces.

Thus consciousness is very grade-able, both among humans with various talents and deficits, and among animals with different senses and brain sizes, as well as within one person in different moods, states of sleep, Dilbertian meetings, etc. It is also the kind of system that presents whatever comes up and doesn't care about what is missing. This again fits the gamma wave system, which wouldn't join up nerves that were not firing, but would just assemble whatever forms a minimal coalition of activity. And lastly, I'll make a stop at synesthesia, where one sensory modality bleeds into another, for instance inducing people to perceive certain numbers as having particular colors wherever they appear. The cause seems to be a literal mixing of neurons in relevant areas of the brain, which has obvious selective implications, both for problems arising from extensive wrong-wiring, and for the possible creativity and special perceptual powers that might result from limited cross-wiring.

What is red? Remember that red is a complete fabrication- our brain's way of mapping qualia to what in physics are numerically different wavelengths of light. The brain can apparently make anything it likes of the stimuli coming in. We should not be surprised if bats see their sonar world in living color as well. The question is why red feels so immediate, and what transaction actually occurs between red and "our perception" of it. In other words, what is the difference between unconscious and conscious information processing?

The difference seems to lie in entry into a privileged process in the brain, such as perhaps the gamma wave system. Other areas of the brain may talk to each other and process information, but that does not enter consciousness. Whatever does enter consciousness comes from relatively high level areas of processing, by virtue of the brain's wiring, which mostly flows from inputs (signals from the eye) to higher levels of processing that are biased towards participating in these gamma wave coalitions (analyses of shape, object identity, face identity, motion, etc.). Some people such as artists may able to "see" a scene without a lot of computational processing and overlaid interpretation, but as it is at a more basic sensory level, implying that they have conscious access to unusual levels of processing.

So consciousness is a part of a unique level of processing, where interpretations are collected and cross-correlated and learned from. Why does that feel real as unconscious processing clearly does not? Incidentally, unconscious processing may be just as complex, as we learn from our social intuitions, which can be incredibly spot-on and discerning about situations we have given little conscious thought to. Dreams also point to enormous depth and richness to unconscious processes, so it may be a mistake to put all this in some kind of higher/lower hierarchy.

A sense of reality may in this theory be the product of pure associational power- the construction of a sufficiently complex and rapidly integrated virtual world becomes what we call consciousness. If everything we see instantly generates a map of spatial, social, artistic, analytical, and other associations, that alone gives the scene "reality", immediacy, and red-ness. Seeing red becomes conscious not through some magical or impossibly recurrent homunculus behind our eyes, but by its particularity in contrast to other sensations and its web of immediate associations with the vast arrays of implicit and explicit knowledge we carry around in the database of our heads.

Suppose you were to see a uniform field of red all the time, day in and day out. You would no longer be conscious of red at all. You would not only be seeing the world through rose-colored glasses, but be seeing only the rose colored glasses and nothing else. Had you no other senses, you would probably lose consciousness altogether, a sensory-deprived victim of extreme torture. But of course we have many other senses- sound, smell, touch, and several newly discovered ones, such as body and spatial sense and the like. We also have dreams- the ability to draw on our voluminous past sensory impressions and their plastic re-arrangements in imagination. So it is literally impossible to do this kind of experiment.

This theory can account for the graded-ness of consciousness, its natural development through evolution, its materiality, and in addition provides a program for the instantiation of such experiences in artificial systems. It informs the common observation that high-level consciousness is dependent on knowledge, such that the study of art history gives us new eyes in museums, and the study of biology gives us new eyes in nature. It also provides a program for the study of consciousness, since its physical correlate should be quite discrete and find-able (by the identity theory mentioned above). It is in the correlation of subjective consciousness to this postulated physical correlate that the test will be found, either confirming detailed correspondence, and driving other theories to extinction, or not.

Lastly is the issue of emotion and pain. The above theory deals with cognitive consciousness, like *appreciating what red is in an immediate way. Emotions are simpler, not necessarily association-based. They are prior to high-level cognition. Past pain is certainly a strong spur to current aversion by association, but how did past pain happen? How was it perceived? There is something elemental and not associative about raw emotions like pain, which form perhaps our first and most traumatic form of consciousness.

Pain can function without consciousness at all. We typically notice that we have pulled our hand from the fire after the deed is done, and wonder at the reflex. But pain can also flood consciousness, issuing its imperative command to fix whatever is wrong. Similarly, other emotions typically operate below consciousness, to the point that our partners are often more familiar with our tics, mannerisms, and broadcast emotions than we ourselves are. As above, the cognitive contents of emotions can be characterized as associative, (what hurts, who did it to us, what is causing pleasure, etc.), but emotions carry extra contents injected alongside as the valence of the pleasure or pain- the command by other systems of the brain that this is something to be done again, or never again.

So it is no surprise that consciousness seems to be a complicated system, requiring rapid and  wide-spread associations to provide the brute contents and textured differentiation of cognition, as well as a brief memory loop that gives continuity to experience. Emotions are injected as extra cognitive contents and feelings, using specially responsive brain areas that date from early evolution and provide more than information: motivation.

I have tried to present a model that is plausible given current evidence, though obviously not very detailed. While there is plenty of circumstantial evidence that something material and eletrochemical in the brain is responsible, it will take a few decades yet to fill in those details.


  • Theologians working at public universities find it getting hot under the collar.
  • Can you lose your sense of smell
  • A billion dollars here, a billion dollars there...
  • One gets the impression that Hamid Karzai doesn't know much about power.
  • Though his isn't the only government beset with unaccountable bloat and corruption.
  • An interesting reflection on the worth of work:
".. while collecting salaries of between £500,000 and £10 million, leading City bankers destroy £7 of social value for every pound in value they generate." ... ".. for every £1 they are paid, childcare workers generate between £7 and £9.50 worth of benefits to society."