Wednesday, January 30, 2008

You Down With the NCC? No, Not Me.

Seed Magazine, the really entertaining science-themed webzine, posted an article called "Questioning Consciousness" today. It's very thoughtful and well written, and makes an argument quite close to my heart--that we are, for the most part, looking at consciousness in radically incorrect terms, and this conceptual confusion is hindering scientific investigation. However, the author--a philosopher from the London School of Economics by the name of Nicholas Humphrey--proposes several radical ideas about consciousness that, it seems to me, take the discourse in the wrong direction entirely.

The predominant theme of the article is that consciousness is an illusion of sorts, akin to the sort of trickery that goes on when a Penrose Triangle is viewed from certain angles--that is, the "problem" of consciousness is only a problem at all because we're trapped in our subjective first-person jail cells. Snip:

Since the qualia are indeed so up-front and remarkable, and since no one knows why this is, we are all, most probably, going to start off by asking what may be a bad question: "How can we explain the existence of these qualia as we experience them?" So here, again, it will only be if we undergo a radical shift in perspective and realize that the "qualia as we experience them" could be a mental fantasy, that we shall move on to asking what may be the good question: "How can we explain why we have the impression that such fantastic qualia exist even if they do not?" But, here is why it is likely to be so difficult to make this move: In the case of consciousness, we cannot simply change our perspective to see the solution. We are all stuck with the first-person point of view. So, the result is we persist with questing for the qualia as such.

Yet if consciousness is a trick, then of course this quest is a fool's errand. It will make no more sense to try to explain the existence of qualia than it would to explain the existence of the impossible triangle. What we should be doing instead is trying to explain just how we have been set up—and why.

I'm immediately suspicious of accounts that try to eliminatively reduce away consciousness as "illusory" or "not really there." Dan Dennett is notorious for this (he's another one who maintains that qualia don't really exist), as is Paul Churchland. It seems to me that trying to deny that we really are conscious is an exercise in futility, somewhat like Wile E. Coyote thinking that he can keep running off the edge of the cliff so long as he doesn't look down. Indeed, if there's one thing that's defining of humanity, it is the elaborately woven tapestry of conscious states that are open to us. I've read Humphrey's article several times through now, and I still can't even begin to fathom what he could even mean by asserting that consciousness in general--and qualia in particular--is just a "trick of the brain."

I really hate to appeal to Descartes since he was so catastrophically wrong about so many things, but this really does seem to go back to his cogito--if I can be sure of one thing in this universe, it is that I am a conscious and thinking thing. That's what's so unique about cognition, thought, and conscious experience: if it seems to me that I am conscious, then by definition I am conscious of something. The trick is all in the seeming, and that's what Dennett, Churchland, and (apparently) Humphrey seem to miss: even if we're spectacularly wrong about what consciousness is, the fact that we've formed an opinion that can even be right or wrong necessarily entails that we're the sorts of things that can have beliefs, desires, ideas, and opinions--in short, that we are conscious creatures.

Humphrey goes on to pose a few questions about the nature of consciousness that he thinks are only answerable (or at least much more likely to be answerable) if we adopt this position that consciousness is really "nothing special." I'd like to take a moment now to pose a few answers of my own to these questions, answers that don't require us to take the phenomenally (pun fully intended) counter-intuitive step of supposing that my experience when I look at a ripe tomato is nothing more than an illusion (again, whatever that means).

1. What exactly is the real-world brain activity that we are engaging with when we say a sensation is like something?

Who knows, but I would bet dollars to donuts that there isn't just one token activity (or even type of activity) that meets this description; this should be a reflection of how much I believe this to be the case, as I don't even eat donuts (and while I don't eat dollars either, many more merchants accept them as payment). Humphrey (along with many others, Searle included) seems to think that we will one day find the "NCC," or neuronal correlate of consciousness. I'm skeptical that we ever will, but for rather atypical reasons (go figure). I'm not a Mysterian, nor am I any breed of Dualist--that is, I think that we will one day be able to give a scientific account of how consciousness happens, and I think that we live in one and only one world that is composed of and only composed of physical particles. I do, however, reject the computationalist notion that the mind is "modular."

Briefly, this theory says something like the following: the best way to understand the mind is to think of it in terms of modules. We have an "arithmetic module" that lets us add and subtract, a "balance module" that lets us stand upright, and a "consciousness module" that's responsible for our subjective and unitary experience of the world. If we "popped out" the consciousness module, we'd be left with p-zombies, creatures that behave as if they are conscious in every way, but actually lack any sort of subjective experience. As I said, this idea does not ring true to me.

Why not? Before I answer that, I'd like to at least take a stab at answering the other questions posed by Professor Humphrey--in the process of posing these answers, I think I'll probably end up addressing this question as well.

2. Why does this activity have the (tricky) properties it has, such that our experience of it is seemingly something so strangely private, not of this world, and indescribable in common terms?

Once again, who knows! That's an empirical question, and needs to be sorted out by science. We can make a few general statements, though. First, it seems that we can be pretty sure that our consciousness (and thus our qualia) is caused by activity in the brain; I take this to be relatively non-controversial. While some still cling to Dualistic theories of mind we, as modern people of science, can set those theories aside with little chance of error, I think. We can also, I think, safely assert that there is no token/token identity between "mental states" (for lack of better vocabulary) and "brain states"--that is, the brain states causally responsible for my experience of red will not be token identical with the brain states causally responsible for your experience of red, even if we're looking at the same object. This is a very important point, and one that we will come back to.

Additionally, I do object to the notion that our qualia are somehow "otherworldly." What could be more "of this world" than the 'what it is like' quality that conscious experience has? I presume that Humphreys means to say something about the peculiarly subjective nature of qualia, but I don't really see why this nature requires any peculiar explanation--if they didn't have that nature, they wouldn't be qualia at all! In other words, consciousness (at least as we define it) is by definition private, subjective, and "other worldly." No further explanation necessary; once we understand what causes consciousness in general, I suspect that we'll see why it can't be anything but "other worldly."

3. What makes this trick work? How is it done?

A similar answer to the above question--its done through the peculiar arrangement of our neurobiology. If you want a more specific answer, give me a few more PhDs and a few more decades and then ask me again.

4. What is the point? Why was it designed like this? What might have been the evolutionary advantage of our having these marvelous experiences?

John Searle likes to compare consciousness (as a process of the brain) to digestion (as a process of the stomach), and I think this fairly apt analogy can aid us in answering this final question. In the stomach, digestion is a high level process that consists in many lower level processes working toward a common "goal." In light of this, it seems rather silly to say something like "I don't see what the evolutionary advantage of digestion is! It seems that we'd be just as well off with only hydrochloric acid to break food down, bacteria to help us absorb nutrients, saliva to act as an enzyme to begin the process, etc." The reason this statement is silly, of course, is just this: what we've just described is digestion, simply seen from a lower level of description! When we say "stomach acid, intestinal bacteria, saliva, etc.," we aren't enumerating components that are operating independently of digestion, but rather components that constitute digestion. Humphrey writes:

The final challenge will be explaining the biological purpose of all this. We can surely assume that the kind of development I have sketched above will not have happened accidentally. It must be the result of natural selection favoring genes that underwrite the specialized neural circuits—whatever they turn out to be—that do indeed sustain the illusion of qualia, giving rise to the magical mystery show observed by the first-person. And it is axiomatic that this will only have happened if those lucky enough to be spectators of this show have somehow been at an advantage in terms of biological survival. Yet, how can this be if, as is widely assumed by theorists, the phenomenal richness of consciousness is of no practical value whatsoever?

Fodor has stated this aspect of the problem bluntly: "There are several reasons why consciousness is so baffling. For one thing, it seems to be among the chronically unemployed. What mental processes can be performed only because the mind is conscious, and what does consciousness contribute to their performance? As far as anybody knows, anything that our conscious minds can do they could do just as well if they weren't conscious.

Humphrey proposes that the solution to this problem lies in recognizing that:

the role of phenomenal consciousness may not be like this at all. Its role may not be to enable us to do something we could not do otherwise, but rather to encourage us to do something we would not do otherwise: to make us take an interest in things that otherwise would not interest us, or to mind things we otherwise would not mind, or to set ourselves goals we otherwise would not set.
But I don't think this is satisfactory either--the behavior he is speaking of could just as easily manifest in p-zombies; that is, there's no functional difference between a creature that takes interest in and does all the activities Humphrey has in mind because it is conscious, and a creature that still performs the activities, but does so without any sense of a subjective experience. Evolution is the ultimate behaviorist, and doesn't care how (or whether) the behavior is being processed internally--as long as the organism survives to reproduce, it is a success from an evolutionary standpoint.

Instead, I'd like to propose that we have consciousness simply because it is a necessary "add on" that comes for free (loosely speaking) with all the other cognitive machinery we rely on; it is the very fact that we are capable of advanced reasoning, representation, and communication that explains why we have consciousness, and it is the fact that we have consciousness that explains why we are capable of reasoning, representation, and communication. Once again, this view is a rejection of the "modular theory" of the mind--in short, we can't consider consciousness as a faculty in isolation from all the other cognitive faculties we enjoy; while consciousness qua consciousness might not provide any specific evolutionary advantages, our other brain-based abilities most certainly do, and the brain structures responsible for these abilities also are causally responsible for consciousness. I consider this view to be a breed of holism and emergentism both.

A careful reader will no doubt notice that the view I'm articulating here has certain leanings toward epiphenomenalism, and I would like to take a moment to assuage these fears. Epiphenomenalism, you may or may not recall, is a broad family of views claiming that the mind cannot possibly have any causal effect on the body. All mental states are epiphenomena, according to this view, and have an existence something like froth on the tops of waves--the froth (mind) is caused by the waves (body), but doesn't really have any effect on them. I'll admit that my view shares certain aspects with epiphenomenalism (specifically that consciousness is a necessary result of brain chemistry), but there is one vital point on which my view differs. Epiphenomenalism carries within it an implicit endorsement of Dualism: it only makes sense to talk about a causal relationship (or lack thereof) between 'mind' and 'body' if those are two very different things. On my view the mind (consciousness, etc.), like digestion in the stomach, is the necessary, routine, and expected result of the human brain working the way it does; there is a causal relationship between mind and body in just the same sense that there is a causal relationship between stomach acid and digestion--digestion simply describes what it is that the stomach acid is doing.

So, why am I not down with the NCC? Individual minds operate differently. A recent study done at MIT illustrates this point quite well: researchers found that individuals with different cultural backgrounds used their brains radically differently to accomplish the same tasks, with totally different areas of the brain being active and yielding the same functional result. There's no reason to suppose that consciousness, which is the result of the complex interplay between the traits measured in this study, would work any differently--in other words, my 'NCC' might be totally different from your 'NCC,' which effectively means that there is no NCC at all.

This is all a very rough sketch, and the result of a good deal of on-the-fly musing on my part. I fully expect and intend to tighten this down over time. In fact, it seems to me that I've got the makings of a pretty good formal philosophy paper here. Hmm....


derekjames said...

I do, however, reject the computationalist notion that the mind is "modular."

Briefly, this theory says something like the following: the best way to understand the mind is to think of it in terms of modules. We have an "arithmetic module" that lets us add and subtract, a "balance module" that lets us stand upright, and a "consciousness module" that's responsible for our subjective and unitary experience of the world. If we "popped out" the consciousness module, we'd be left with p-zombies, creatures that behave as if they are conscious in every way, but actually lack any sort of subjective experience. As I said, this idea does not ring true to me.

How do you account for phenomena such as blindsight, in which there is function without subjective experience?

And modularity is one of those often-misused terms. Do you think the brain is one big distributed network, ala Lashley? How are you using the term "modular", and are you really saying that you don't think there's modularity at any level?

Jon said...

Good questions--hopefully I can give equally good answers.

Blindsight is really interesting. For those who are unfamiliar with the cases, here's a brief summary.

A while back (as in a few decades ago), an experimental new surgery was proposed to help control severe epilepsy in patients who had had no luck with more conventional therapy. This new treatment involved surgically severing the corpus collosum, the bundle of nerve fibers that connects the left hemisphere of the brain to the right hemisphere, allowing the two to cooperate in cognitive tasks. While this surgery was actually quite effective at curbing seizures (it prevented the 'electrical storm' of activity that causes a seizure from spreading across the whole brain), it also led to some rather strange side effects--one of these is the so-called 'blindsight' effect.

Suppose patient with a severed corpus collosum is shown a picture of (say) a cat. If he looks at it normally, both the left hemisphere (which is 'hooked up' to the right eye) and the right hemisphere (which is 'hooked up' to the left eye) are receiving information about the environment, and so performance is normal (more or less). However, if you cover one eye (say, the right one) and present the same picture, something more interesting happens.

Since only the right hemisphere is receiving information, the subject will be unable to report seeing anything at all (verbal processing happens primarily in the left hemisphere). However, if you put a pen in the subject's left hand (right hemisphere), he will be able to draw the object before him quite accurately, despite saying that he has no idea why he's drawing what he is.

My off-the-cuff answer to your question would be this: while the subject might not be conscious of his experience in just the same way that he was prior to the surgery, he most certainly still is conscious of something, and the blindsight experience is certainly like something. We shouldn't really be surprised that the "tone" of consciousness changes when radical changes are made to the neurobiological structure responsible for its creation--I don't think we can honestly claim, though, that blindsight is a case of truly unconscious information processing in the way that computationalists want it to be.

As to your second set of questions, I'm honestly not certain I've researched the modularity thesis enough to give a satisfactory answer; most of my knowledge of the theory comes from discussion with fellow philosophers, rather than formal research (ah, the curse of temporarily not being a student!). I'd love some book recommendations for further reading!

Thanks for the comments!!

Michael said...

Hi Jon,

I'll have to read the Humphrey article I suppose. Having admitted I haven't, I'll leave some comments anyway.

The idea that consciousness is part of the sideshow seems wrongheaded, non-occamish. Where can you point at things that evolved on us that are just along for the ride? Why would the signal most distinguishing feature of us (and I would imagine all higher animals) be the one thing that violates that?

A p-zombie doesn't do what humans do. Can anybody really believe a p-zombie would read books about consciousness and free will and discuss them as though they mattered? I love the fact that you diss Descartes even as you appeal to him, let me do the same except for the dissing part. Without proving it, the world is not a trick. Stuff hasn't been put here to pretend to be one thing while really the world is another.

It seems intuitively appealing, and therefore denied at peril, that consciousness is intimately related with drive. I spent the day today pounding on an information theory problem about the error rates in different transmission systems. I have no reason NOT to think that a lot of the mathematical reasoning parts of my brain are modular, they could go without me going to. But I have NO reason to think I could have made much progress without my consciousness. It seems I drove myself the entire time I was working on the problems.

My recollection is you don't really believe in free will. My own suspicion is that free will and consciousness are part of the same beast. The evolutionary advantage is: conscious animals are a lot more motivated to do clever and difficult things to survive than p-zombies. My laptop shows absolutely no discernable aversion to being destroyed with a sledge hammer. My gerbil, on the other hand, is appears highly motivated to avoid death.

There is a theme in much of the talk on free will and consciousness that seems like it could be eliminated with progress made by doing so. That theme being: things are NOT as they seem. We can theorize that things are REALLY different than they seem, that it feels like we make choices even as we don't, that it feels like things feel like things, even though we only think they feel like things (Humphrey's position?). The starting point of reasoning about the world is most productively to keep the horses in front of the carts, to keep the most straightforward basic perceptions of things as the most basic straightforward indications of how those things are.

I am reminded of the theory that the female orgasm did not exist which I have been told had prominence in 100 years ago in academic circles. I can just imagine thousands of non-academic women and men coming all over the place in London while the poofter dons of Oxford contemplated a world which was academically rich, but which had little connection with reality in many ways. Are we doing this with free will? Is Humphreys doing this with consciousness?

I googled "female orgasm didn't exist" and came up with a bunch of hits. Apparently, people even now are debating it! I am not a female, but I really have not realistic doubt that they DO exist. Just as I have little realistic doubt that you are conscious, and I can't even see you.

Anyway, I hope you enjoy these comments and don't take offense at my colorful language and/or examples. I've thought that one advantage of working on problems in the modern world is less of a consensual blindness or buy-in to some nominally "academic" approach to things, and consequently more of an openness to things that might work even if they come from outside the respectable paradigms. (Did I manage to make talking dirty sound valuable? Then my job here is done.)

Have a good one,

Jon said...


Your comments are thought-provoking and well expressed. I certainly don't care what sorts of examples you appeal to in making your point, so long as those examples make sense in context (which they certainly do). I'm delighted by your appeal to naive realism--that is, your appeal to the intuition that things are exactly as they appear--and certainly agree with this position. I think you raise enough troubling points that your comments deserve a separate post, which I shall be making in the next few days. Thank you!