Friday, April 28, 2006

Consciousness vs action and memory

on this comment we talked about the distinction between consciousness and desicion making and memory. Koch claims that much desicion making occurs without consciousness.

Is the other way around also true? Could you be conscious of something but not be able to remember it (right away) or act upon it? (I am not talking about an impaired motor system, but a motor system that is fine, just that the output of the NCC is blocked). I am not claiming that it is not possible, but it sounds to me that it is unlikely (at least for me it is impossible to imagine). Most likely, the concept behind the NCC as an independent 'perceptual' system that feeds to the desicion making apparatus is the erroneous one, and an integrative view of the brain would be better.

Explicit Representation

In page 33 Koch says:

"Underlying every direct and conscious perception is an explicit representation whose neurons fire in some special manner."

I am a little confused with the explicit representation idea. How explicit are we talking? Infinitely explicit? That is, it only responds in the whole world to the represented object/concept? Or kind of explicit? That is, it responds to the represented object more robustly than to other objects? What about small assembles of cells that can represent something unambiguously (like the examples he talks about, such as motion perception in MT or face recognition in IT)? Do those count as an explicit representation?

Columnar Organization?

On page 32 the first paragraph says:

"Those attributes that are made explicit -- that is, for which there exists a columnar organization -- are the ones that are sufficient, under the right circumstances, for conscoius perception."

Is he claiming that columnarly organized representations are the only ones that are sufficient for conscious perception? Does he think that the Bill Clinton cell is in a Bill Clinton column?

He had claimed that columns were important before, but I thought that he would eventually explain why he is giving so much importance to it (maybe he will later). But at the moment, there is no reason to believe that columnar organization has any relationship to conscious perception, or that explicitness is correlated to columnar-ness (nice word ah?).

Thursday, April 27, 2006

2.1 The machinery of the cerebral cortex (3)

What is meant by Explicit and Implicit?

"An explicit representation is one that has more logical depth than an implicit one because it is, in essence, the summation of all the implicit information."

Things to note. Here Koch is careful enough to use the concept of explicit and implicit in the context of representation, and not necessarily of cells' representations. Even though all his examples are representations by single cells. Here the problem is, that if we continue his claim, the logical outcome is that there must be an explicit representation of the whole field of consciousness at some level. And we all know that this is not going to happen at the single cell level.

Something that I would have liked Koch to discuss is what exactly has to be represented. It seems as though in all the examples he presents, the cells do in fact have explicit representations of some things, but at the same time implicit representation of others. As I said before, if there was indeed a pure explicit representation of consciousness it would have to be of the whole field of consciousness.
For example, he shows in figure 2.4 a cell in IT of monkeys that responds preferentially when the monkey views a side views of bearded people. This cell might be representing explicitly 'bearded man seen from the side' but at the same time it might form part of an implicit representation of 'uncle Joe' who has a beard.

2.1 The machinery of the cerebral cortex (2)

Elections as a metaphor for neuronal competition

All throughout this section Koch makes me feel as if he is conflating the concept of attention with the concept of consciousness. He said previously that he thinks that attention and consciousness are separate phenomena, but then he seems to put them in the same bag. It's confusing :S

2.1 The machinery of the cerebral cortex

Here Koch introduces us to the concept of a coalition (an array of active cells that represent something, such as an object). He says:

"The NCC are closely related to this suppresion of competing cellular assemblies, representing alternative interpretations of the scene. Usually only a single coalition survives -- the one whose properties you are then conscious of. Under some conditions -- when the neural representations don't overlap -- two or three coalitions may coexist peacefully, at least for a while."

Here it seems to be as if he was talking about a NCC for 'scene perception' rather than 'attribute perception' (unless you are experiencing binocular rivalry, but that is not usually the case). Because, why else would only one (or a few) coalitions survive? I would expect many many coalitions surviving at the same time all the time if it was NCC of attibutes as he seems to imply elsewhere.

Tuesday, April 25, 2006

1.5 Recapitulation (page 19)

In the second paragraph Koch says:

"My goal is to identify the specific nature of this activity, the NCC, and to determine to what extent the NCC differ from activity that influence behavior without engaging consciousness"

Unconscious perception research goes to the heart of the definition of consciousness. Something is unconsciously perceived if the subject performs better than chance in a 2-altenative forced choice task, even though he claims that he was guessing (something like blindsight in normal subjects). So with this definition, together with the cited extract, we can see that maybe Koch is claiming that being conscious is the same as being able to verbally report it? Or is it that the verbal machinery is conected only to the NCC of everything, and not conected to anything else?

1.4 The neuronal correlates of consciousness (3)

In page 18, the 4th paragraph says:

"Eventually, a theory that bridges the explanatory gap, that explains why activity in a subset of neurons is the basis of (or, perhaps, is identical to) some particular feeling, is required."

Isn't this the whole issue? I thought that's what this book was trying to address... now I guess we'll have to find a new purpose for the book.

side comment

There are pleasure systems in the brain and pain related systems in the brain,
What would happen if I stimulate both at the same time? Would the animal feel a pleasurable pain? Or a painfull pleasure? Would one overcome the other? Would the behavior seek this stimulation while knowing that it is bad? (like a drug)

1.4 The neuronal correlates of consciousness (2)

In page 18 paragraph 2 Koch says:

"... the NCC activity must affect other neurons in some manner. This post-NCC activity influences other neurons that ultimately cause some behavior. This activity can also feed back to the NCC neurons and to previous stages in the hierarchy, significantly complicating matters"

Something bothers me about this way of thinking. It is too static. There is an NCC, and the NCC does something (generates consciousness and affects other non-NCC cells). Reaching the NCC is an unambiguous state. You either are there or not. No questions asked. When we know from many lines of research that this is not so true (unconscious perception, near-threshold perception, attention research, etc).

On the side, the issue of feed-back is a fascinating one. We have very little knowledge regarding the role of these massive projections.
As an example, let's say I am looking at an apple on a black room. Let's say that the NCC for the apple lights up. So now I am conscious of the apple. Then this NCC projects all over the place (to frontal, parietal, occipital and temporal lobe structures) including the cells in lower visual cortical areas that generated in the first place this NCC activity. This means that the actual NCC of the apple is going to change slightly on the next 'consciousness wave' (post-feed-back). This change might have an effect on how we perceive the apple or no effect at all possibly. If it does have an effect, what would be the mental effect? Maybe getting distracted and thinking about the movie you're watching tonight? Maybe seeing slight changes in the image? Maybe paying more attention to some aspect of the apple? Can you ever have one phenomenal state for more than a few milliseconds?

Monday, April 24, 2006

1.4 The neuronal correlates of consciousness

Here we get our first definition of Neuronal Correlates of Consciousness of NCC.

"(NCC) are the minimal set of neuronal events and mechanisms joingly sufficient for a specific conscious percept"

For an in depth analysis of this definition check David Chalmer's article on the topic.

Once again we stumble upon the limitation on the definition of consciousness. Some of the difficulties that will arise are;

- Is there a NCC for each visual attribute? or for a whole scene? For example, when I see a tree with an apple hanging, should I be looking for the NCC of red, green, apple, tree, sky, ant in the grass, butterfly flying by? or should I just be looking for the NCC of the whole scene?

- Furthermore, should I be separating visual consciousness from the rest of the conscious field? John Searle (article) argues that we should not. That the right way of approaching the problem is to understand the conscious field, and then understand how sensory experiences change this conscious field. Not how they create it.

- Is there a NCC for a concept? or for raw visual attributes? For example, in this picture should we be looking for the NCC of the dark and white contours? Or should we be looking for the NCC of the dalmatian? Before you 'see' the dog, you can certainly see the visual features that constitute the dog, so something must change when you identify it as being a dog. Is that an NCC also?

-How do I know I am conscious of something? Because I feel it, you would say. But if I feel it, I can talk about it. I can change my behavior. Is that necessary to be conscious of something? Can I be conscious without it affecting my behavior somehow? (note that not-acting on something perceived is a behavior as well). If the answer is no, then should we be looking for the NCC of conscious perception? Or conscious decision making and motor act? Are they separable?

-If the time we exprience as continuous is in fact illusory, what happens with the neural activity correlated with it? For a NCC to exist we need a temporaly coherent conscious state. But we know it is not (change blindness, illusory motion, flash-lag, etc). The reason why we believe that time goes by smoothly and linearly is because we feel it that way. But just bear in mind William James hillarious and right-on-target commentary about introspection. He says it is like “trying to turn up the gas quickly enough to see how the darkness looks."

- Can we ever generalize a conscious state? Is it possible that my current conscious state can be replicated? Hillary Putnam claims that each conscious state is unique and unreplicable. Can we then find a NCC? Sure, but it is not going to be useful in the future, because is a NCC for a conscious state long gone and never to come back.

1.3 My approach is a pragmatic, empirical one (2)

How can consciousness be approached in a scientific manner?

Here Koch tells us that he is going to focus on the study of vision as a more experiment-friendly way of approaching consciousness. He says:

"Underlining my choice is the tentative assumption that all the different aspects of consciousness (smell, pain, vision, self-consciousness, the feeling of willing an action, of being angry and so on) employ one or perhaps a few common mechanisms."

I am not sure how I feel about this assumption. It made me think about the kind of explanation that would satisfy people like me. A mechanistic explanation about how we make a desicion based on reward, expectations, sensations would be fine for me. We could perfectly well build a robot that made desicions based on these factors. And I would be willing to admit that however it 'feels like' to make a desicion, this feeling is shared by the robot. But on the other hand, sensory feelings seem to be a different story. With a robot that senses light is not as simple to admit that it has the same feeling as I do when I sense light.
So to assume that vision and the feel of willing action share common mechanisms might be to go a little bit too far.

Sunday, April 23, 2006

1.3 My approach is a pragmatic, empirical one

A working definition

Here Koch tells us why he didn't define consciousness. He says that a formal definition of consciousness "is likely to be either misleading or verly restrictive, or both". Fair enough. So he gives an operational definition; "consciousness is needed for nonroutine tasks that require retention of information over seconds". Hmm... it smells like a conflictive operational definition. I don't want to judge it before we see what he'll do with it, but it seems to me that it can have lots of caveats. Such as; when I do a routine task I'm not conscious?; what if I forget what I was conscious of before a second passes by?; what if I retain the information for several seconds affecting my behavior but can't recall it verbally?.

1.2 A spectrum of answers (part2)

Consciousness is an emergent property of certain biological systems

Here Koch puts forward a compelling analogy with the pre-DNA era and heredity. He cites William Bateson saying:

"The properties of living things are in some way attached to a material basis, perhaps in some special degree to nuclear chromatin, and yet it is unconceivable that particles of chromatin or of any other substance, however complex, can possess those powers which must be assigned to our factros or gens. The supposition that particles of chromatin, indistinguishable from each other and indeed almost homogeneous under any known test, can by their material nature confer all the properties of life surpasses the range of even the most convinced materialism."

If we translate this into the mind-body problem, I guess Koch is suggesting that in a few years they might site someone as saying:

"The properties of the mind are in some way attached to a material basis, perhaps in some special degree to neuronal activity, and yet it is unconceivable that neuronal activity or any other activity, however complex, can possess those powers which must be assigned to our minds. The supposition that neuronal activity can by their material nature confer all the properties of the mind surpasses the range of even the most convinced materialism."

I think this analogy here is a little bit missleading for one simple reason. In the heredity debate, the form of the answer was already known, whereas in the mind-body debate we don't know what or how the answer could potentially look like. In other words, had I asked the incredulous Bateson how would an answer look like, he might have said: "Well, unlikely as it is, you would have to find a mechanism by which to store lots of information in molecules. " What would the mind-body skeptic say?

Saturday, April 22, 2006

1.2 A Spectrum of answers

Here Koch goes over several different approaches and explanations regarding the consciousness problem.

Some of them are:

Consciousness is illusory
Here he is talking about the theories proposed by Dan Dennett. In page 6 Koch says: "(Dennett) argues that consciousness as most people conceive of it is an elaborate illusion". And then "Given the centrality of subjective feelings to everyday life, it would require extraordinary factual evidence before concluding that qualia and feelings are illusory".
It is important to point out the "as most people conceive of it" in Dennett's proposal. He is not saying that subjective feelings don't exist. He is saying that they do exist, only that they are dispositions rather than magical 'quales'. It seems to me that Koch's own way of seeing qualia (that we will analyze in subsequent chapters) is much much closer to Dennett's point of view than he acknowledges.

Consciousness requires behavior
This is a very interesting proposal. O'Regan and Noe propose that: "The behaving organism embedded in a particular environment is what generates feelings".
Koch doesn't like this account, and I don't really like it either in the way O'Regan a Noe propose it. But it is clear that there is something to this way of seeing the problem. They point out that neuronal activity doesn't have a tag that says "I am visual" or "I am auditory" and that there aren't a priory reasons to feel activity coming from the eyes as visual and not some other way. So, sure enough, there must be a developmental interplay between sensory, motor and reward related activity to generate what we call today a visual perception.
This is indeed interesting and mind-opening, but their account of phenomenal experience didn't sound convincing. Koch points out that while dreaming we have phenomenal states that don't relate directly with the external world. In this article they directly address this point, in fact not answering it at all! Here is the section where they describe this:

"It is often claimed that dreaming, or other types of mental imagery, provide a counterexample to our denial that the brain must represent what is seen. Since dreams and mental images are apparently pictorial in nature, this seems to show that we are, after all, capable of creating an internal iconic image. Penfield's classic observations (e.g. Penfield & Jasper [1954]) of visual memories being created by stimulation of visual cortex might also be thought to indicate that there are internal pictorial representations.
It is easy to be misled by these arguments, which for some reason are peculiarly compelling. But it is important to appreciate that they are misleading. Whether dreams, hallucination, or normal vision are at stake, these arguments are another instance of the error of thinking that when we see things as picture-like (be it when we look at reality or when we have a dream), this must be because there is some kind of internal picture. But this is as misguided as the supposition that to see red, there must be red neurons in the brain. The supposed fact that things appear pictorial to us in no way requires there to be pictures in the head. Therefore the fact that we dream, hallucinate and imagine does not provide evidence in favor of the view that brain contains pictures of the detailed environment
A corollary of this confusion about dreams and mental imagery is the idea, expressed by a number of authors (e.g. Zeki [1993], Kosslyn [1994], Farah [1989]) that feedback from higher brain areas into the retinotopic cortical map of area V1 would be a good way of creating mental imagery. This argument is somewhat misleading. It could be taken to be based on the implicit assumption that mental imagery occurs because of activation in V1: the topographic, metric layout of V1 would make it a good candidate for the cortical areas that possess what Zeki [1993] has called an "experiential" quality -- i.e. the capacity to generate experience. But again, the metric quality of V1 cannot in any way be the cause for the metric quality of our experience. It is as though in order to generate letters on one's screen, the computer had to have little letters floating around in its electronics somewhere.
There may also be a second confusion at work in the argument from dreaming that we are considering. We have already noted that from the fact that dreams are pictorial, it does not follow that, when we dream, there are pictures in the head. But do we really have reason to believe that dreams are pictorial? People certainly do say that they are. But does this give us reason to believe it is so? Just as we have observed that the idea that seeing is pictorial reflects a kind of naïve phenomenology of vision, it may very well be that the claim that dreaming is pictorial is similarly ill-founded phenomenologically. Certainly it is not the case that when we dream, it is as if we were looking at pictures. A hallmark of dream-like experiences is the unstable and seemingly random character of dreamt detail. For example, the writing on the card is different every time you look at it in the dream[16]. This suggests that without the world to serve as its own external model, the visual system lacks the resources to hold an experienced world steady."

Ok, but where in this explanation do they address the fact that the dreamer is indeed having a visual experience, and that this experience is not directly related to anything in the real world? Is there anything, apart from neuronal activity, required for this particular experience(call it visual or not)? Ok, so you need a real world to mold the brain in a way that visual experiences are possible, but that doesn't mean that at this very moment my visual experience is generated by the brain AND the external world. They seem to be mixing up the development of sensation with the mechanisms of sensation.
Maybe I am missing something in the explanation, but at the very least they don't make it clear.

Friday, April 21, 2006

1.1 What needs to be explained

Definition of Consciousness

Anyone that reads this section will realize a problem. There is no definition of consciousness. The book is about something that is not defined. If we browse to the end of the book, page 332 offers a definition of consciousness. It says,

"Consciousness: What this book is all about. At this early point in the scientific exploration of this phenomenon, it is difficult to define it rigorously.... " and then is goes on giving some characteristics of consciousness.

So what needs to be explained?
"what is the relation between the conscious mind and the electro-chemical interactions in the body that give rise to it" whatever the "conscious mind" happens to be.

Sensory consciousness vs desicion making and memory

Then we go into a bit of detail on what the conscious mind is. "Sensory qualities, (are) the building blocks of conscious experience". And by sensory qualities he refers to qualia.

But then, in page 3 things get a little bit more complicated when the concept of unconscious processes makes its appearance. For example:

"... the point of training is to teach your body to quickly execute a complex series of movements -- returning a serve, evading a punch, or tying showlaces -- without thinking about it." and
"Much high-level decision-making and creativity occurs without conscious thought..."

Is tying my shoelaces not conscious? Sure enough I am not thinking about every movement, but I am certainly conscious that I am tying my shoelaces. So apparently now we added to the bag of "consciousness definition" memory and decision-making, besides qualia.

The quest for consciousness

This blog is going to be dedicated to discuss topics and ideas that emerge while I read "The quest for consciousness" by Cristof Koch.
These are not restricted by ideas proposed in the book, but also ideas evoked by the book.

Please feel free to joing the discussion (or grab the book and propose new ideas)