Books : reviews

Alva Noe.
Action in Perception.
MIT Press. 2004

Alva Noe.
Out of Our Heads: why you are not your brain, and other lessons from the biology of consciousness.
Hill and Wang. 2009

rating : 3.5 : worth reading
review : 19 September 2013

Alva Noë is one of a new breed—part philosopher, part cognitive scientist, part neuroscientist—who are radically altering the study of consciousness by asking difficult questions and pointing out obvious flaws in the current science. In Out of Our Heads, he restates and reexamines the problem of consciousness, and then proposes a startling solution: Do away with the two-hundred-year-old paradigm that places consciousness within the confines of the brain.

Noë suggests that rather than being something that happens inside us, consciousness is something we do. Debunking an outmoded philosophy that holds the scientific study of consciousness captive, Out of Our Heads is a fresh attempt at understanding our minds and how we interact with the world around us.

What is consciousness? Where is it located? How does it arise? Noë tackles these questions by describing how consciousness is something we do, by virtue of being embodied in and interacting with a complex environment. It is not something we are, something internal to us, merely by virtue of having a brain.

p.xii. In a way our problem is that we have been looking for consciousness where it isn’t. We should look for it where it is. Consciousness is not something that happens inside us. It is something we do or make. Better: it is something we achieve. Consciousness is more like dancing than it is like digestion.

Not everything that is important can be reduced to static, physical properties; not everything that is important is a noun. Some important properties are dynamic, and are relations between other entities; some important properties are verbs, are actions, are interactions.

pp3-4. … there’s nothing about this piece of paper in my hand, taken in isolation, that makes it one dollar. It would be ludicrous to search for the physical or molecular correlates of its monetary value. The monetary value … depends on the existence of practices and conventions and institutions. The marks or francs or pesos or lire in your wallet didn’t change physically when, from one day to the next, they ceased to be legal tender. The change was as real as it gets, but it wasn’t a physical change in the money. Maybe consciousness is like money.

[For those wondering what “real change” is being referred to above: it is when many European countries changed their currency to the Euro.]

Here the important dynamic property is the interaction between the brain-body system of the organism, and the environment within which it is embodied.

p10. … to understand consciousness—the fact that we think and feel and that a world shows up for us—we need to look at a larger system of which the brain is only one element. Consciousness is not something the brain achieves on its own. Consciousness requires the joint operation of brain, body, and world. Indeed, consciousness is an achievement of the whole animal in its environmental context. I deny, in short, that you are your brain. But I don’t deny that you have a brain. And I certainly don’t deny that you have a mind. To have a mind, though, requires more than a brain. Brains don’t have minds; people (and other animals) do.

Noë discusses modern advances in brain imaging. Some people say that these show us pictures of thoughts in the brain. Noë explains how this is an over-simplified picture of what is really being measured. It is in some real sense only half the picture. What is happening in the brain is only part of the story, and leaves out the interactions between the brain, body, and world that go along with the brain activity. Noë contends that consciousness requires this interaction, indeed in some sense, is this interaction.

When a model is proposed like this, my automatic scientific reflex is to pick holes in it, to push it to limits, to test it. So, if consciousness requires, or is, interaction, two test cases that immediately spring to mind are: persistent vegetative state, and locked-in syndrome. One would imaging, in Noë’s model, neither of these cases would be considered conscious. Yet we believe that people with locked-in syndrome are conscious – that’s one of the things so horrifying about it – and they confirm that once they are provided with a system that allows communication. Are we to assume that they are confabulating such consciousness when not communicating?

Noë does talk about both these cases, and appears to be saying that both are conscious, to some degree. I fear I didn't really follow the argument at this point, as it seemed to be contradicting the main thesis. These kind of edge cases need to be fully examined and explained within the model.

A different point that Noë does talk about in detail is the “problem” of solipsism: how can we reliably infer that other people have minds, and are conscious, rather than being mere zombies? Here he states there is in fact no problem: we actually start from a position of assuming others have minds, we don’t infer it, because of our embodied consciousness, we are born and develop immersed in the consciousness of others.

But what about those theory of mind experiments? In such, the location of a toy or a sweet is changed in the presence of the child, but not of a third party, and the child has to say where the third party will look. Young children give the new location, and this is used to demonstrate that they do not have a belief in the consciousness of others. Well, these experiments could have a different explanation:

p32. It may very well be right that young children take for granted that they and others share a common world and also common interest in that world, and that this assumption leads them to be unable to fathom that the attention-grabbing pull of the chocolate itself won’t be experienced by others as it is experienced by them. But far from showing that they have no conception of the minds of others, this shows that they have no conception of the minds of others as private and unobservable.

So, it’s not that they don’t understand that the other is conscious, it’s that they don’t (yet) understand that we are all separately conscious. This seems to make more sense. After all, if the child doesn’t think that the other is conscious, why would it think that they would look anywhere at all? Presumably, it’s only because the child thinks that the other shares its own conscious desire for the treat that they would search for it.

So solipsism isn’t a problem to be solved, it is a pathology to be avoided. There is no “view from nowhere” regarding other people: our embodied consciousness requires us to start from the position of the reality of others.

p33. … no sane person can take doubts about other minds seriously. From the standpoint of these collaborative mutual involvements, the problem of other minds can’t arise: not because there is not in fact a detached, theoretical standpoint from which the problem can be raised … but because, crucially, we do not and cannot occupy that standpoint, at least as long as we want to carry on together in cooperation. … Intimacy and commitment sometimes simply leave no room for theoretical musings. I cannot both trust and love you and also wonder whether, in fact, you are alive with thought and feeling …. A certain theoretical detachment is incompatible with our joint mutual commitment.

The move from a physics to a biology viewpoint requires studying the active alive organism embodied in its environment. The very act of drawing the boundary, of identifying an organism, is non-reductionist.

p41. To do biology, we need the resources to take up a nonmechanistic attitude to the organism as an environmentally embedded unity. When we do that … we also secure the (at least) primitive mentality of the organisms. The problem of mind is that of the problem of life. What biology brings into focus is the living being, but where we discern life, we have everything we need to discern mind. … You can’t both acknowledge the existence of the organism and at the same time view it as just a locus of processes or physicochemical mechanisms. And once you see the organism as a unity, as more than just a process, you are, in effect, recognizing its primitive agency, its possession of interests, needs, and point of view. That is, you are recognizing its at least incipient mindfulness.

Again, my reflex is to test the model. All life is (at least incipiently) mindful? Noë seems to mean this. He explicitly talks about bacteria – they are his starting point – and they unarguably have some degree of agency. He then moves on to animals – more obviously mindful – but neglects plants (and fungi). Yet plants are organisms too, and are as much more complex than bacteria as are animals, so in Noë’s model should also have “at least incipient mindfulness”. It is just more difficult to see plants as having agency, since they are not motile. Well, actually, they are: just watch a timelapse video of flowers following the sun, or tendrils wrapping round supports. Our own embodied consciousness works at too fast a timescale to notice the slower vegetative timescales of plants. (Venus flytraps, anyone?)

However, the agency of (most) animals is certainly greater than that of plants. And it is that agency, that interaction with the world, that leads to consciousness. The animal’s brain is only part of the overall interacting system.

p47. … what matters for consciousness is not the neural activity as such but neural activity as embedded in an animal’s larger action and interaction with the world around it. … the brain’s job is that of facilitating a dynamic pattern of interaction among brain, body, and world. Experience is enacted by conscious beings with the help of the world.

We are embodied in our environment from birth, and we grow up into it, adapting to it, adapting it to us. Separating us from it damages us, diminishes us, or at least changes us. (In the case of adaptation to a damaging environment, damaging separation may nevertheless be the best option.)

p51. Maturation is not so much a process of self-individuation and detachment as it is one of growing comfortably into one’s environmental situation. We grow apart, but we attach to the world without. We integrate. In learning to walk or mastering language, in developing friendships, in acquiring an occupation, in learning to navigate and use technology, we root ourselves in the practical environment. This is one reason, certainly, why radical changes to one’s environment, especially occurring later in one’s life—for example, in the course of migrating from one country to another, upon the loss of a spouse, during a period of rapid technological change—are enormous, maybe even devastating personal challenges. The loss of a feature of the environment with which ones daily activities are intimately interwoven is the loss of a part of oneself.

This loss can be particularly bad in the case of elderly people. Moved to a hospital, or into what is euphemistically called a “home”, they lose their familiar, adapted to environment, and can become severely diminished.

One of the things we continually adapt to is our own growing, changing, ageing body. Sudden changes to that can also cause disruption in our habits.

p76. … the loss of a hand does not all at once obliterate the behavioral setting on which your having or believing yourself to have a hand depends. … the absence of your hand is not real until it fails to be at your disposal when you prepare to reach with it or stop your fall with it. A limb is quasi-present as a phantom limb when the behavioral, environment-involving attitudes and engagements outlive the loss of the limb. Only when you fully adapt to your new circumstances—only when you break the habit of acting with and on your hand—will your ghost hand finally be put to rest.

Whether the changes are sudden, or slow, we adapt. And that adaptation changes the way we perceive the world, forming a feedback loop.

p79. As our body schema changes, our relation to the world around us changes, and so how we perceive the environment changes. How big a parking spot looks will be affected by the size of the vehicle you’re driving; how steep a hill looks has been shown to vary, depending on the weight of the pack you are carrying. Indeed, it has been shown that the apparent size of the baseball varies in direct correspondence to the hitter’s batting average. The better you are hitting, the bigger the speeding balls you are trying to hit will seem! When you’re slumping, the balls actually appear to shrink!

This adaptation process allows us to extend ourselves. [I have previously speculated on some potential futuristic consequences of this.] We quite happily extend ourselves using physical tools (sticks, hammers, cars), and mental tools (language, writing, science), so why assume that our consciousness has to be confined within the boundary of our skull?

p80. Where do we stop, and where does the rest of the world begin? What these reflections on the body schema show is that there’s no principled reason even to think that our bodies stop where we think they do. Parts of me—tools—can be spatially discontinuous with me: What makes them me, what makes them part of my body, is the way my actions take them up. And insofar as I act in and feel with my extended body, my mind is extended too.

The particular extension of ourselves that is language depends very heavily on other people. Noë gives some interesting examples of how language is an interactive process. This might sound obvious: we talk to other people, not ourselves (well, not all the time). But here he is talking about the co-creation of the linguistic environment, where the group can do so much more than the individuals alone.

p87. Much of our cognitive lives—vast stretches of our linguistic lives, our cooperative work lives, as with the air traffic controllers—requires not only landmarks and tools (such as language or paper strips) but also other people.

Dispite the potentially dizzy intellectual heights of group-talk, much of what we say and do is nevertheless mundane and habitual, literally “everyday”.

pp97-98. Human beings are creatures of habit. Habits are central to human nature. Roboticists should take heed; they’ve directed their energies to making clever robots—robots that can make chess moves or avoid obstacles. A better goal would be to make robots with habits. My hypothesis: Only a being with habits could have a mind like ours.

Habits are enabled by the relative constancy of the environment. We could not develop habits if the world was different every day. These habits help us get through life with reduced cognitive burden: we can do many things automatically. We have exported this burden to the environment. We are a lot less intellectual than we might like to believe.

p107. One of the very many false ideas about language is that its primary function is to express information or communicate thoughts. Speech has many functions but surely a large put of it is more like the grooming behavior of chimpanzees or the shepherding behavior of dogs than it is like reasoned discourse among parliamentarians. We bark so that our kids get out the door in time to get on their bus and so that they feel safe and loved; we purr so that our colleagues and coworkers know we’re on the job and ready to be called on.

[“reasoned discourse among parliamentarians”? A non-obvious choice of example there!]

One thing that puzzles some researchers is how children perform the amazingly difficult computational task of inferring the rules of grammar. Noë explains how: language isn't arbitrary (and in fact experiments indicate that even language savants have difficulty learning “non-standard” grammars), but was invented by us for our use.

p110. Language may be an immensely complicated symbolic system, but we didn’t just happen upon it: it doesn’t just happen to be that we can figure out (miraculously!) how to use it. We ourselves have built language—collectively, over thousands of years—precisely to be a way of collaborating and communicating that is easy for us.

As might be expected for embodied beings, we are well-adapted to recognising patterns in our environment. Our exceptional face recognition abilities appear to be a special case of this general ability. [FFA = “fusiform face area” in the brain]

p116. … the FFA is activated by nonfacial objects, but only when the objects belong to a class in which the perceiver has expertise. Birds light up the FFA in birders, cars light up the FFA in car buffs, and so on. … it is the fact that … everyone is a face expert that explains why the FFA lights up in everyone for faces; only some of us are birders or car buffs, so only some of us show activation in the FFA for those different sorts of objects.

For Noë, a key factor in our embodiedness is vision. This is an active process. We don’t have passive camera-like eyes. We don’t just see; we look, we move our eyes and head, we engage.

p145. When you go to the theater or a baseball game, you sit up and look around and move your eyes and your head; in this way you engage with the event in front of you. (Indeed, even when you try to be still, your eyes move on their own, making saccades three or four times a second.) Seeing is a kind of coupling with the environment, one that requires attention, energy, and, most of the time, movement.

Noë discusses the case of blind people who learn to “see” using a system that “imprints” the current view through a head-mounted camera onto their skin. They learn very quickly to assimilate the sensation as a form of vision. This requires the camera to be on their head, so that the sensory experience of their environment is correlated in a visual-like way with their own movements in that environment. A “third person” view is no good.

It seems that, for Noë, this embodied vision is a crucial aspect of our consciousness, anchoring and embodying our major sense in the environment. Again, a test spring to mind: if vision really is so crucial for our sense of consciousness, how does this, if at all, affect the consciousness of blind people? Noë however restricts his discussion to kittens who are prevented from seeing when young, and who then never become able to see. (This seems to contradict the experience of the blind people with the skin-seeing device, but Noë does not juxtapose these two examples.)

Instead of discussing blindness, Noë talks about visual experience within dreams, and how it is fundamentally different from vision in reality.

p179. One striking difference between normal perceptual experience and experience in a dream is that the former, in contrast with the latter, is stable. Indeed, according to one dream researcher, Stephen LaBerge, it is a universal feature of dream experiences that detail is never stable across scenes in the dream. For example, if you read a sign in a dream and then, in the dream, turn away and then turn back, the words on the sign will have altered. This should not be surprising, from my perspective at least. After all, in normal perception—in contrast with dreams—we don’t have to do the work of stabilizing detail. The detail is there, in the world. Reality anchors us. Whatever actions we take—shutting our eyes, turning away, getting distracted—things around us remain unaffected. In a dream, however, detail shows up as a feat of creative imagination. The fluidity and shifting grounds of dream experience reflect precisely the fact that in dreams, but not in normal perceptual experience, we are decoupled from the world around us. What determines content in a dream is precisely not what is there in front of us. We can see what we want to see, or what we are afraid to see, or what we wonder what it might be like to see. Which is just another way of saying that dream seeing is not really seeing at all.

[The sign reading example here is different from my own dream reading experiences: whenever I try to read text in a dream, it always annoyingly evaporates. This is one thing that lets me know that I am dreaming.]

Reality is that which, when you stop believing in it, doesn't go away.

-- Philip K. Dick. “How To Build A Universe That Doesn't Fall Apart Two Days Later”, 1978.

This is a key point: we don’t have to do everything, because reality is out there doing a lot for us. Moreover, we can’t do everything, because reality is out there independently doing a lot for us. Our consciousness is entwined with and in this independent reality, and our interactions with it.

p186. If we are to understand consciousness—the fact that we think and feel and that a world shows up for us—we need to turn our backs on the orthodox assumption that consciousness is something that happens inside us, like digestion. It is now clear, as it has not been before, that consciousness, like a work of improvisational music, is achieved in action, by us thanks to our situation in and access to a world we know around us. We are in the world and of it. We are home sweet home.

There’s a lot of thought-provoking stuff here, and Noë bolsters his argument it with a range of good examples. I feel he could have gone further, investigating some of the potentially problematic consequences of his model, to do with locked-in syndrome, plant incipient mindfulness, and blindness, at least. Such will need to be done for the model to become acceptable, and could well reveal deeper insights, but even without that, this is an intriguing approach. I am looking at this from the point of view of embodied robotics, but it should be interesting to anyone interested in consciousness.