Baroness Susan Greenfield is a crossbencher in the House of Lords, Professor of Pharmacology at the University of Oxford and director of the Royal Institution of Great Britain

Susan Greenfield

Sensational minds

--Will the day come when we can run a brain scan or take a blood sample and say, that's a certain type of consciousness at work? Susan Greenfield thinks it will. Here she proposes a new way to look at this most subjective of experiences.

HOW does a wrinkled lump of grey matter weighing little more than a kilogram manage to think, love, dream and feel such widely different sensations as raw pleasure and numbing depression? Philosophers, physicists and computer modellers have been pondering these questions for decades,
wondering how your brain creates your consciousness - your personal inner world of thoughts and feelings.

Thus far, their deliberations have not been entirely fruitful. My own view is that we should put this big question - the "water into wine" problem of how the bump and grind of brain cells translates magically into subjective experience - to one side for the moment, and concentrate on a much less glamorous approach. I think we can try to establish a correlate of consciousness-the particular physical state of the brain that always accompanies a subjective feeling. If we could do so we may at last be ready
to develop a testable model of what happens in the brain when you are conscious.

My suggestion is that the depth of consciousness varies according to the number of brain cells working together at any moment in time. At its most basic level I am proposing that consciousness is synonymous with raw emotions, and at its fullest extent with inner reflection and self-awareness. Consciousness is like a dimmer switch, it grows as brains grow, but it also varies from moment to moment as neurons are coordinated into vast but highly evanescent working assemblies. These assemblies are modified in turn by feedback from the body, and communicate their state to it. Hence, consciousness, in my view, is also a dialogue between the three great control systems in the body: the nervous, hormonal and immune systems.

Soon we may even be able to monitor this dialogue, or at least to measure these assemblies as an index of consciousness, and so perhaps gain a better understanding of what other people or animals are experiencing. Most usefully, this model might also suggest new ways to treat mental illnesses, many of which I see as caused by an inappropriate degree of consciousness at any one time.

The best way to begin to explain consciousness is to draw up a shopping list of the features or properties we expect. Then neuroscientists can go back to their labs and see how the brain could deliver.

First, 1 don't believe we should be looking for one special brain region. Many regions are active while you are awake, but as you become unconscious, they all shut down in a fairly uniform way. When someone has been anaesthetised, there's no one region that lights up or gets extinguished.
There is no single specialised "centre for consciousness".

Secondly, although consciousness comes from more than one brain area, at any one moment you have only one consciousness. The world seems of a piece.
So we can expand the first item on the list to say that while consciousness is distributed all over the brain, somehow the activities of the different regions are coordinated. And if there's no special centre or neurons for consciousness then the neurons and areas that generate it must do other jobs as well. The physical manifestation of consciousness must be something that happens in or to ordinary brain cells at certain times, but not others.

Also on my shopping list is the notion that the more complex the brain the deeper the consciousness. The idea of degrees of consciousness helps answer questions such as when a fetus becomes conscious, and which other animals are conscious. 1 can't see a physical Rubicon when the brain of a
developing fetus changes suddenly, nor any obvious cutoff in the animal kingdom between a nervous system that generates consciousness and one that does not. We should think instead of a continuum: a rat is conscious but not as conscious as a dog; a dog is conscious but not as conscious as a
primate; and so on, Even an ant will have a tiny modicum of consciousness.

If you think of consciousness like this-as something that varies by degree - there are two interesting consequences. The first is that we may be more conscious at some times than at others, hence our experience of states of "heightened awareness", and the conviction that we can "raise" or "deepen" our consciousness. The second, crucial consequence is that we will have finally converted consciousness from a qualitative to a quantitative phenomenon. We can then look for a measure of the depth of our consciousness as it varies from one moment to the next, and search the brain for something that contracts or expands with it. 1 think that the most logical place to look is in very large networks - " assemblies " - of brain cells.

You're born with pretty much all the brain cells you'll ever have, but as you mature these cells develop more interconnecting branches. Our brains are incredibly plastic, and these connections grow and change with every experience. Babies evaluate the world in purely sensory terms-how sweet, how fast, how cold, how loud. But gradually these abstract sensations coalesce into people and objects with meaning and associations. It's these personal connections and associations that 1 think of as the "mind". The mind is your personalised brain, which allows you to see the world in terms of what you have experienced already. Even if you're a clonethat is, an identical twin-your mind will be unique. You see the world in terms of things that have happened to you alone.

If we see a familiar person, our visual system activates a "hub" of brain cells that corresponds not only to the shapes, movements and colours of a face, but to all the associations set up in our mind by our experiences of that person. That can all happen without our being aware of it.
Consciousness, 1 believe, is generated as this active, hard-wired hub corrals huge numbers of other brain cells around it to form a vast working assembly that lasts for just a trice. The image I have is like throwing a stone into a puddle, producing ripples of consciousness.

We now know the brain to be capable of forming such highly transient assemblies. Amiram Grinvald at the Weizmann Institute in Rehovot, Israel, has shown that in response to a flash of light, as many as 10 million brain cells become active together, coordinated into a working assembly that lasts for less than a quarter of a second - exactly the space and time scales 1 think we should be exploring.

The assembly will be slightly different every time. Partly it will depend upon the size and strength of the stimulation of the hub, but also on the levels of a variety of chemical messengers - neurotransmitters which change moment by moment. These transmitters "modulate" the activity of large groups of cells and mediate arousal levels, your sleep-wake cycle and your dreaming. In physiological terms, these put cells on "red alert" - they can predispose brain cells to be recruited into the working assembly,
triggering lots of covert associations.

I think it is the activity of these transient neuronal assemblies that correlates with the depth of your consciousness at any one moment. To test the model, let's take some examples of the different types of assemblies formed and see how they relate to different types of consciousness.

One time you'd expect to see unusually small cell assemblies would be when you didn't have much connectivity in the first place, as in a young child's brain. What do we know about an infant's consciousness? One feature is that their centre of attention varies depending entirely on the sensory quality of what they're seeing. They live in the press of the moment, in a rather abstract world with little meaning, reacting to everything in a simple, emotional way. Infants are like little sensory sponges: they lack any accumulated experience with which to interpret the world. They haven't yet forged multiple connections - they haven't yet developed a "mind". Each burst of brain activity will come from only a small hub of cells, which will create small, short-lived ripples of consciousness.

This is, I believe, the most primitive kind of consciousness we have, with a small assembly associated with strong emotions and an immature mind. So my own view is that emotions are the building blocks of consciousness, and that you can't have consciousness without some sort of emotion. That's why I for one don't put much of a premium on computer models of consciousness: such models focus on tasks such as learning and memory, which an ordinary PC can do without subjective inner states.

There are times when adults too have diminished consciousness. You would have small assemblies, as in childhood, when you're dreaming. However, the reason would be different. In this case you have no strong sensory input, so there's little to stimulate the neuronal hubs, and you're dependent on internal residual neuronal activity. This perhaps explains why dreams have a disconnected, flimsy narrative. At the time they seem very real, with high emotional content, but in retrospect we wake up and judge our dreams as irrational with the checks and balances of our cognitive adult minds.

We can chemically alter our level of consciousness, too. So a third situation in which you might have a small assembly would be if the work of the brain's chemical messengers was disrupted, affecting the ease with which the working assemblies formed. Taking drugs such as ecstasy can interfere with one such chemical, serotonin. And in schizophrenia, levels of another messenger, dopamine, are effectively in excess. In both cases the ease with which assemblies form would change, the net size would be smaller and consciousness would seem childlike or dreamy. People may take
the world at face value, see it in sensory terms and display flimsy logic.

Another time you would find only small assemblies is when you are in a rapidly changing environment with such competition that the assemblies don't have a chance to form properly. Fast-paced sports like white-water rafting, bungee-jumping or skiing would do it, as would a rave.

The opposite of such states would be a large cell assembly, where one would expect the outside world to seem remote. Your senses would be reduced, you might feel emotionally numb, yet extremely self-conscious. You would have a highly logical, perhaps persistent train of thought. These symptoms often occur in clinical depression. Perhaps depression is due to the malfunctioning of the chemical modulators, resulting in overly large assemblies. We know the drug Prozac and related agents influence those chemicals.

Although my theory seems to predict what to expect in different types of consciousness, the assemblies of neurons I'm positing do not all on their own generate consciousness. Assemblies are merely an index - a correlate of your prevailing inner state. Something else must happen. I believe assemblies report to the rest of the body, and the rest of the body reports back to them, and this iteration somehow translates into subjective consciousness.

Neuroscientists, at their peril, often ignore the fact that the brain is in a body. We know that feedback from the rest of the bodymost noticeably the immune system and hormones - can influence our state of mind, and similarly our state of mind can influence other control systems like our immune status. And we know that the nervous, endocrine and immune systems are interlinked. I think the links must be chemical, and for my money peptides are very good candidates. These substances coexist with traditional transmitters, but are only released under special circumstances, as neurons become more active. There are many different peptides, so you would never have exactly the same amounts or combinations twice. Moreover, we know that peptides can interface with the immune, nervous and endocrine systems: some peptides are also hormones, and this puts them in a good position to be, if you like, trilingual.

The way I see it is that at any one moment, transiently formed cell assemblies would release a signature profile of peptides into the body.
These peptides influence the endocrine and immune systems, and in return the systems would release peptides that would determine the size of the brain assemblies. That iteration of peptides between the three great control systems of the body is, in my view, what happens when you are conscious. One day it may be possible to test this hypothesis, by recording profiles of peptide availability in the blood and trying to correlate these with the prevailing state of consciousness.

I think at a clinical level this exercise would be useful. It might suggest new ways of treating conditions such as depression with novel types of drugs, or of developing non-drug treatments that might drive the formation of a certain size of assembly and alter the type of consciousness in a beneficial way.

How this all translates into the elusive subjective inner state of consciousness is a completely different question, and I'm not pretending to have answered it. On the other hand, I do think that we can use this model in the future to design experiments and help us understand depression and emotions, why people take drugs, and perhaps most mystifyingly, why people go bungee-jumping.

Baroness Susan Greenfield is a crossbencher in the House of Lords, Professor of Pharmacology at the University of Oxford and director of the Royal Institution of Great Britain

Further reading: The Private Life of the Brain by Susan Greenfield is published by Penguin (2000)