The Electric Brain
How does a three-pound mass of wet gray tissue (the brain) succeed in representing the external world so beautifully? In this interview with noted neuroscientist Rodolfo Lliní¡s of the New York University School of Medicine, find out how the rhythm of electrical oscillations in the brain gives rise to consciousness, and how failures in this rhythm can lead to a variety of brain disorders.
Why brains are important
NOVA: Let's start by talking about why one needs a nervous system—or a brain—in the first place.
Rodolfo Llinás: That's a very intriguing issue. The nervous system is about 550 million years old, and it first came about when cells decided to make animals. Basically there are two types of animals: animals, and animals that have no brains; they are called plants. They don't need a nervous system because they don't move actively, they don't pull up their roots and run in a forest fire! Anything that moves actively requires a nervous system; otherwise it would come to a quick death.
Why would it die if it didn't have a nervous system?
Because if you move, the variety of environments that you find is very large. So if you happen to be a plant you have to worry only about the very small space you grow into. You don't have to do anything other than maybe move up and down. And you're following the sun anyhow, so there is no planned movement, and therefore there is no necessity to predict
what is going to happen if, which is what the nervous system seems to be about. It seems to be about moving in a more or less intelligent way. The more elaborate the system, the more intelligent the movement.
So you need a nervous system in order to be able to predict the future?
Yes, and in order to predict you have to have, at the very least, a simple image inside that tells you something about the purpose of the outside world. That is common to all nervous systems of all forms that we know of. Each animal has a different universe—the universe it sees, the universe it feels, the universe it tastes. Earth probably looks very different not only for all of us as individual humans, but also for different animals.
"We assume we have free will, but we don't"
How does consciousness come into this view of the brain? Is consciousness a mysterious phenomenon, in your opinion?
I don't think so. I think consciousness is the sum of perceptions, which you must put together as a single event. I seriously believe that consciousness does not belong only to humans; it belongs to probably all forms of life that have a nervous system. The issue is the level of consciousness. Maybe in the very primitive animals, in which cells did not have a single systemic property—in which each cell was a little island, if you wish—there may not have been consciousness, just primitive sensation, or irritability, and primitive movement. But as soon as cells talked to one another there would be a consensus. This is basically what consciousness is about—putting all this relevant stuff there is outside one's head inside, making an image with it, and deciding what to do. In order to make a decision you have to have a consensus.
But it all just boils down to cells talking to one another?
Some people believe we are something beyond neurons, but of course we are not. We are just the sum total of the activity of neurons. We assume that we have free will and that we make decisions, but we don't. Neurons do. We decide that this sum total driving us is a decision we have made for ourselves. But it is not.
So this mass of wet gray tissue that is our brain is made up of neurons?
The brain is made out of cells. It is a long and very distinguished group of cells—about 550 million years or so old. These cells have a small mass. Our brain is about one-and-a-half liters, or three pounds, but it has 1010 cells, which is a huge number of cells. Ten billion cells. And each cell has 1,000 to 10,000 or so synapses—the connections between the cells. So the brain has trillions of synapses.
How does the brain keep all these different neurons communicating in synch?
Neurons like one another very much. They respond to one another's messages, so they basically chat all day, like people do in society. "Where can I park?" "How much is it going to cost?" "Am I going to get a ticket?" One set of neurons talks to another set of neurons, and they talk back, so we have a dialogue between different components in the brain. And the dialogue is not between one cell and another cell, but rather between many cells and many other cells. It's like having a huge number of people holding hands, dancing together, making ever-changing circles and organized together in such a way that every cell belongs, at some time, to some circle. It's like a huge square dance. Each dancer belongs to a particular movement at a particular time.
And there's music that keeps them all dancing together?
Right. It's generated by the neurons themselves. Neurons have an intrinsic rhythm, a bit like a hum. They generate this electrical dance at a given frequency because they have similar rhythms—they hum in unison. But as in the case of choirs and dancing, you can have two groups doing different things at the same time. Now imagine that each group doing something represents an aspect of an external event, like a color.
That's the brain's job—to represent an external event?
Right. Imagine many cells making an activity circle, with electrical activity going around and around like a windmill. Imagine that out of this circle come a few cells into the center and perform a particular dance that the other cells see but are not necessarily part of. Now these cells that do this particular dance may be cells that have learned something from the outside that they want to put in the context of whatever else is happening in the nervous system. The brain, when awake, is continually generating a picture of the outside world. When new information from the outside comes in it has to be put into context with whatever else was happening just before.
"The brain is the ultimate organ. It can make a reality. It can dream it."
Is there ever nothing going on—no dance—inside the brain?
Sometimes there is nothing as far as consciousness going on, like when you fall asleep. At that time you are not generated. That particular dance, that is you, is not being created by the brain at the moment. If you are asleep but dreaming then the brain makes another dance during which you exist but you don't care about the external world. You can exist on your own with dreams, hallucinations, or deep thoughts, or you can relate to the outside world. Normal people want to relate to the outside world. If you happen to be a schizophrenic you may not. You may want to hallucinate somewhere in a corner. You are consumed by your thoughts. You are fascinated by whatever is inside your head.
What part of the brain does this coordinating? We've all heard of different areas. Frontal lobes do this, another area does that....
If we look at the nervous system there are basically two functions. One is sensory—the ability to respond to the outside world—and the other is the ability to do something about it, the ability to modify the world. As the nervous system gets more complex in higher animals there's another totally astounding property, which is the ability of the nervous system to invent things inside the head, which it can then make into reality. All my life, even as a child, I have been amazed that you can think of something that doesn't exist and then by using the motor system—painting, talking, constructing—you can make them be part of the external world.
Make a pie, for example.
Make a pie that didn't exist before. So it gives you an idea of the unbelievable ability of the system. Not only can it see or move, but it can also make a reality. It can dream it. From that point of view, it really is the ultimate organ. Each part of the brain has a particular function with respect to the nervous system. The visual cortex has one function, the frontal lobes have another function, the auditory system has yet another function. And yet when we look at the external world we see things as having properties that are inseparable from the object itself.
Can you give us an example?
Imagine I have a little bird on my hand. I can see the bird. I can see its color. I can see its shape. I can hear it sing. I can feel its weight on my hand. It might peck me. All of these things occur simultaneously, so we say that the bird has those properties. But all those properties are put together in different parts of the brain. So one wonders how the brain makes a collage of all these sensory inputs to generate one single precept—the bird—out of all the different sensory systems activated. This is called the binding property. Since we don't know for sure how it works, we call it the binding problem.
We are not sure exactly how it happens, but there are good ideas about how it may happen. One of the ways we can attempt to understand how it happens is by studying people who have mental or neurological problems. Someone with a lesion on his or her visual cortex would be able to hear and feel and move but would not be able to see, so you know that he or she injured the conscious component that sees but not the other conscious components. So consciousness has parts.
Consciousness interrupted
How else can consciousness get damaged?
The other thing that can happen is that deep in the brain there is a structure called the thalamus. If the thalamus is damaged—and this is a central entity of the brain, some sort of gateway into the brain—if the gate is damaged then you have the same problems that you have with cortical damage. If the part of the thalamus that connects to the visual cortex is damaged then you don't see.
So externally they'll look like the same problem, or for the person they will feel like the same problem?
Absolutely the same. Even a neurologist cannot tell the difference, so he or she will have to do an MRI to see where the lesion is, either cortical or thalamic. So this central nucleus, the thalamus, and the cortex seem to be very deeply related. Consciousness can be damaged at the cortex, at the thalamus, or at the connections between the thalamus and the cortex. These two levels are organized as a feedback system in which the thalamus speaks to the cortex and vice versa in a dialogue.
If the thalamus and the cortex weren't able to "bind" all these different sensory inputs, what would the world look like?
It wouldn't look like anything. It would be a hodgepodge of things, like when people drink too much and get into problems of timing and cannot move properly or think properly and see things that might not be the way things actually are. You can also have binding problems in neurology. A German colleague told me about a patient diagnosed as having a psychiatric condition. When they did a brain scan they found that she had a bilateral lesion on the temporal lobe. The problem was that she couldn't see things that were in motion, so things were continually appearing and disappearing. If she had to cross the street she would be terrified because she wouldn't see cars coming.
Until they stopped.
Yes, but then they just appeared, and where did they appear from? So she had a problem of binding movement with objects. There are issues of binding that are less complicated. For instance, some dyslexic children have binding problems and so their timing is slightly off, they think a little slower. To them everyday events happen a bit too fast for them, a bit like action movies, so they have problems following fast events or reading or even hearing the sounds of words properly.
This goes back to the concept of the brain's need for rhythm?
You need rhythm to put everything together. It is like dancing. Rhythm tells you when to move and at what speed to move, whether you are dancing a waltz or doing the cha-cha. Likewise the brain has a rhythm. In order to bind things many parts of the brain must fire and be active together, and they must have a rhythmicity. By having this rhythmicity—the thalamo-cortical rhythm—they make one function into an event.
"Consciousness is soluble in a local anaesthetic or even in alcohol."
Is the binding, this rhythm, where consciousness comes from?
Yes. Binding allows the different parts to be transformed into one cognitive experience. Now, interestingly enough, all of these things can be dissolved with a local anaesthetic. So if we put a local anaesthetic in someone's visual cortex, although we are not damaging the brain or changing the synapses or the neurons, the neurons are now not capable of having electrical activity. That function disappears. So consciousness is soluble in a local anaesthetic or even in alcohol.
How is this thalamo-cortical rhythm generated?
The cells in the brain, like the heart, have intrinsic rhythm. They move, they oscillate, like the waves in the ocean at a certain speed, a certain velocity, a certain frequency. Cells can have different frequencies. They can oscillate very slowly, and when that happens consciousness disappears. You are asleep, deep asleep; you are not dreaming. Cells are negatively charged; they are negative with respect to the outside world, by about 60 to 70 millivolts. If the membrane potential, the voltage across the membrane of the cells, is modified so it becomes a little less negative, the rhythm of the cells changes. They wake up. They oscillate at a high frequency.
When you are asleep, how many times per second do the cells oscillate?
When we are asleep most neurons oscillate at a very slow rhythm. That's how cells control consciousness—by voltage, by becoming electrically inactive or active. If you take a drug, if you fall asleep, if you get hit in the head, the brain does not generate the functional state that is you. The you disappears.
Your image of yourself disappears.
Yes, but to you that's all that being you is! I tell my students when they fall asleep in my lecture that they disappear. Their body is there but they're not there. As someone gently elbows them on the ribs they reappear. This is known as waking up.
Insights into cures
If the thalamus is so important in maintaining consciousness, can you trace specific psychiatric or neurological disorders to problems in the thalamus?
This is a beautiful story that is just beginning to be understood now because it is such a different point of view. Because of these rhythms, which are known to be related to the sleep-waking cycle and which we have been studying since the beginning of the last century, we ask certain questions. I ask, for example, isn't it incredible that we fall asleep as a single entity and then wake up as a single entity? It would be a tragedy if you were to wake up and you could see but you couldn't hear. The system seems to be able to shut itself off as a whole and turn itself on as a whole.
What would happen if for some reason some part just stayed off?
If you look at Parkinson's patients, they have a great deal of difficulty moving. They are almost paralyzed—they walk with very small steps, have very little facial expression—and they have tremors. It turns out that their thalamic cell properties make these cells oscillate at a low frequency—four to six times per second—even when they are awake. One part of the brain is asleep while the rest of the brain is awake. That is a problem. These neurons are not dancing the right way. There is friction between neighboring areas, one of which is asleep and one of which is awake. Something called the edge effect apparently occurs, that is, there is a discrepancy. Some parts are stuck in a particular mode while the rest are computing other things. We came to the conclusion that in Parkinson's the inability to move in a coordinated manner was created by the fact that the frequency of oscillations did not allow for generation of proper movement orders to the muscles. And the edge effect, in contrast, was a continuous order to move.
The edge effect—the continuous order to move—causes the tremors?
Yes. And if that is the case then if you wake up the appropriate part of the thalamus this problem should disappear. If you put an electrode in the part of the thalamus that polarizes it, this problem should immediately disappear.
And this has been tried?
Yes, we have spectacular films of such results. You can see how when the person is stimulated the results are immediate and the person begins to function. You are bringing the circuit back into the correct timing by either returning the cells to their correct rhythm or by taking them out of the loop. So, the question is, what happens if that occurs in another part of the thalamus? What about focusing on a part that hears? What you would expect to happen happens—the edge effect now produces a sound, rather than a tremor. And what property does this sound have? It is on all the time. It has a certain ongoing frequency, like a persistent acoustic tremor. The person hears something not from the outside but from a signal on the inside. This is the case in tinnitus.
Ringing in the ear?
Right. So what happens if something like this happens in a part of the brain that feels pain—the cingular cortex? What happens is that you have central pain. Like when you lose an arm but the hand that should be there hurts, you have a phantom limb. [For more on phantom limbs see From Ramachandran's Notebook.] If you put in an electrode, and you stimulate the correct part of the thalamus, you get rid of the pain. And what happens to depression? Same thing. Hallucinatory events in the visual system are probably the same thing too. So there are a whole lot of psychiatric and neurological problems that might be related to a thalamo-cortical dysrhythmia.
Do you think that someday you might be able to insert electrodes in the appropriate part of the brain to solve all these different problems?
Well, hopefully it is only a temporary solution while we think of better ways to solve the rhythm problem. A pacemaker with an electrode is a very rough solution.
"The nervous system doesn't repair itself for a very good reason."
Most people, it seems, don't view the brain as electric—they view it as a big gland secreting neurotransmitters. So that if you have an imbalance of serotonin you might have depression, or if you don't have enough dopamine, you have another problem—Parkinson's. How do neurotransmitters fit into your view of the brain?
Dopamine and serotonin work by modifying directly and indirectly the electrical activity of neurons. Too much or too little generates medical conditions. With Parkinson's there is no dopamine. What happens then is that the thalamus becomes hyperpolarized and starts firing continuously. In the case of Parkinson's, dopamine goes down, and then the thalamus begins to oscillate, and you get tremors and paralysis.
What other diseases might involve this timing problem, or lack of rhythm, between the thalamus and the cortex?
We think that this thalamo-cortical dysrhythmia is responsible for some types of epilepsy, Parkinson's, depression, obsessive-compulsive disorder, some aspects of schizophrenia, central pain, and tinnitus. They are all part of the same disease.
And you think this because you have seen evidence of it?
Yes.
How do you actually view these rhythms?
You can either put the patients in an instrument, like an EEG [electroencephalogram], or better an MEG [magnetoencephalograph], or else you can put electrodes in the thalamus during surgery and see abnormal activity.
Supporters of biofeedback therapy have cited your work as support for their theories of why biofeedback could work in treating epilepsy. Does that make sense to you?
It makes a lot of sense. With biofeedback, though, there's only so much you can do. If someone had enough of an intact brain to reorganize, then there would be hope. The best way to correct these things is with specialized designer drugs that would go to specific cells and change the properties.
But there aren't many of those.
Not at the moment, but rational drug development in neuropharmacology will result in some, at some point in the near future. We will dream some up. But you have to know what the problem is first, right?
So what is coming out of this research is not necessarily pacemakers for the brain but more targeted drugs?
Pacemakers aren't my idea of heaven. It's the same with cancer treatment. These are molecular problems, and we are attacking them with spoon- or toothpick-sized tools. Obviously it's wrong, but that's all we have. Rational drug design is what's going to happen. At the moment we use drugs that hit everywhere, and so they have unwanted side effects. There are bad side effects to neurosurgery too and to having cables coming out of your head. There are side effects in all of these things we do.
Does the normal thalamo-cortical rhythm change as you grow older?
Yes, we have looked at that. Forty hertz [number of cycles per second], or beta gamma band activity, as it's known, is not as well organized with age. There are low frequencies that begin to creep in. So people don't hear so well, they have a little ringing in the ears, and they don't see so well. You begin to see how the ability to put things in time very crisply disappears. A 25-year-old is agile, intelligent, sees things better, and can memorize and remember better. He has the disadvantage of not having experience, hence the saying, "if only the young knew or the old could." The young don't know but they can and the old know but they can't.
What do you think about this work that has come out recently that suggests that the brain is always generating new neurons?
It's a bit controversial. I haven't seen many people getting better with time. There's a very good reason for it, by the way. The nervous system has its own developmental history. Most of the nerve cells are as old as we are, and they have known one another for as long, so they have established their connectivity—they have a common history. So if a new cell comes in, forget about the new-kid-on-the-block problem. How are you going to give functional context to that cell?
The nervous system doesn't repair itself for a very good reason. It lacks context if it starts growing in all directions. A paralyzed leg is a big problem, but worse than a paralyzed leg is one that moves in the wrong direction or at the wrong time. That one can easily kill you.
Do you find, spending so much time thinking about how the brain thinks, that it's hard, even when you're not at work, to stop thinking about how how your brain thinks?
Sure. I find such thought fascinating and often downright amusing. Esthetically the brain has been beautifully assembled by evolution, and it's also, fundamentally, about the nature of our own existence.