Agents and anarchists

THE INHABITANTS OF the British Isles have had a rather fraught relationship with their kinfolk on the European mainland for a very long time. Over the millennia, they have fought again and again, sometimes winning, sometimes losing, against Italians, Spanish, French, Danes, Dutch, Poles, Bulgarians, Austrians, Hungarians, Albanians and Yugoslavians. But over the last century or so, the Germans have been the most significant adversaries, and not only have there been battles involving the military, there have also been some that involved intellectuals.

In the nineteenth century several German scholars argued for an unusual idea. They knew that we can feel the positions and motions of our limbs and that we are aware of exerting forces and making a physical effort. At the time they thought they knew that there were no sensory organs inside the body that might be sending messages about these things, so where were the feelings coming from? Their answer was that they are produced by the command signals that the brain sends to the muscles in order to exert a force or move the body. The Germans proposed, therefore, that these feelings are not based on messages from sensors in the body, but upon the signals that tell muscles what to do. This is often called the outflow theory of the propriosenses (kinesthesis and the senses of force and effort) because the signals from which the feelings derive flow out from the brain. 

When propriosensors in muscles and joints were discovered late in the nineteenth century, the Germans were not inclined to change their view. They were happy to concede that these sensors existed, but argued that they have nothing to do with propriosensory perception and our conscious bodily awareness. Rather their sole purpose is to provide the body with certain reflexes that operate entirely non-consciously. Propriosensors are indeed involved in a number of these, the best known of which is the muscle stretch reflex.129 When you stretch a muscle, sensors within it detect the stretching and transmit messages to the spinal cord where they are transformed into signals that are sent back to the muscle causing it to contract and resist the stretching. This is the same reflex that causes your leg to kick out when a physician taps your patellar tendon – the so-called knee-jerk reflex – and the Germans knew all about it, claiming that such reflexes are what propriosensors are for.

Aristotle held that if something is to count as a sense or sensory faculty, there must be a sensory organ: only feelings that derive from messages coming into the brain from sensory organs count as sensory feelings. Most people today would be in agreement. From this perspective, a feeling that derives from commands that flow out from the brain is not a sensory feeling, it is something else. If the German outflow theory is right, then the propriosenses are not like the classical five senses, but more like the sense of entitlement or the sense of humour, not 'sensory senses' at all. British thinkers of the late nineteenth century objected strongly to this German nonsense and dismissed it as continental tosh, a common response to what goes on across the English Channel. British neuroscientists Henry Bastian, David Ferrier and Charles Sherrington argued that propriosensory feelings derive from messages that flow into the central nervous system from propriosensors, the inflow theory. According to the British, therefore, these feelings are sensory in exactly the same way as the experiences associated with the traditional five senses. But they faced stiff opposition led by two giants of the German-speaking scientific establishment, Herman von Helmholtz and Ernst Mach. Against such powerful and influential adversaries, the battle would not easily be won. Indeed, it would not be won at all. 

 

Inflow and Outflow

CHARLES SHERRINGTON WAS no minor figure himself: he made substantial contributions to our understanding of the nervous system, coined the term synapse for connections between nerve cells, and was awarded the Nobel Prize in Physiology or Medicine in 1932. He also argued for the inflow theory. He didn't have any conclusive experimental results that established the truth of it, but he did have some good arguments. The easiest to understand is his observation that propriosensory feelings are possible when there is no outflow to the muscles. According to the outflow theory, if there is no outflow, then we should not know where our body parts are unless we can see them or feel them touching one another. Sherrington observed that when our muscles are completely relaxed and hence receiving no outflow, our ability to perceive the posture of body parts when the eyes are closed is not significantly affected. This contradicts the outflow theory, but inflow theory can easily explain the observation: there are messages coming from propriosensors when the muscles are relaxed, so there is inflow, and this is responsible for kinesthetic perception of relaxed limbs. 

Another argument of Sherrington's was based upon the abilities of a man with an unusual deficit. He was a patient of two French doctors who assessed him as having no tactile and no propriosensory capacity between his neck and his waist.130 They investigated his ability to sense the heaviness of objects that he held in his hands and found that he was unable to perceive any difference in the heaviness of objects that differed substantially in weight; he said that they all felt weightless. He was also unable to feel any difference between hand-held objects when he squeezed them, even though some of them, such as a crumpled newspaper, would have been easily crushed, whereas others, such as a piece of wood, would not. Holding an object or squeezing it requires muscular force and hence outflowing commands, with more force and greater outflow being required the heavier or less squeezable the object. If the outflow theory were correct, then the patient should have had different experiences of force exerted and effort expended when the outflow was greater, but this was not what happened. The patient reported no differences in his experiences. This is what would be expected if his experience were derived from messages sent from sensors in the limbs that held or squeezed the objects: there were no messages, so there were no differences. 

This second argument of Sherrington's is weaker than the first since it is based on observations obtained in 1890 from just one very unusual person. It would be more conclusive if there were more recent, verifiable cases of tactile and propriosensory loss. If such cases could be found, it would perhaps enable us not only to determine if the inflow theory really is supported, but also what it would be like to experience propriosensory loss. We can imagine what it is like to be blind or deaf and can approximate the experience by shutting our eyes or plugging our ears. We can imagine what it is like to lose the sense of touch as we might have experienced some partial loss due to the cold, local anaesthesia or to the kind of pressure on a nerve that produces pins and needles. It's hard, however, to imagine what the loss of propriosensory feeling would be like. Sherrington thought that the loss would be bizarre and so beyond everyday experience that it was 'doomed to remain indescribable',131 but it's quite possible that the loss would hardly be noticed given that we are hardly aware of propriosensory feeling when we have it. There is, however, an old bit of folk wisdom that says 'you don't know what you've got till it's gone'.132

 

IN 1971 IAN WATERMAN, then a young man of nineteen, contracted a virus. He became quite ill and felt overwhelmingly weak and exhausted. After a few days the sickness passed, but the exhaustion persisted, and he spent most of his time in bed. Less than a week later, he lost the ability to properly coordinate his movements and collapsed on to the floor when he tried to get up. His speech became slurred, he felt tingling sensations in his hands and feet, and found that they had lost their sense of touch. Soon afterwards he found he couldn't feel touch on any part of his body below the neck. He also found he couldn't do anything: he couldn't get out of bed, he couldn't eat, couldn't reach for anything, turn the pages of a book or remove items of clothing, he couldn't even sit up. It wasn't that he couldn't move – he could – he just couldn't make his body do what he wanted; he couldn't make it move in a purposeful way. It is hard to imagine how frightening and bewildering this experience must have been for the young man.

Ian's sensory nerves had been severely damaged by what was believed to be an autoimmune reaction that occurred as a consequence of the viral infection. The affected fibres were those ending either with propriosensors or with sensors that respond to skin pressure, contact or stretch – tactile sensors. These sensors and nerve fibres comprise what is known as the large-fibred sensory system of the body because the fibres are thick (large) and have a sensory function. There are thinner ones that comprise the small-fibred sensory system, but these were not affected. The sensors of the small-fibred system respond to temperature, to tissue damage, to metabolites and other chemical substances, and to certain types of squeezing and stretching of the internal tissues. Thus, Ian lost all of his sense of touch with the exception of temperature and skin pain (if we count these as components of touch). If Sherrington and the British neuroscientists were correct, he should have lost propriosensory feeling as well. If Helmholtz and the Germans were correct, propriosensory feeling should have been largely unchanged. 

How did Ian Waterman's propriosensory loss affect his bodily experience? The answer is that if he couldn't see his body, it seemed to disappear. This disappearance was more profound than the kind of recession into the background that occurs when we don't pay attention to the body. When he shut his eyes he completely lost track of where his body parts were, he couldn't locate his limbs or deter­mine their posture, and no amount of effort trying to attend to his body parts would enable him to locate them. He also lacked proprio­sensory reflexes: when a physician tapped his patellar tendon, nothing happened. He could contract his muscles by acts of will but couldn't make purposeful movements or feel the movements that his body actually made. He was, according to the title of a BBC television doc­umentary about his life, 'the man who lost his body'.133 Of course, because his small-fibred system was intact, he would not have com­pletely lost bodily sensation. He could feel aches, pains, hot and cold, and some aspects of sensual touch. These sensations were felt at par­ticular locations because his body-model framework was presumably still present, but without propriosensory updating these locations would typically not correspond to reality.

Ian's prognosis was grim: he was told that his condition was incurable, that he would never recover and faced spending the rest of his life in a wheelchair. But he refused to accept these dire predictions. Over the months and years following his illness Ian taught himself to make purposeful movements again using vision to replace what his illness had taken away. He never regained his tactile and propriosensory feeling, but he did learn to walk again and was able to live an independent life. The remarkable story of his efforts and their ultimate success are described in the book Pride and a Daily Marathon written by his physician, Jonathan Cole. I will return to Ian's condition later. What matters for us right now is that it confirms what we would expect based on the inflow theory of propriosensory feelings: when the brain receives no propriosensory messages, propriosensory feeling vanishes. Thus, cases like Ian Waterman's (there have been one or two others) would seem to hand victory to the British and their inflow theory, consigning the Germans and their outflow theory to the rubbish heap of history. But rarely is anything quite as conclusive as it seems, and this particular battle was destined to end in a draw.

 

AS OFTEN HAPPENS in debates between two factions with opposing views, instead of trying to reach a compromise, both sides remain obdurate in their refusal to accept that the views of the opposing faction have any merit. It typically turns out that both sides have a part of the truth and if they were only willing to acknowledge this, progress would be quicker. This is certainly the case for the inflow and outflow theories – both are partly correct, but both are wrong in claiming that only inflow or only outflow is involved in propriosensory feeling. It is now generally accepted that kinesthesis, together with the sense of exerted force and heaviness, involves contributions from both inflow and outflow.134 I'll call it the inflow+outflow theory. 

When we are doing things – walking, standing and talking, drawing or writing – we are contracting muscles and moving our bodies. Outflowing messages are telling our muscles what they should be doing, and sensory messages are flowing in from sensors in the muscles and joints. According to the inflow+outflow theory, the outflowing messages go to two places: to the muscles and to circuits where they are combined with the inflowing sensory messages to generate propriosensory perception and feeling. Thus, the outflowing signal is split into two: one that actually affects the muscles and another that is destined for the circuits responsible for propriosensory perception and feeling. Scientists usually refer to the latter as an efferent copy;135 the word 'efferent' simply means flowing out from somewhere in the nervous system.

When the muscles are completely relaxed, there is no command signal and so there is no efferent copy either; only inflowing sensory messages will be reaching the propriosensory circuits in the brain. As Sherrington pointed out, we can perceive the positions, postures and movements of our completely relaxed, unseen limbs, implying that inflow alone is sufficient for kinesthesis (no sense of effort or exerted force would be expected since no force or effort is exerted in the absence of outflow). From the presence of kinesthesis in the absence of outflow we can conclude that the brain does not combine inflowing and outflowing signals by multiplying them, since multiplication by nothing (no outflow) gives nothing. Some kind of addition or subtraction would work, though. We might expect, therefore, that propriosensory percepts and feelings will be present when there is outflow but no inflow. However, the picture that emerges is not completely clear. Ian Waterman's experiences suggest that kinesthesis is absent when there is no inflow, but there are other cases in which some residual kinesthetic experiences seem to persist. Some amputees, for example, not only sense the presence of their phantom limbs and feel pains and tingles in them, but are able to move them around as well. When they try to move the phantom as if it were a real limb, they feel it move in the manner intended, but they have no propriosensors in the missing limb, which seems to support the outflow theory. However, they do have propriosensors in what remains of the limb and where the limb once connected to the rest of the body, and movements of these parts would produce some inflow. Perhaps this inflow is responsible for the phantom movements.

In 2006 a group of researchers based in Sydney, Australia attempted to resolve the matter by conducting a definitive experiment.136 The idea was to remove all sensory inflow from a person's right hand and forearm, paralyse the muscles, hide the arm under a screen so that it could not be seen, and then determine whether the person could feel a change in the position of their numb, paralysed hand when they tried to move it. Trying to move the hand implies that a person is sending an outflowing command signal to the muscles: if they feel the hand's position change, then the feeling must have been produced by this outflow alone. So, what actually happened? 

The people who took part moved a pointer to indicate whether or not they thought, felt or otherwise perceived that their right hands had moved. They all moved the pointer in the direction that they had tried to move their hands. You might think that this demonstrates that kinesthetic percepts can be produced by outflow alone, but I'm not completely convinced. Participants moved the pointer in the direction of the attempted movement, but that doesn't mean that they had any kinesthetic perception or feeling of the hand moving. It may be that the person inferred that their hand must have moved and adjusted the pointer accordingly. They would try to move, feel no movement of the hand and no tactile sensations to indicate that movement was blocked, and infer that the hand must have moved in the intended direction because that is what it would normally do. They would have been aware of trying to move, but this is not the same as having the kinesthetic experience of hand movement. Alternatively, trying to move a numb, paralysed hand could produce a strong belief that a movement had been made, which in turn could cause someone to convince themselves that they experienced the movement or even to hallucinate it. So, the movement of the pointer could indicate a genuine perception of movement, an inference that movement occurred, a false conviction that they had actually experienced a movement, or a hallucination. Given these alternatives, we cannot know for certain whether or not outflow alone produces kinesthetic perceptual experiences. After more than one hundred years the question of whether outflow alone can produce kinesthetic experiences still isn't completely settled. The same is true for the other propriosensory feelings: it is not yet decided whether feelings of effort and force can be produced by outflow. Nevertheless, when all the evidence is taken into consideration, it seems to me to be highly likely that propriosensory experiences can be evoked by outflow alone.137

What we now know is that the inflow+outflow theory is correct in asserting that both are involved in kinesthesis. So, neither the British nor the Germans were right. Both had identified part of what was going on, but only by coming together were they able to get closer to the truth. The inflow+outflow theory is intriguing because what might seem to be a sensory feeling – the movement of a limb, the heaviness of a hand-held object – is actually partly sensory (the inflow part) and partly not sensory (the outflow part). It shows us that perceiving the body is not only a matter of sensing it (inflow) because what we are doing with it (outflow) also makes a contribution. 

But the role of outflowing messages is not restricted to their influence on propriosensory experience. Cognitive neuroscientist Sarah-Jayne Blakemore and her colleagues in London have proposed that outflow explains why someone else can tickle you, but you can't tickle yourself.138 When you are being tickled, there is sensory inflow from the tickling but no outflow. When you try to tickle yourself there is both inflow and outflow and the latter suppresses the ticklishness that would have been experienced if only the inflow were present. Outflowing messages can also enable us to determine whether we did something or not. It may never have crossed your mind to wonder how it is that when your body moves, you can tell whether you moved it or not. Indeed, you might wonder why anyone would wonder about something so blindingly obvious. Surely it's impossible not to know or to be deceived about whether you made a movement or not. But as with everything else psychological, hiding within the apparently simple and obvious is the complex and mysterious. 

 

Agency

EVERYTHING YOU DO involves moving your body and the movements you make depend upon what you're doing. The movements needed to hold a conversation are rather different from those needed to dig a hole or play the piano. You make a huge variety of different movements as you go about your business, but despite their variety they come in only two basic types, we call these voluntary and involuntary. This terminology can lead to misunderstandings, so let's begin by clarifying what it means. The first thing to be careful about is that the word voluntary in this context doesn't mean done of your own free will, it means brought about by an act of will, an act that could be free or not. If someone forces you to make a movement by threatening to cause you pain if you don't, then the movement isn't done of your own free will, but it still counts as a voluntary movement. 

The second thing to bear in mind is that a voluntary movement is not necessarily one that you intend to make or even one that you are aware of making. Consider voluntarily saying a word. This will likely involve a lot of subtle movements of the lips, jaw, tongue, larynx and chest. You are almost certainly unaware of making most of these and even those you are aware of aren't really intended: you aren't aware of an intention to move your tongue or your jaw when you speak. So, it isn't the movements themselves that are intended, it is the utterance of the word. Likewise, when digging a hole or picking up a cup, it isn't the movements of your body that are intended (most of which you are also unaware of), rather it is the removal of earth or the raising of the cup. A voluntary movement is not typically an intended movement, it is a movement made when carrying out an intended act like raising a cup. In order for a movement to count as voluntary, it simply has to occur as part of an intended act. 

Now you know what a voluntary movement is, it's easy to say what an involuntary movement is – it's one that doesn't occur as part of an intended act. Your body is producing such movements all the time and you don't need to do anything to make them happen: your heart is beating, your chest is moving in and out as you breathe, your eyelids are blinking. Other involuntary movements are produced in response to something happening. The knee jerk you make when someone taps your patellar tendon is an example, another is the sneeze produced in response to irritation of the nasal passages – in both these cases, the movements you make are reflexive movements. Involuntary movements can also occur as the result of something pulling on you or pushing you. For example, if you hold your arm up and then relax, it will fall – an involuntary movement caused by the force of gravity. These latter kinds of involuntary movements are called passive movements; those produced by your own muscular activity are active movements.

When your body moves, you know immediately what kind of movement it was – voluntary, involuntary, passive or active – your experience of these different kinds of movements is different. When your body moves passively, you experience the movement as involuntary and caused by some external force or agency. When your body moves as a result of your own muscular contractions, you experience it as one you made yourself. This is true of both voluntary movements and active involuntary movements. If you blink or breathe, then it's you doing the blinking and breathing. If you get some finely ground pepper up your nose, you sneeze, and you experience it as something you did involuntarily. What enables you to determine what kind of movement it was? When you make voluntary movements, you have the experience of willing movement to occur – or more accurately, of willing something to occur by means of bodily movement – and of it occurring as a direct consequence of your act of will. You experience the act of will as the cause of the movement and of its intended consequences. Philosophers have long wondered whether this feeling of being the causal agent – the sense of agency as it is called139 – corresponds to what is really happening. Are your acts of will really the causes of your bodily movements? This is an open question; what matters now is that voluntary movements are accompanied by a sense of agency, whereas involuntary ones are not, so it's obvious to you what kind of movement your body has made. 

Suppose your body moves involuntarily. There is no sense of agency, no experience of an act of will, but you are able to determine whether the movement was active (you did it) or passive (an external force did it) and the experiences are quite different. Involuntary movements that you produce, such as reflexive movements and the movements made when breathing, are associated with an experience of self-causation, but without an act of will, a sort of sense of knowing that you did it. Making active movements involves producing out­flowing command signals and these, in the form of efferent copies, could be the basis for discriminating between active and passive involuntary movement. But there are also differences in the sensory messages produced: active muscle contractions stimulate proprio­sensors that are not stimulated by passive movements140 and being pushed or pulled usually evokes tactile sensory messages from where the push or the pull is applied to your body. Normally, of course, both these inflowing and outflowing signals are available to tell you whether a movement was active or passive. But Ian Waterman and others like him can tell if an involuntary movement that they see themselves making (they can't feel them) was something that they did or was produced by an external force. This means that outflowing signals are alone sufficient to distinguish involuntary movements that you make from those produced by external forces.141 However, it is not sufficient that there simply be outflowing signals that cause the movement, the outflow must be coming from the right place in the brain.

In Chapter 2, I described how prior to brain surgery, Wilder Penfield electrically stimulated the surface of his patients' brains. Recall that stimulation of a strip of cortex called the postcentral gyrus evoked sensations on the surface of the patient's body. Electrical stimulation of another strip of cortex located immediately in front of the postcentral gyrus called the precentral gyrus was found to evoke movements of the body. The patients did not experience these movements as self-produced and would deny having made the movements themselves, saying 'you did that' rather than 'you did something that made me do that'. Electrical stimulation of the precentral gyrus causes a signal to be transmitted out of the brain, down the spinal cord and out to the muscles. This outflow signal does not give rise to an efferent copy that causes a person to experience an involuntary movement as one that they produced themselves – the movement is experienced as being caused by an external agent. The signals that are responsible for feeling that a movement is self-produced must be found deeper in the workings of the brain. 

ACCORDING TO ALAN TURING, the Austrian philosopher Ludwig Wittgenstein was a 'very peculiar man'.142He certainly asked some peculiar questions, one of the best-known being, 'What is left over if I subtract the fact that my arm goes up from the fact that I raise my arm?' Part of Wittgenstein's appeal is that it is never quite clear exactly what his pronouncements mean, and this is no exception. But let's suppose that Wittgenstein is talking about a voluntary raising of the arm and so is actually asking something like, 'If you voluntarily raise your arm but somehow the arm's going up is taken away, what remains?' The situation vaguely resembles the experiment described earlier in which the person tries to move their hand, but because the muscles are paralysed, the hand does not move. But the experiment does not match the situation Wittgenstein is asking about because the person could not feel the paralysed limb due to it being anaesthetised. There are, however, some other studies that might help us answer our modified version of Wittgenstein's question.

In 2008 Michel Desmurget and his colleagues at the French National Centre for Scientific Research studied the effects of direct brain stimulation on patients who were about to undergo surgery to remove brain tumours. The situation closely resembled that of Wilder Penfield's studies: small electrical shocks were delivered to the exposed surface of the patients' cerebral cortices at various locations. These locations were slightly different to those Penfield stimulated: some were behind (caudal to) the postcentral gyrus in a region of the cortex called the posterior parietal lobe. The effect of stimulating these sites was to produce experiences of what the researchers called the 'intention and desire to move',143 though no movement was actually produced. Patients would say things like 'I felt a desire to lick my lips' and that they had a feeling 'like a will to move'. If the strength of the stimulation was increased, the experience changed to include a feeling that a movement that they intended to make had taken place, but again, no movements were produced. It is as if stimulation of these sites in the parietal lobe produced what our modified version of Wittgenstein's question asks for – the thing that exists when an intended movement is not present. The answer being, the experience of willing a movement and in some cases the feeling of that movement having taken place, even though it didn't. An experience of willing a movement and a feeling that it took place when it didn't is a sense of agency for an illusory movement.

One conclusion from Desmurget's study is that the experience of intending and willing a movement develops in the parietal lobe of the cerebral cortex. So, activity in this part of the brain, but not elsewhere,144 gives rise to the experience of self-causation, at least for voluntary movements. What we cannot know, however, is whether acts of will themselves (or whatever actually causes voluntary movements to occur) are generated in the parietal lobe to spread subsequently to other regions of the brain and be transformed into actual motor commands. Additional findings of Desmurget and colleagues, together with those of other researchers, have shown that the precentral gyrus and regions of the frontal lobes in front of it are involved in the production of movement, but not with feelings of willing a movement or of intending to move. These areas are candidates for where intentions are converted into signals that produce the movements. It has been found in monkeys that stimulation of some locations in the frontal lobes can produce complex patterns of movement associated with the performance of actions directed towards achieving some useful outcome, like picking up an object.145Desmurget and colleagues reported similar results in the patients they studied, but although complex patterns of movement were evoked, the patients were unaware that that had produced them. You have to wonder what their experiences would have been had they been aware that they had produced these movement patterns without any intention to do so. 

 

Anarchy

STEPHEN ORLAC WAS an accomplished concert pianist who lived in France. He was returning to Paris from Fontainebleau when his train was involved in a serious accident. Orlac survived but was very seriously injured. Most distressingly, both his hands were badly crushed and mangled; it looked as if he would lose them and never play the piano again. Fortunately, he had the opportunity to receive a double hand transplant. The operation was successful, but Orlac was unable to use his new hands to play the piano; he found that they had a different skill: they could throw knives and soon developed a strange and autonomous tendency to do just that. Distressed and disturbed, Orlac sought out the doctor who performed the transplant and explained that his hands 'have a life of their own. They feel for knives. They want to throw them, and they know how to . . . They want to kill.' This was no surprise, for the hands were those of none other than Rollo, the convicted murderer and circus knife thrower, who was on the way to his execution when he was killed in the same train accident that destroyed Orlac's hands. Such is the plot of the 1935 movie Mad Love, starring Peter Lorre, which was adapted from the novel, The Hands of Orlac by Maurice Renard. The theme of someone's hands having a mind of their own and doing things that are neither wanted nor intended, at least not by their owner, has featured in one or two other movies. The most famous is Stanley Kubrick's black comedy Dr Strangelove, in which the eponymous doctor, played by Peter Sellers, has trouble with his right hand. It has a tendency to make Nazi salutes and at one point tries to throttle him so that he has to defend himself by wrestling it with his other hand. Most bizarrely, however, it turns out that possession of a hand that seems to act of its own volition is not merely a figment of the creative imagination.

Between 1906 and 1914, the famous neurologist Kurt Goldstein was a resident physician at the Psychiatric Hospital of the University of Königsberg and it was here that he encountered an extraordinary patient. Known only by her initials, HM, she made some bizarre claims and exhibited odd behaviours that had never previously been reported in the medical literature. Her left hand, she explained, had grabbed her by the throat and tried to throttle her. And that was not all; it was continually doing things that she didn't want it to. When she buttoned a shirt with her right hand, the left hand would undo the buttons she had just done up. She claimed that there was 'an evil spirit occupying the hand' and that the hand would do 'what it likes'.146 She would get angry and frustrated with her hand and often resorted to hitting it or talking sternly to it in an attempt to stop it behaving in an unwanted fashion. Goldstein's account of the strange claims and behaviour of his patient was the first time such a case had been documented. It would be thirty-seven years until another was described. Since then, several more cases have been reported with the common feature that only one hand is affected and it moves in a way that achieves specific outcomes, like undoing the buttons on a shirt or picking up an object, but without any intention on the part of the patient. To someone watching, the behaviour usually looks normal and exactly as if it were intended. It is hard to imagine that it wasn't, yet the person obdurately insists that they did not mean to do it. They lack a sense of agency for these actions and are unable to stop them unless they grab the affected hand with their unaffected one. To the sufferer, their affected hand behaves as if it had a mind of its own.

Those who suffer with this condition are said to have an anarchic hand and it can occur without other significant symptoms, particularly when the right hand is affected.147 Sufferers have normal tactile and propriosensory feeling in the errant hand, experience it as being a part of their body, and when it's not misbehaving, they can make voluntary use of it – they experience no paralysis. The weird, almost comic nature of the condition has caught the public imagination: in recent years it has been the topic of a TV documentary and featured in episodes of the medical drama series House M.D., the comedy horror series Scream Queens and the cartoon series South Park. But bizarre though it is, the condition is not particularly difficult to understand. 

We know from studies like the one of Desmurget and colleagues described earlier, that patterns of upper limb movement associated with goal-directed actions can be evoked by electrically stimulating locations in the frontal lobes without evoking any sense of agency for the movements produced. In normal circumstances, nerve cells at these locations become active in response to incoming signals associated with the intention to act, signals that may originate from within the parietal lobes of the brain. Perhaps what is happening in people with an anarchic hand is that the frontal lobe circuits are becoming active but without the normal 'intention to act' signals. How could this come about? One clue lies in the fact that the anarchic hand only carries out an unintended act if the sufferer becomes aware of the presence of something that they can grasp or handle: they see a food item and they pick it up and try to put it in their mouth, they see a comb and they pick it up and start trying to comb their hair. This suggests that the frontal lobe circuits are being activated by the sight of something. Visual messages are activating the circuits involved in actions appropriate for it – the sight of an apple activates the circuits that generate pick-it-up-and-bite-it behaviour, the sight of a comb activates pick-it-up-and-comb-your-hair-with-it behaviour. There is a large body of evidence to show that this kind of activation does happen, but normally it does not result in the behaviour actually being carried out unless the person wants to. So, in patients with an anarchic hand, something has happened in their brains that allows the visual activation to elicit the behaviour; what is it?

Having an anarchic hand is associated with damage to a particular region of the cerebral cortex centred on a part called the supplementary motor area, or SMA for short. The human cerebral cortex has something in common with another part of the human anatomy, one accidently referred to by Winston Churchill when he was delivering a speech to the French National Assembly. Churchill is alleged to have declared, 'quand je regarde mon derrière je vois qu'il est divisé en deux parties'.148 He wanted to say something to the effect that when he looked back on his life, he saw that it had two distinct parts. What he actually said translates as, 'when I look at my backside, I see that it is divided into two parts', which though anatomically accurate, is not what he wanted to convey. Like Churchill's derrière, the cerebral cortex is divided into two halves with a deep cleft between them (see Figure 2.1, p. 38). There is an SMA in both halves; they are found at the top of each, just in front of the precentral gyri, and they extend a small way down into the cleft. Anarchic hand results when the SMA and nearby parts in one half of the brain are damaged. Most commonly, the damage is in the left half, and the right hand is anarchic.

The obvious conclusion is that the SMA and nearby regions exert some kind of suppressive control over the frontal lobe circuits responsible for goal-directed upper limb movements. Suppression of activity in these circuits ensures that simply seeing an apple cannot evoke a grab-apple-and-stick-it-in-mouth action – it's like a kind of veto on unwanted acts. In order to grab the apple voluntarily, the veto needs to be briefly lifted and then the act will be evoked as it is in the anarchic hand case, except that lifting the veto is associated with a sense of agency. Sergio Della Sala, an Italian neurologist working at the University of Edinburgh, expressed this idea nicely when he wrote, '[f]rom this perspective it looks as if our brain may have a free "won't" rather than a free will'.149

- 'How you feel: The story of the mind as told by the body' is written by by James Tresilian, and published by Little, Brown Book Group. It's out now.