Читать книгу Philosophical Foundations of Neuroscience - P. M. S. Hacker - Страница 32

Penfield’s training

Wilder Penfield (1891–1976) was born in Spokane, Washington. After graduating from Princeton in 1913, he won a Rhodes Scholarship to Oxford and entered the School of Physiology there to begin his medical studies under the inspiring influence of Sherrington. He followed Sherrington’s interest in histology and, in particular, in neurocytology. After obtaining his BA in physiology at Oxford, he went to the Johns Hopkins Medical School, where he finished his medical degree in 1918. His first research concerned changes in the Golgi apparatus of neurons after axonal section. In 1924 he began to study the healing processes of surgical wounds in the brain. On Sherrington’s advice, he spent some time in Madrid working with Pio del Rio-Hortega, learning to use the histological methods of his brilliant teacher Ramón y Cajal. To this end, surgical specimens of brain scars were collected from patients who had been operated on for post-traumatic epilepsy.

Penfield’s achievement

Penfield was aware of the studies on cortical localization in the primate brain that Sherrington had carried out, and which have been described above. In 1928 he went to Breslau to work with Otfrid Foerster, to learn his method of gentle electrical stimulation of the cortex of epileptic patients while they were under local anaesthesia during the excision of epileptogenic scar tissue. During these procedures he learned the method of operating under local anaesthesia, using electrical stimulation to identify the sensory and motor cortex to guide the surgical excision. This technique was to be used to singular effect by Penfield in Montreal, where, in 1934, he established the famous Montreal Neurological Institute at McGill University, which was devoted to the study and surgical treatment of focal epilepsy. Such stimulation made it possible to locate exactly the position of the sensorimotor cortex and of the cortex subserving speech, so that these vital areas could be spared during the surgical excision. In some instances the stimulation might activate the more excitable epileptogenic cortex and reproduce a portion of the patient’s habitual seizure pattern. This enabled the surgeon to identify the site of the physiologically deranged epileptic focus. Penfield’s mastery of these procedures was subsequently summarized in a series of monographs on brain surgery for epilepsy.

Penfield noted in 1938 that stimulation of certain parts of the temporal cortex in patients occasionally excited the vivid recall of previous experiences. It became evident that almost half of the patients afflicted with epilepsy had seizures that could be shown to originate in one or other of the temporal lobes. This work on temporal lobe epilepsy led to very important observations regarding the hippocampus and memory function, as well as the localization of the cortex subserving the latter. By 1951 Penfield, together with Milner, had shown that removal of one hippocampus on the medial aspect of the temporal lobe resulted in severe memory disorder in patients who were later found to have damage to the hippocampus on the opposite side. Thus the bilateral loss of function of the hippocampus led to the complete inability of these patients to remember any post-operative occurrence. This memory loss was not accompanied by any loss of intelligence or attentive capacity. Penfield’s analyses of the electrical stimulation of the cortex of 1,132 conscious patients undergoing brain surgery greatly extended our knowledge of functional localization, especially with regard to memory and to that most human of capacities, speech.

Penfield’s methodological commitment

Already in his student days, Penfield had had a ‘sense of wonder and a profound curiosity about the mind’. When he turned from the study of the animal to the human brain, his ‘planned objective’, he later wrote, was ‘to come to understand the mechanisms of the human brain and to discover whether, and perhaps how, these mechanisms account for what the mind does’.²² Studying under Sherrington, he came to ‘the realization that the brain was an undiscovered country in which the mystery of the mind of man might someday be explained’. He was, of course, fully aware of Sherrington’s views on the relation of mind and brain. In the final paragraph of the foreword to his great book The Integrative Action of the Nervous System (1906), Sherrington had remarked, ‘That our being should consist of two fundamental elements offers, I suppose, no greater inherent improbability than that it should rest on one only.’ Penfield, however, took the view that the neuroscientist should endeavour to explain the behaviour of animals, including humans, on the basis of neuronal mechanisms alone. Only if that failed, he thought, should one have recourse to alternative forms of explanation. And throughout his career as a neurosurgeon, he retained this methodological commitment.

Penfield on the mind

Towards the end of a long life dedicated to neurosurgery and neurology, Penfield published a small volume entitled The Mystery of the Mind. This was, he wrote, ‘the final report of my experience’ – an overview of what he had achieved in respect of his youthful objective. ‘The nature of the mind’, he averred, ‘presents the fundamental problem, perhaps the most difficult and most important of all problems’ (MM 85). What he wished at last to do, he wrote in the preface, was to ‘consider the evidence as it stands, and ask the question Do brain mechanisms account for the mind? Can the mind be explained by what is now known about the brain?’ (MM, xiii). Referring explicitly to the above-quoted remark of Sherrington’s, Penfield judged that ‘the time has come to look at his two hypotheses, his two “improbabilities”. Either brain action explains the mind, or we must deal with two elements’ (MM 4). Despite his methodological commitment, Penfield found himself driven towards a Cartesian view not unlike that of his great teacher. ‘For my own part’, he wrote,

after years of striving to explain the mind on the basis of brain-action alone, I have come to the conclusion that it is simpler (and far easier to be logical) if one adopts the hypothesis that our being does consist of two fundamental elements … Because it seems to me certain that it will always be quite impossible to explain the mind on the basis of neuronal action within the brain, and because it seems to me that the mind develops and matures independently throughout an individual’s life as though it were a continuing element, and because a computer (which the brain is) must be programmed and operated by an agency capable of independent understanding, I am forced to choose the proposition that our being is to be explained on the basis of two fundamental elements. This, to my mind, offers the greatest likelihood of leading us to the final understanding toward which so many stalwart scientists strive. (MM 80)

What led him to this conclusion? Two features in particular had impressed Penfield. First, given his specialization in epilepsy cases, he was, unsurprisingly, impressed by the phenomena of epileptic automatism. Second, he was powerfully struck by the responses elicited from patients in reaction to electrode stimulation during surgery.

Penfield’s interpretation of epileptic automatism

A patient, suffering an epileptic seizure that has induced automatism, will often continue to execute whatever more or less stereotypical tasks he was engaged in. He will, however, be in a fugue condition – that is, after recovery he will remember nothing of what he has done during the seizure. Penfield interpreted automatism as showing that the epileptic seizure disconnected the mind from what, following Hughlings Jackson,²³ he called ‘the brain’s highest mechanism’ (a precursor of Eccles’s ‘liaison brain’). He took it that the brain, during the period of automatism, is controlling the behaviour of a ‘human automaton’ in accordance with antecedent ‘programming’ by the mind. Just as the programming of a computer comes ‘from without’, so too the programming of the brain, which is, Penfield claimed, a biological computer, is effected by the mind via the brain’s highest mechanism. Purpose comes to it from outside its own mechanisms. Short-term programming obviously serves a useful purpose, making possible automatic continuation of routine tasks, and this is visibly and strikingly manifest during periods of such epileptic seizures.

That this highest mechanism, most closely related to the mind, is a truly functional unit is proven by the fact that epileptic discharge in gray matter that forms a part of its circuits, interferes with its action selectively. During epileptic interference with the function of this gray matter … consciousness vanishes, and with it goes the direction and planning of behaviour. That is to say, the mind goes out of action and comes into action with the normal functioning of this mechanism.

The human automaton, which replaces man when the highest brain-mechanism is inactivated, is a thing without the capacity to make completely new decisions. It is a thing without the capacity to form new memory records and a thing without that indefinable attribute, a sense of humour. The automaton is incapable of thrilling to the beauty of a sunset or of experiencing contentment, happiness, love, compassion. These, like all awarenesses, are functions of the mind. The automaton is a thing that makes use of reflexes and skills, inborn and acquired, that are housed in the computer. (MM 47)

Though Penfield ventured no testable hypotheses about how this interaction occurs, he claimed that the highest brain mechanism is, as it were, the mind’s executive. It accepts directions from the mind, and passes them on to the various mechanisms of the brain (MM 84). The mind directs the brain in action. It has no memory of its own. But the contents of the stream of consciousness are recorded in the brain (as seems evident from the inadvertent retrieval of long-lost memories during cortical stimulation of the brain during operations). So, when the mind needs to retrieve a memory, in a flash it opens the files of remembrance in the brain through the highest brain mechanism (MM 49).

Penfield’s interpretation of phenomena consequent on electrode cortical stimulation

Reflection on some of the phenomena consequent on cortical stimulation during operations led Penfield to similar conclusions. So, for example, a patient, whose ‘speech cortex’ was interfered with by an electrode, exhibited exasperation when he could not identify a picture of a butterfly. On withdrawal of the electrode, he said, ‘“Now I can talk. Butterfly. I couldn’t get the word “Butterfly”, so I tried to get the word “moth”!’ It is interesting to see how Penfield construed and explained this temporary impairment of the patient’s normal identificatory powers.

It is clear that while the speech mechanism was temporarily blocked he could perceive the meaning of the picture of a butterfly. He made a conscious effort to ‘get’ the corresponding word. Then, not understanding why he could not do so, he turned back for a second time to the interpretative mechanism … and found a second concept that he considered the closest thing to a butterfly.²⁴ He must then have presented that to the speech mechanism, only to draw another blank.(MM 52)

According to Penfield, concepts are stored away in the mind’s concept mechanism in the brain, from which the mind selects the concept it requires. That concept is then presented in the stream of consciousness, and if the mind approves of the selection, the highest brain mechanism flashes this non-verbal concept to the speech mechanism, which, when functioning normally, will present to the mind the word that is appropriate for the concept (MM 53).

Penfield was equally impressed by the fact that when neural stimulation to the brain caused a hand movement, the patient invariably responded, ‘I didn’t do that. You did.’ And equally, when cortical stimulation caused vocalization, the patient said, ‘I didn’t make that noise. You pulled it out of me.’ It was striking that no form of electrical stimulation to the cortex could induce a patient to believe or to decide (MM 77). It is not surprising that Penfield drew the conclusion that belief and volition are functions of the mind.

The mind and its interaction with the brain via ‘the highest brain mechanism’

A man’s mind, Penfield concluded, is the person (MM 61). It is the mind that is aware of what is going on, that reasons and decides, and understands (MM 75f.). The person walks about the world, depending always upon his ‘private computer’ (i.e. his brain), which he programs continuously (MM 61). The highest brain mechanism is the meeting place of mind and brain, the psychophysical frontier (MM 53). The mind, in making decisions, causes the highest brain mechanism to send neuronal messages to other mechanisms in the brain, and data stored in the brain are admitted to consciousness. How is interaction effected? Here Penfield speculated that there must be a second form of energy (other than the electrical energy that is used by the highest brain function to innervate the nervous system) which is available to the mind. This, he conjectured, must be made available to the mind in its waking hours by the highest brain mechanism.

The mind vanishes when the highest brain mechanism ceases to function due to injury or due to epileptic interference or anaesthetic drug. More than that, the mind vanishes during deep sleep.

What happens when the mind vanishes? There are two obvious answers to that question; they arise from Sherrington’s two alternatives – whether man’s being is to be explained on the basis of one or two elements. (MM 81)

Penfield thought it preposterous to suppose that the mind is merely a function of the brain, and so ceases to exist when it ‘vanishes’ in sleep or epileptic automatism and is re-created afresh each time the highest brain mechanism functions normally. Rather, he concluded, the mind is ‘a basic element’, and has a ‘continuing existence’. ‘One must assume’, he wrote, ‘that although the mind is silent, when it no longer has its special connection to the brain, it exists in the silent intervals and takes over control when the highest brain mechanism goes into action’ (MM 81). So, the highest brain mechanism switches off the power that energizes the mind whenever one goes to sleep, and switches it on again when one awakens.

Is the explanation improbable?, Penfield queried.

It is not so improbable, to my mind, as is the alternative expectation [explanation] – that the highest brain mechanism should itself understand, and reason, and direct voluntary action, and decide where attention should be turned and what the computer must learn and record, and reveal on demand. (MM 82)

Penfield’s neo-Cartesianism

Penfield’s neo-Cartesianism is no advance over that of Sherrington and Eccles. But if we are to learn anything from his errors, we must not simply dismiss them as misguided and move on to other matters. That will merely ensure that we learn nothing from his endeavours. We must ask what went wrong, what drove one of the greatest neurosurgeons and neurologists of all times to embrace such a misconceived view of the mind and brain?

Shared presuppositions

(1) The Cartesian conception of the mind

It should be noted that there are at least two fundamental presuppositions that Penfield shared with Sherrington and Eccles. The first was a Cartesian conception of the mind. Like Descartes, Penfield conceived of the mind as an independent substance (or, as he puts it, ‘a fundamental element’ that has ‘continuing existence’). Like Descartes, he identified the person, the ‘I’, or the ‘self’ with the mind, rather than with the living human being. Like Descartes, he took the mind to be the bearer of psychological attributes,²⁵ and consequently conceived of human beings as subjects of psychological predicates only derivatively. And, like Descartes, he took the mind to be a causal agent that can bring about changes in the body by its actions.

(2) The assumption that the question of whether brain mechanisms can account for the mind is an empirical one

The second presupposition is that the question which so deeply disturbed him – namely, whether brain mechanisms account for the mind, whether the mind can be explained by reference to what is known about the brain – is an empirical question. Like Sherrington, Penfield conceived of the matter as a choice between two different empirical hypotheses. Either we can explain everything the mind does – for example, thinks and believes, reasons and concludes, has wants, forms intentions and purposes, and decides to act – by reference to neural states and events, or we must conceive of the mind as an independent substance in immediate causal interaction with the brain, and hence with the body. The choice between these two hypotheses is to be determined by the evidence that supports them severally and by their relative explanatory powers.

Criticisms of Penfield’s presuppositions:

(1) Misconceptions about the nature of the mind

Both presuppositions are misconceived. The mind, as we have already intimated, is not a substance of any kind. Talk of the mind is merely the form in which we represent to ourselves human powers and their exercise and talk about human powers and their exercise. We say of a creature (primarily of a human being) that it has a mind if it has a certain range of active and passive powers of intellect and will – in particular, conceptual powers of a language-user that make self-awareness and self-reflection possible. The idioms that involve the noun ‘mind’ have as their focal points thought, memory and will. And they are all readily paraphrasable into psychological expressions in which the word does not occur (we shall discuss this matter in some detail in §4.8).

A person is not identical with his mind. A mind is something ( but not some thing) a person is said to have, not to be. In having a mind, an animal (that is thereby also a person, and a bearer of rights and duties) has a distinctive range of capacities. And it is obvious both that an animal cannot be identical with an array of capacities, and that if a human being loses enough of those distinctive capacities, he can cease to be a full-blooded person (and exist only in a ‘vegetal state’). It is not the mind that is the subject of psychological attributes, any more than it is the brain. It is the living human being – the animal as a whole, not one of its parts or a subset of its powers. It is not my mind that makes up its mind or decides; it is not my mind that calls something to mind and recollects; and it is not my mind that turns its mind to something or other and thinks – it is I, this human being. Hence, too, the mind is not a causal agent that brings about changes in the body and its limbs by its actions. On the contrary, it is human beings that deliberate, decide and act, not their minds.

(2) Whether the brain can ‘account for’ the mind is not an empirical question

Consequently, Penfield’s second presupposition is misguided. Whether we can ‘account for the mind’ in terms of the brain alone, or must account for the (supposed) activities of the mind (e.g. thought, reasoning, wants and purposes, intentions and decisions, voluntary and intentional actions) by reference to the mind itself, conceived of as an independent substance and therefore causal agent, is not a matter of choice between two empirical hypotheses. If these were empirical hypotheses, then either could in principle be true; that is, both would present intelligible possibilities, and it would be a matter of empirical investigation to discover which is actually the case. But that is not how it is at all.

It is neither the brain nor the mind that is the subject of psychological attributes

First, it is not the mind that thinks and reasons, wants things and has purposes, forms intentions and makes decisions, acts voluntarily or intentionally. It is the human being. We do indeed characterize someone as having a clear, rigorous or decisive mind. But these are merely ways of characterizing his dispositions in respect of thought and will. If we want to understand why a normal human being reasoned the way he did, thought what he did, has the goals and purposes that he has, and why he decided as he did, formed such-and-such intentions and plans, and acted intentionally, no neurological account will clarify for us what we wish to be clarified. To this extent Penfield was right. Where he was wrong was in the supposition that what we need is an explanation in terms of the activities of the man’s mind – where the mind is conceived of as an agent with causal powers. Rather, what we want is an explanation in terms of his reasoning, hence too by reference to what he knew or believed, and, in the case of practical reasoning, by reference to his goals and purposes. And if our explanation renders his reasoning intelligible, no further information about neural events in his brain can add anything. All a neural explanation could do would be to explain how it was possible for him to reason cogently at all (i.e. what neural formations must be in place to endow a human being with such-and-such intellectual and volitional capacities), but it cannot rehearse the reasoning, let alone explain its cogency.

Neither the causal agency of the brain nor the causal agency of the mind explains intentional action

Similarly, if we are puzzled by someone’s actions, if we wish to know why A signed a cheque for £200, no answer in terms of brain functions is likely to satisfy us. We want to know whether A was discharging a debt, making a purchase, donating money to charity or betting on a horse – and once we know which of these is the case, we may also want to know what A’s reasons were. A description of neural events in A’s brain could not possibly explain to us what we want to have explained. If we wish to know why A caught the 8.15 a.m. to Paris, a description of neural events cannot in principle satisfy our need for an explanation. But the answer that he had a committee meeting there at 2.00 p.m. that was to decide upon such-and-such a project for which A is responsible may give quietus to our curiosity. If A has murdered B, we may wish to know why. We may be given a reason, and still remain dissatisfied, wishing to understand more – but the ‘more’ we wish to understand is most probably A’s motive, not what neural events occurred at the time of the killing. We want to know whether he killed B out of revenge or out of jealousy, for example, and that requires a quite different narrative from anything that neuroscientific investigation could produce. Explanation of action by redescription, by citing agential reasons, or by specifying the agent’s motives (and there are many other forms of explanation of related kinds) are not replaceable, even in principle, by explanations in terms of neural events in the brain. This is not an empirical matter at all, but a logical or conceptual one. The type of explanation is categorially different, and explanations in terms of agential reasons and motives, goals and purposes, are not reducible to explanations of muscular contractions produced as a consequence of neural events (see chapter 16). But equally, such explanations are not couched in terms of the activities of the mind, conceived as an independent substance with causal powers of its own. In this sense, Penfield’s dilemma is a bogus one. He was perfectly right to think that one cannot account for human behaviour and experience in terms of the brain alone, but wrong to suppose that the idea that one might be able to do so is an intelligible empirical hypothesis as opposed to a conceptual confusion. He was also wrong to suppose that the alternative is accounting for human behaviour and experience in terms of the causal agency of the mind, and wrong again in thinking that that too is an empirical hypothesis. There is no need whatsoever to impale oneself on either of the horns of Penfield’s dilemma.

The hypothesis that mind–brain interaction can explain human behaviour is logically incoherent

Once these presuppositions are jettisoned, it becomes easier to see why the explanation of human behaviour in terms of the interaction of the mind (conceived as an independent substance) and the brain is misconceived. It is not a false empirical hypothesis, but a conceptual confusion. For inasmuch as the mind is not a substance, indeed not an entity of any kind, it is not logically possible for the mind to function as a causal agent that brings about changes by acting on the brain. This is not an empirical discovery, but a conceptual clarification. ( But it is equally mistaken to suppose that substituting the brain for the Cartesian mind is any less confused. That too is not an empirical hypothesis, but a conceptual muddle, which likewise stands in need of conceptual clarification.)

Neither epileptic automatism nor electrode stimulation of the brain support dualism

Consequently, Penfield was mistaken to think that what so impressed him – namely, the phenomena of epileptic automatism and the various facts that characterize electrode stimulation of the brain – constitute empirical support for a dualist hypothesis. Epileptic automatism does not show that the mind has become disconnected from the ‘highest brain mechanism’ to which it is normally connected, and by which it is supplied with energy of an as yet unknown form.²⁶ What it shows is that during an epileptic seizure, as a consequence of the abnormal excitation of parts of the cortex, the person is temporarily deprived of some of his normal powers (including memory, the ability to make decisions, emotional sensitivity and a sense of humour), while other powers, in particular those for routine actions, are retained. The phenomena are indeed striking, but they amount to a dissociation of powers that are normally associated, not to a disconnection of substances that are normally connected. They do not show that the brain is a computer or that the mind is its programmer. The brain is no more a computer than it is a central telephone exchange (the previously favoured analogy), and the mind is no more a computer programmer than it is a telephonist. It is perfectly true that the ability to continue routine tasks unreflectively is useful (and the expression ‘short-term programming’ is an apt metaphor here). It is also true that the purposes pursued by someone are not the purposes of his brain. But it does not follow from this that they are the purposes of his mind. They are the purposes of the human being – and they are to be understood in terms of facts about human life, social forms of life, antecedent events, current circumstances, agential beliefs and values and so forth, not in terms of neural events and mechanisms. But it is, of course, true that, but for the normal functioning of the brain, a human being would not have, and would not be able to pursue, the normal kinds of purposes that we do pursue.

The various phenomena that characterize electrode stimulation of the brain are similarly misconstrued by Penfield. The case of interference with the ‘speech cortex’ does not show that there is any such thing as a ‘concept mechanism’ in the brain that stores non-verbal concepts that can be selected by the mind and then presented to the speech mechanism to be matched to the word that represents the concept. That is picturesque mythology, not an empirical theory. Words are not names of concepts, and do not stand for concepts, but rather express them. Concepts are abstractions from the use of words. The concept of a cat is what is common to the use of ‘cat’, ‘chat’, ‘Katze’, etc. The common features of the use of these words is not something that can be stored in the brain or anywhere else independently of a word (or symbol) that expresses the concept. The patient whom Penfield describes could not think of the word ‘butterfly’ with which to identify the object in the picture presented to him. He knew that the object belonged to a class which resembles a different class of insects (viz. moths), and tried, equally unsuccessfully, to think of the word for members of the second class. This temporary inability is incorrectly described as knowing the concept but being unable to remember the word for it. The supposition that the mind might be presented with non-verbal concepts from which to choose presupposes that there is some way of identifying non-verbal concepts and distinguishing one from the other independently of any words or symbols that express them. But that makes no sense (see §§15.1–15.2).

It is certainly interesting that Penfield found that electrode stimulation could not induce either belief or decision. But this does not show that believing and deciding are actions of the mind, any more than it shows that they are not actions of the brain. It is true that they are not actions of the brain – but that is not an empirical fact that might be shown to be the case by an experiment. Rather, there is no such thing as the brain’s believing or deciding (any more than there is such a thing as checkmate in draughts). But it is also true that they are not actions of the mind either. My mind does not believe or disbelieve anything – I do (although, to be sure, that is no action). Nor does it decide – it is human beings that decide and act on their decisions, not minds.

That the exercise of mental powers is a function of the brain does not show that behaviour and experience are explicable neurally

Penfield objected vehemently to the suggestion that the mind is a function of the brain, and supposed that if it were, then the mind would cease to exist during sleep or epileptic automatism. The suggestion is unclear, but one may surely say that the distinctive powers of intellect and will of a creature that has a mind are a function of the creature’s brain (and of other factors too). It does not follow (as Penfield evidently feared it would) that the behaviour and experience of such a creature in the circumstances of life is explicable in neural terms. But nor does it follow that the mind ceases to exist during sleep or epileptic seizure – any more than one’s knowledge and beliefs, intentions and projects, cease to exist when one is asleep. Penfield was rightly impressed by the fact that ‘the mind develops and matures independently throughout an individual’s life as though it were a continuing element’ (MM 80). But he was misled by his unquestioned assumption that the mind is a kind of agent. Had he thought of the mind in more Aristotelian terms as a set of powers and capacities (second-order powers), he would have been closer to the truth, and less prone to conceptual illusion. For the continuous possession of powers and capacities is not interrupted by sleep or even by epileptic automatism, even though the agent cannot exercise some of his normal powers during the seizure. And the developing unity of a person’s mind is not the development of a substance distinct from the human being himself, but rather the emergence of a determinate character and personality, an intellect with certain distinctive characteristics and a will with a coherent array of preferences, all of which are traits of the person – of the living human subject who has rights, duties and obligations, and who is responsible for his acts, actions and activities.

What Penfield thought the less plausible alternative to dualism is the view currently favoured

Penfield thought that a form of Cartesian dualism is more probably correct than what he conceived to be the alternative: namely, ascribing understanding, reasoning, volition and voluntary action, as well as deciding, to the brain itself. It is very striking and important that the strategy that Penfield conceived to be altogether improbable is precisely the route that is currently adopted by the third generation of neuroscientists, who ascribe psychological functions to the brain. This is the subject of the next chapter.