Madness Explained, page 21
Different brain circuits use different neurotransmitters. The most intensively researched are acetylcholine, gamma-amino butyric acid (GABA), glycine, serotonin (5-hydroxytryptamine), noradrenaline and dopamine. However, a large number of peptides may also function as neurotransmitters (a recent textbook of biological psychiatry54 lists twenty-nine candidate substances). Until only a few years ago, it was thought that each neurone used only one type of neurotransmitter. However, it now appears that some neurones use both a classical neurotransmitter and a peptide. To complicate matters further, there are also a number of different kinds of receptors for each neurotransmitter. For example, in the case of dopamine, five main types are known, which, by convention, are labelled D1 to D2.
Figure 7.3 The structure and function of the dopamine synapse (reproduced from R. W. Heinrichs (2001) In Search of Madness: Schizophrenia and Neuroscience. Oxford: Oxford University Press). Dopamine is synthesized from tyrosine and stored in vesicles. When released into the synapse, it binds with receptors on the post-synaptic membrane. Monoamine oxidase breaks dopamine down to yield homovanillic acid. Dopamine is also taken back into the pre-synaptic cell. Drugs such as amphetamine and cocaine release and prevent the uptake of dopamine. Anti-psychotic drugs such as haloperidol and chlorpromazine occupy post-synaptic receptors and thereby block dopamine occupancy.
The biochemical approach to psychosis attempts to explain how dysfunctional neurotransmission can lead to symptoms. Theories of this kind have usually been stimulated by the accidental discovery of drugs that either mimic psychosis or ameliorate patients’ symptoms. For example, in 1938 the Swiss chemist Albert Hoffman first synthesized the powerful hallucinogen, lysergic acid diethylamide (LSD). At the time he was exploring the therapeutic properties of compounds derived from naturally occurring ergot in the hope that these would prove useful for controlling muscle spasms during pregnancy. Animal tests indicated that LSD had no unusual behavioural effects, and there the matter might have rested had Hoffman not accidentally inhaled a dose of the drug (it is active in microgram quantities) some years later. He described his experiences afterwards as follows:
In the afternoon of 16th April, 1943… I was seized by a peculiar sensation of vertigo and restlessness. Objects, as well as the shapes of my associates in the laboratory, appeared to undergo optical changes. I was unable to concentrate on my work. In a dreamlike state I left for home, where an irresistible urge to lie down overcame me. I drew the curtains and immediately fell into a peculiar state similar to drunkenness, characterized by exaggerated imagination. With my eyes closed, fantastic pictures of extraordinary plasticity and intensive colour seemed to surge towards me. After two hours this state gradually wore off.55
Not surprisingly, this discovery encouraged some researchers to argue that schizophrenia might be caused by some kind of endogenous (self-created) hallucinogenic substance.56 In 1962, a dramatic breakthrough was reported which seemed to support this theory. Researchers saw a ‘pink spot’ (which was later found to be the chemical 3, 4-dimethoxyphenylethylamine) when the urine of schizophrenia patients was allowed to diffuse on chromatography (blotting) paper. The spot was absent when the procedure was repeated with urine from ordinary people.57 Unfortunately, it later turned out that the spot was of dietary origin and could be found in the urine of anyone who ate institutional food.
Following the discovery that LSD interferes with the neurotransmitter serotonin, some neurochemists embarked on a search for abnormalities in the relevant neural pathways of schizophrenia patients.58 However, this approach rapidly fell out of favour for two main reasons. First, as will be discussed in detail in a later chapter, the experiences induced by hallucinogenic drugs are quite dissimilar to those reported by psychotic patients.59 Second, and more importantly, the serotonin hypothesis was displaced in the minds of most scientists by an alternative theory, which seemed to promise a simple and complete biological account of madness. This theory was, of course, the famous dopamine hypothesis, which was stimulated by Laborit’s accidental discovery of the anti-psychotic properties of chlorpromazine.
Arvid Carlsson, a pharmacologist at the University of Lund in Sweden, first discovered the neurotransmitter function of dopamine in 1957. Soon afterwards, with his colleague Margit Lindqvist, Carlsson obtained evidence from animal studies that chlorpromazine and haloperidol (another neuroleptic) occupied dopamine receptors, blocking them from being stimulated by dopamine.60 Subsequent experiments by the pharmacologist Paul Jannsen and others confirmed that the potency of the various neuroleptic drugs that had, by that time, become available, correlated with their capacity to do this.61 The more readily they occupied the receptors, the lower the dose required to produce an anti-psychotic effect. This discovery led a number of researchers to conclude that schizophrenia must be caused by some kind of abnormality in the dopamine system.
A parallel observation that seemed to support this hypothesis concerned the effects of abusing the stimulant drug amphetamine. Early reports that this sometimes led to a psychotic reaction were confirmed in 1958 when a British psychiatrist, Philip Connell, published a monograph describing forty-two patients who had become ill in this way. Usually, the patients became paranoid and experienced visual hallucinations, but these symptoms typically remitted within a few days of discontinuing the drug.62 Connell’s observations were later supported by the results of a remarkable series of experiments conducted in the United States by Burt Angrist and Samuel Gershon, in which normal volunteers were persuaded to take large hourly doses of amphetamine. The effects of this procedure were sometimes dramatic. For example, a volunteer:
who had taken 465 mg of amphetamine over 22¾ hr abruptly experienced a florid paranoid psychosis. Before the experiment he had made a ‘deal’ with the attendant on the ward, to whom he owed several dollars. As he became psychotic, he ‘heard’ a gang coming on the ward to kill him (sent by the attendant). His paranoid feelings included the experimenter who he assumed had ‘set up’ the ‘trap’. He was at times quite hostile. Explanations that his feelings were amphetamine-induced were rejected with sardonic mock agreement… He had visual hallucinations of gangsters, of doors opening and closing in the shadows, and visual illusions, in which papers on the bulletin board ‘turned into’ ‘a gangster in a white coat’.63
As it was known that amphetamine causes a functional excess of dopamine in the brain, the irresistible conclusion was that over-activity of the dopamine system causes psychosis whereas dampening down the system by blocking dopamine receptors makes psychotic patients better. This theory not only neatly explained the benefits of neuroleptics, but also some of their negative effects as well. Parkinson’s disease, characterized by stiffness, tremor and other movement disorders, is caused by a depletion of the brain’s supply of dopamine. Neuroleptics often cause Parkinsonian symptoms as side effects, which is precisely what should be expected if they act by dampening the dopamine system. Conversely, the drugs used to treat Parkinson’s disease (for example, levo-dopa, which is metabolized into dopamine after crossing the blood–brain barrier) sometimes cause hallucinations and delusions in people who have previously had no experience of psychosis.
Unfortunately, time has not been particularly kind to the dopamine hypothesis. As we saw in Chapter 4, one problem is that the theory has rested on a false premise – the specificity of neuroleptics for schizophrenia. For this reason, Carlsson has recently acknowledged that the theory should be renamed ‘the dopamine hypothesis of psychosis’.64 Unfortunately, other deficiencies of the theory are much less easily overcome.
Whereas blockade of the dopamine receptors is usually achieved within an hour or so of taking a single dose, patients normally have to take their medication for several weeks before experiencing any benefit.65 More embarrassing for the theory is the fact that some psychotic patients fail to obtain any benefit at all, even after many months of treatment.66 In the 1980s it became possible to use PET scans to estimate the extent to which receptor blockade is achieved in individual patients. This was accomplished by detecting photon emissions from patients who had been injected with a neuroleptic that had been ‘radio-labelled’ (made radioactive). Studies using this technique established that receptor blockade is as complete in non-responding patients as in responding patients.67 To make matters even more complicated, a number of atypical neuroleptic compounds have since been discovered and, although effective at reducing symptoms, these drugs do not seem to have the same affinity for dopamine receptors as the older, more established compounds.68 Taken together, these findings add up to compelling evidence that dopamine receptor blockade is not a sufficient condition for bringing about a reduction in psychotic symptoms.
An even greater problem for the theory is researchers’ failure to demonstrate direct evidence of dopamine abnormalities in the brains of patients, despite several decades of hard labour. According to the most obvious interpretation of the theory, schizophrenia should be associated with an excess of dopamine. Some investigators have attempted to test this prediction by measuring dopamine levels in the brain at post mortem, but the results have been inconsistent. Others have therefore attempted to measure homovanillic acid (HVA, a metabolite or breakdown product of dopamine) in the cerebrospinal fluid of living patients. These levels should be high if there is excessive dopamine in the brain. However, the majority of studies that have addressed this issue have reported no difference between schizophrenia patients and ordinary people.69
A second interpretation of the dopamine hypothesis, which follows logically from the failure of the first, suggests that psychosis is associated with an excess of dopamine receptors. Although early postmortem studies seemed to support this version of the theory, it later transpired that increased receptor density could be the consequence of consuming neuroleptics for many years. (Animal studies have confirmed that the brain responds to prolonged neuroleptic treatment by sprouting more receptors.)70 This objection led to an experiment which British psychiatrist Peter McKenna has(wrongly, in my view) described as, ‘arguably the most important… in the history of psychiatry’71 – the attempt to measure the density of dopamine receptors in living patients who have never received drug treatment.
This was first achieved in 1986 by a group of researchers at Johns Hopkins University in the USA, who used PET to assess ten drug-naive patients after they had been injected with a radio-labelled neuroleptic in sufficient dose to ensure complete dopamine receptor occupancy.72 Their announcement that DA2 receptor density was elevated in the patients appeared to herald a major breakthrough in the search for the biological origins of mental illness, but within months this was followed by a report from another group who were unable to obtain the same result.73 Several years later, the Johns Hopkins University group announced that they had found increased DA2 receptor density in bipolar disorder patients who were suffering from psychotic symptoms.74
More recent research has far from clarified these findings. For example, in a longitudinal investigation, a group of researchers in Finland administered PET scans at intervals of between six and twenty-four months to four schizophrenia patients who had never taken neuroleptic drugs. Two patients who were highly psychotic at the outset showed a considerable reduction in their symptoms between scans. One showed a corresponding 24.7 per cent decrease in the density of his DA2 receptors, whereas the other showed an increase in receptor density of 13.5 per cent. The remaining two patients experienced only very modest changes in their symptoms, but one showed a 56 per cent reduction in DA2 receptor density, while the other showed a 24.6 per cent increase.75
As the dopamine hypothesis has unravelled, investigators have studied the role of other neurotransmitters in psychosis, although with equally inconclusive results.76 It is therefore difficult to escape the conclusion that biochemical research into psychiatric disorders has failed to live up to its initial promise. Of course, we should not therefore infer that this type of research is doomed to failure. Anyone who has ever visited a pub knows that changes in the chemistry of the brain can bring about dramatic changes in behaviour and experience. However, even if biochemical peculiarities can be found in psychiatric patients, they may tell us very little about the origins of psychosis.77 Consider the results of two recent studies. In the first, researchers at the Department of Psychiatry at Columbia University used functional imaging techniques to record the brain’s biochemical reaction to a dose of amphetamine. They reported that the consequent increase in dopamine synthesis was excessive in schizophrenia patients who were currently ill. However, the responses of patients who were currently well were quite normal.78 This finding suggests that, although dopamine abnormalities may be associated with the generation of symptoms, they may not be of primary aetiological significance. In the second, researchers at the Ralf H. Johnson Veterans Affairs Medical Centre in Charleston, USA, measured dopamine beta-hydroxylase (an enzyme that converts dopamine to noradrenaline) in the plasma of patients suffering from post-traumatic stress. Higher levels of the enzyme were found in a subgroup of the patients who were suffering from psychotic symptoms.79 This finding suggests that, when dopamine abnormalities are present in psychotic patients, they may be part of the brain’s response to emotional trauma.
The Limits of Biology
We embarked on this review of biological studies of madness in order to answer two questions. First, we wanted to establish whether the discovery of brain abnormalities in mad people precludes a psychological analysis of their problems. Second, we wanted to evaluate the competing claims about the disease status of psychosis made by the neoKraepelinians and Thomas Szasz (recall that both presume that this issue hangs on whether or not neuropathology can be demonstrated in patients). A sceptic might attempt to address both of these questions by asserting that the presence of biological abnormalities in psychosis remains unproven (the position taken by Szasz). But to make this claim is to doubt too much. Whereas many of the biological hypotheses that we have considered appear to be unsupported by data, it does appear that the brains of psychotic patients are (perhaps subtly) different from the brains of ordinary people. The most compelling evidence is that obtained from structural imaging. Whereas early studies clearly exaggerated the extent to which the cerebral ventricles of patients are enlarged in comparison with those of ordinary people, the less dramatic differences found in recent and more carefully conducted investigations cannot be dismissed so easily.
In order to address the first of our two questions, therefore, we need to think more carefully about the implication of these observations. Many biological psychiatrists have argued for a crude form of biological reductionism in which it is assumed that brain abnormalities reflect aetiological processes that are unconnected to the social environment. However, this assumption is plainly false, because, as we have also seen, brain abnormalities alone provide few clues about aetiology. Because our brains are affected by our experiences, peculiarities in the size of its anatomical components, in neuroactivations when we perform particular tasks, or in the biochemical transactions between neurones, can be as readily attributed to the impact of the environment as to causative biological factors such as early brain damage, viruses or endogenous neurotoxins. (This is not to say that brain damage, viruses or endogenous neurotoxins play no role in psychosis, but that other kinds of evidence – which I will discuss in later chapters – will be required to demonstrate that they do.) Curiously, this limitation of brain research was better understood by Emil Kraepelin than by many contemporary psychologists and psychiatrists. It was for this reason that Kraepelin regarded aetiology and pathological anatomy as providing, in principle at least, independent methods of classifying patients.
The problem here seems to be a persisting naivety about the relationship between the biological and the psychological, which has deep roots within our culture. Its origins predate Descartes, who, three and a half centuries ago, articulated the philosophical doctrine of dualism, which holds that the mind and brain are different kinds of substances, the former non-material and the latter physical. This idea is so entrenched in our language and our folk psychology that, even today, many people find it difficult to think in any other way. It is this prejudice that has led both biological researchers and anti-psychiatrists to assume that the processes underlying abnormal behaviour must be attributed either to biology or to psychology. The remedy for this confusion is, of course, a better understanding of relationships between the brain, mental events such as thoughts and feelings, and the social and physical environment. Clearly, it will not be possible to provide a completely satisfactory account of these relationships in a few paragraphs. (The problem of consciousness is particularly intractable, and continues to test the best minds in the business.) For present purposes, it is sufficient to note that most attempts to solve this problem assume that statements about brain biology and statements about mental processes represent different levels of descriptions of the same phenomena – in other words, that the mind and the brain are one.
One implication of this unity is that we are likely to make more progress in understanding the biology of psychosis if we first attempt to understand the psychology. Once we know how the environment influences patients’ complaints, and the mental processes that underlie them, we will be in a much better position to make sense of the data obtained from biological investigations.
