Timothy is trying to explain what it feels like to have a near-constant stream of spoken thoughts intruding into his mind from outside. He holds the remote-control handset in front of the TV. ‘Suppose I want to watch 2. Well, the thoughts in my head will say, ‘Why don’t you press 3? We want to watch 3′.’
That’s just the trivial, everyday irritation of an illness that has profoundly affected Timothy’s whole life. In fact, he has lost almost everything. He has not seen his son for eight years. His marriage broke down more than a decade ago and he hasn’t had a relationship since. He has given up his business and his flat. For someone whose life has been torn apart by disease, he does not seem bitter. But then perhaps he is just being polite.
Advertisement
Timothy (not his real name) has schizophrenia. Half a million others in Britain, and 2 million in the US, will suffer from the disease at some time in their lives, making it one of the commonest health problems in industrialised societies. For each sufferer, a string of relatives will also be affected. One estimate suggests that schizophrenia costs Britain about £2.7 billion a year, or 2 per cent of GNP.
Bizarre symptoms
Yet our knowledge of the roots of the disease and our ability to treat it remain primitive. Almost a century after the syndrome was first labelled, psychiatrists must still rely on signs and symptoms alone to diagnose schizophrenia. Drugs have improved, but their effect is mostly to dampen down the most bizarre symptoms, leaving equally disabling aspects of the disease untouched.
Meanwhile, public ignorance about schizophrenia seems as great as ever. While a scattering of violent attacks has led the media to stereotype schizophrenics as a threat to society, the truth is that overall, they are no more likely to be violent than anyone else. The only person a schizophrenic is likely to harm is himself or herself; one study found that almost a third of schizophrenics had tried to commit suicide at some time.
All this might sound bleak, but research into schizophrenia is slowly beginning to gather momentum. Imaging techniques are helping to identify physical abnormalities in schizophrenics’ brains, and now scientists are trying to link these abnormalities with specific symptoms of the disease. Geneticists are searching for the ‘schizophrenia gene’ or genes (see ‘Elusive genes’), applying the techniques that have been successful in pinpointing the genes behind disorders such as cystic fibrosis. Molecular biologists are analysing differences identified in the architecture of nerve cells in the brains of schizophrenics and abnormalities in receptor molecules on the surfaces of these cells. And a few researchers even claim to be making progress in understanding the disease through experiments on animals.
The main message from all these separate strands is clear; schizophrenia is a biological disorder of the brain, not some mysterious demon. Increasingly, the evidence points to abnormalities that arise very early in life, probably before birth, which disrupt the normal development of the brain. The challenge facing scientists is not only to uncover the precise nature of those abnormalities and devise methods of treating them, but also to show how they relate to the disturbances of mind suffered by schizophrenics. The task will keep researchers busy for years, but if they succeed the benefits of their work will bring them up against some of the biggest questions in neuroscience – the neural roots of personality, language and perhaps even consciousness itself.
First, a rapid tour of what little we already know. Schizophrenics do not, as popular myth holds, have split personalities. Rather, the disease causes a wider fragmentation of their intellect and social selves, attacking the very qualities that make us human. Schizophrenics often have difficulty communicating in language or facial expression and are unable to put themselves in someone else’s shoes, or to judge others’ intentions towards them. There are clear physical and structural differences in schizophrenics’ brains. For example, studies using computerised tomography (CT) scans have shown that the lateral ventricles tend to be significantly bigger than in healthy brains. And a form of magnetic resonance imaging has revealed that schizophrenic brains often have a smaller volume of tissue in the left temporal lobe than normal brains. But the links between such abnormalities and symptoms are still unclear.
Ferrying impulses
In chemistry as well as structure, the schizophrenic brain differs from the normal brain. For example, there are complex differences which have yet to be fully understood in the way it handles dopamine, a neurotransmitter, that ferries impulses between nerve cells (see ‘The dopamine link’).
Increasingly, scientists believe that schizophrenia may be a syndrome caused by more than one underlying disease. If the different diseases could be singled out, scientists argue, then the search for genes and other causative factors might be much easier.
In psychiatrists’ jargon, schizophrenics’ symptoms are divided into so-called ‘positive’ and ‘negative’ categories. Positive symptoms include hearing voices or others discussing you in the third person, paranoia and thinking that thoughts are coming into your head from elsewhere.
Frightening as these experiences must be, it is the so-called negative symptoms that may be more disabling in the long term. People with schizophrenia tend to be extremely poor communicators. Many lack the ability to make small talk and if you ask questions you will tend to get one-word answers. Critically, they also seem to have difficulties with what psychologists call working memory – the ability to keep information in mind long enough to use it for a specific task. For example, our brains use working memory to speak in sentences, holding the beginning of a sentence in mind until we finish it. And schizophrenics tend to lack the will to initiate actions, relying instead on external stimuli to do so.
So far, such knowledge has grown slowly, but now the pace is accelerating. One fashionable area involves the use of scanning techniques such as PET (positron emission tomography), which measures the flow of blood in the brain and allows scientists to identify which bits of the brain ‘light up’ as people perform psychological tests. Its enthusiasts hope to reduce schizophrenia to specific cognitive processes mapped to particular networks of neurons in the brain.
One such team, led by Chris Frith and others funded by the Medical Research Council at the Hammersmith Hospital in London, is comparing normal and schizophrenic brains as they tackle specific psychological tests. With PET, you can – in theory – watch how these tests affect the brain. But, warns Frith, it is vital to study normal brains first and, because of the broad range of symptoms that are lumped together as schizophrenia, essential also to study subgroups of patients who all have the same symptoms, such as intruding thoughts, rather than patients with varying symptoms who have all been diagnosed as schizophrenic. So far, no one has found any specific abnormalities common to all schizophrenics, but they have found that particular symptoms go hand in hand with particular abnormalities.
Cognitive tests are a way of modelling a thought process. Take, for example, the observation that schizophrenics suffer from lack of will. In cognitive terms, says Frith, this can be modelled by asking a person to guess what colour the next card in a pack will be. The point is not to guess correctly, but to be able to initiate thoughts about the colours. Healthy brains tend to generate a random sequence, such as ‘red, red, black, red, black, red’. Schizophrenics are more likely to come up with a stereotyped response, such as ‘red, black, red, black, red, black’; or ‘red, red, red, red, red, red’.
So far, studies by Frith’s group and others in Sweden and the US consistently find reduced brain activity in the prefrontal cortex in schizophrenics with negative symptoms. Occupying almost a third of the entire cortex, this cortex has evolved in humans to greater complexity and size than in any other organism. The PET results also reveal abnormalities in other brain structures, including the hippocampus, which is concerned particularly with learning and memory. Most researchers believe there is no single brain structure at fault.
In a separate study, Philip McGuire at the Institute of Psychiatry in London and colleagues scanned schizophrenics with positive symptoms while they were ‘hearing voices’. Using the technique of single-photon emission tomography (SPET), the team found that during these episodes of voice-hearing, the blood flow was greater than normal to the part of the brain known as Broca’s area, which has long been linked with articulated language. The same people were scanned again under identical conditions some weeks later when their voices had stopped; this time, there was no significant increase in the flow of blood to Broca’s area.
Language flow
Daniel Weinberger at the National Institute of Mental Health (NIMH) in Washington DC has studied identical twins where one is schizophrenic. This approach has the advantage that the twins act as each other’s controls, differences in the brain of the affected twin are far more likely to be due to the disease itself than to other factors. Weinberger’s findings add to the evidence that the prefrontal area functions abnormally in schizophrenia. They also show that in affected twins the hippocampus is smaller and becomes overactive during tasks that require working memory. Indeed, the less active the prefrontal cortex in the affected twin, the smaller the hippocampus. This strengthens the idea that schizophrenia involves faults in different areas of the brain rather than one single region.
Studies like these are still in their infancy, and there are pitfalls, warn the researchers. For example, if patients are being scanned while they attempt some task that they find difficult (such as guess the colour of the next card), isn’t it to be expected that the pattern of brain activity will differ from the pattern in people who find the task easy? Can that difference necessarily be attributed to schizophrenia? How do you account for the effects of drug treatments? And, with the exception of twin studies, how do you ensure you have a proper control group?
The research also makes unusual demands on their participants. Timothy has been involved in several studies with the Institute of Psychiatry that have required him to spend long periods of time in the scanner. ‘When they did one scan, the thoughts were really bad that day,’ he says. But he thinks the research is worthwhile. ‘So it sounds masochistic, but in a way I was pleased,’ he says.
There are more fundamental concerns, too. Some scientists, such as Robert Knight at the University of California, Davis, think that the rush to use PET may be naive unless researchers make careful use of the results. There is a risk, says Knight, that the approach could degenerate into a kind of latter-day phrenology. PET can tell you that specific areas of the brain have been activated, he says, but not how they are inter-acting. Moreover, broad brush neuroanatomical insights of the kind produced by PET may in the end reveal little about the detailed molecular or cellular basis of conditions such as schizophrenia.
Another, more extreme, attempt to reduce schizophrenia to its component parts uses animals. The idea here is to take the mental abilities that are believed to break down in the disease, look for their counterparts in animals and locate the neuronal networks that are activated when the animals display these abilities. The high priestess of this approach is Patricia Goldman-Rakic at Yale University, who is studying working memory in monkeys. You might reasonably ask how an animal without language skills can tell you anything about schizophrenics, but Goldman-Rakic is adamant: ‘There is a lot of reason to believe that the breakdown of working memory function is the essence of thought disorder.’
Her model of working memory in monkeys focuses on the animals’ ability to hold a visual image in mind. Goldman-Rakic says the team has identified subsets of neurons in the prefrontal cortex that are ‘programmed’ specifically to hold visual images in working memory. Different subsets of neurons are dedicated to handling different types of information, she says.
The researchers believe they have more support for their theory from observations of people with localised brain damage in the prefrontal cortex – caused, for example, by a tumour or a stroke. Often, such people suffer specific symptoms, such as apathy and social withdrawal, similar to certain aspects of schizophrenia. ‘There is a profound similarity in the symptoms exhibited by schizophrenics and patients with prefrontal lobe damage,’ says Goldman-Rakic.
In her latest study, she asked a group of people with prefrontal brain damage and a group of schizophrenics to perform a modified version of the monkeys’ working-memory task. Both groups’ performance on the test was impaired, she says, compared with healthy volunteers and people with damage in other regions of the brain.
Fatty substances
Goldman-Rakic and others, such as Weinberger at the NIMH and Jay Pettegrew at the University of Pittsburgh, are now taking the next step to try to identify the biochemical changes that make networks of neurons malfunction in schizophrenia. No one has made a breakthrough yet, but several teams have detected imbalances in constituents of the fatty substances called phospholipids that make up the cell membranes of neurons. Alterations in the membrane could affect the communication between cells and disrupt the performance of specific receptors for neurotrans-mitters such as dopamine or acetylcholine. For example, researchers in Scotland have found abnormalities in two long-chain fatty acids in the blood cells of people with negative symptoms. Breakdown products of the two substances are involved in the dopamine system.
While the reductionist approach of these studies may produce intriguing results, not everyone is satisfied. Most scientists seek a more complete theory to explain schizophrenia. Frith, for example, contends that the essential problem is a breakdown in the brain’s ability to monitor the information it handles. This would tend to make the brain ‘think’ that internal events were externally driven, he says – for example, voices or thoughts that are generated within the mind but perceived as someone else’s.
Flailing room
There are parallels for this monitoring system: for example, when your eyes scan around a room, the room appears to stay still. By contrast, if you gently move your eyeball from side to side by mechanically shifting it with a fingertip, the room starts to flail about alarmingly. More than a century ago, scientists concluded from this observation that the brain must be monitoring visual information and correcting it to make sense.
Frith’s theory finds some indirect support from work by Robert Knight in California. A healthy brain, says Knight, constantly filters out irrelevant stimuli – the sound of air conditioning in an office, for example – but will react sharply to new information – such as the sound of a fire alarm. By measuring the electrical activity of the brain in response to specific stimuli, Knight has shown that stroke patients’ with damaged prefrontal brains are both less efficient at screening out irrelevant information, and less likely to enhance new stimuli, than healthy brains.
No one is yet saying that Knight’s findings in stroke patients can explain the intruders in schizophrenics’ heads, but his theory of damaged filtering finds striking similarities with the way schizophrenics describe their symptoms. ‘Everything was very loud,’ said one in interviews with Fiona Macmillan, a researcher who collected the experiences of many schizophrenics during the 1980s. ‘I could hear the ash cracking off the end of the cigarette and hitting the floor.’
Attempts like these to devise an overall theory of the malfunctions that add up to schizophrenia may help scientists to frame the right research questions. But they cannot help to solve the mystery of what causes the disease, and on this matter the jury is still out. After years of supposing that the abnormalities in the brain arose at the time of the onset of disease – usually early adulthood – neurobiologists are now suggesting something very different.
Weinberger argues instead that the disorder begins early in the development of the brain, probably before the end of the second trimester of pregnancy. Either because of a genetic defect or an environmental insult such as a viral infection, the migration of neurons to their correct places in the forming brain may be disrupted, he argues. As a result, the cortex of the brain develops faultily and there are abnormal connections between the cortex and the limbic system, the group of structures including the hippocampus and the hypothalamus that are thought to be concerned with emotion and motivation.
The latest evidence for this idea comes from the Institute of Psychiatry in London, where Robert Kerwin and his colleagues have made postmortem studies of the brains of people with schizophrenia. They have found abnormalities and reductions in the numbers of some types of glutamate receptor, particularly in the regions of the temporal lobe that are physically abnormal. It is already known that certain types of glutamate receptor are involved in the development and migration of embryo neurons.
But if the seeds of schizophrenia are sown before birth, why does the disease only become noticeable a couple of decades later? One idea is that, once damaged brain architecture sets the scene, certain features of normal adult brain development trigger the disease, and this is where dopamine and environmental factors may come into play, says Weinberger. The basic idea is not new but has been given fresh impetus from experiments with rats. Based on the results, Weinberger claims that faulty development of the connections between the limbic system and the prefrontal cortex may lead to symptoms in young adulthood that are triggered by environmental stress.
Some researchers are bent on finding a deeper explanation for the causes and orgins of schizophrenia. One of the most controversial ideas comes from Tim Crow at the Clinical Research Centre at Northwick Park Hospital in Harrow, north London. Schizophrenia, he argues, is the evolutionary price we pay for the human brain’s highly developed capacity for language and social activity.
Crow’s starting point is the surprisingly uniform incidence of schizophrenia worldwide, between 0.7 and 1.4 per cent in all societies studied to date, when the disease is narrowly defined. Crow argues that this suggests a genetic disease independent of environmental triggers such as viruses. He also points out that the disease remains common even though schizophrenics are less likely to have children than the general population. So, he argues, the gene responsible must carry other advantages that help to conserve it in the gene pool.
Crow and others claim to have evidence – disputed in some circles – that schizophrenics’ brains are less likely than normal brains to show specialisation of one side of the brain. Hemispheric specialisation – or ‘asymmetry’ – is a normal characteristic in humans which varies from person to person. Numerous teams have also found that the structural and functional abnormalities in the brains of schizophrenics tend to be concentrated in the left hemisphere, which deals with language in right-handed people. Crow argues that a diversity in the population of different degrees of asymmetry is associated with normal variation in personality and intelligence. If we suppose there is a gene or genes governing asymmetry, then in Crow’s theory, schizophrenia – and other forms of psychosis such as manic depression – are the extreme and disadvantageous ends of the range, resulting from particular forms of the gene or genes.
For people with schizophrenia like Timothy, no theory is much comfort unless it can hasten the development of better drug treatments and therapies. Nevertheless, he continues to participate in research because he knows that schizophrenia runs in families – his own included – and he wants to see progress for the next generation. By the time children born into schizophrenic families today reach adulthood, there may be some hope, says Goldman-Rakic. ‘In 20 years there won’t necessarily be a cure,’ she says. ‘But I think we will maybe understand better the genes that are involved, and the changes in receptors.’ She foresees a more sophisticated armoury of drugs for the disease, which will not merely suppress the positive symptoms but also maintain normal cognitive processes, such as working memory. ‘None of this is saying that we will be able to make schizophrenics normal,’ she says, but it might make their lives more tolerable.
* * *
Elusive genes
In the past gene hunters have claimed to find evidence for a ‘schizophrenia susceptibility gene’ on chromosomes 5, 11, 22 and the X chromosome. Some of these claims have fallen flat, others remain unsubstantiated, and few teams expect to find such a gene tomorrow. Nevertheless, scientists are confident that schizophrenia has a large inherited component, partly because in studies where both twins suffer from the disease, a higher proportion are identical than nonidentical.
In the 1990s, scientists are using molecular methods to speed up the hunt for schizophrenia genes. Some, such as Peter McGuffin and Mike Owen at the University of Wales College of Medicine in Cardiff, are using the techniques known as positional cloning to search through the DNA of individuals from at least 200 pairs of siblings affected by schizophrenia in the hope of locating genes linked with the disorder.
The Cardiff team and others in the US are also planning another approach. There is some evidence, still controversial, that schizophrenia develops at earlier and earlier ages in successive generations of families heavily affected by the disease. This pattern has parallels with certain other conditions such as fragile X syndrome and Huntington’s disease, which are known to be caused by defective genes made up of DNA ‘stutters’ – repeated sequences of triplets of base pairs of varying length. The longer the repeat, the more severe the syndrome and the earlier the disease starts. If schizophrenia has a similar genetic basis, some scientists believe that hunting for trinucleotide repeats in the genomes of affected family members may help them to close in on their quarry.
* * *
The dopamine link
The strongest evidence to support the involvement of dopamine in schizophrenia is that drugs which block certain dopamine receptors in the basal ganglia reduce the symptoms of the disease, while other drugs that stimulate the dopamine system, such as amphetamines, can produce schizophrenia-like symptoms even in healthy people. These observations led scientists to predict that the dopamine system was somehow ‘turned up’ in schizophrenia: either there was too much dopamine in the schizophrenic brain, or there were too many receptors.
It seems that neither explanation is right. The disturbance of the dopamine system must be secondary to some other molecular effect, scientists believe, such as imbalances in components of cell membranes.
Lyn Pilowsky and her colleagues at the Institute of Psychiatry in London and at the Institute of Nuclear Medicine have used single-photon emission tomography (SPET) to track a radiolabelled mimic of dopamine in the brains of schizophrenics and to watch the activity of drugs that are believed to block its receptors. There are five known types of dopamine receptor, dubbed D1 to D5, and probably many more waiting to be discovered. The team has found no overall differences from normal brains, but has detected subtle abnormalities in D2 receptors in the basal ganglia, including an apparent increase in the density of receptors. This higher binding capacity seems to be more pronounced in men than women and seems to be restricted to the left hemisphere.