Haydn Ellis, a neuropsychologist at the University of Wales in Cardiff, has recently pointed to several reasons for doubting the psychodynamic account of the Capgras delusion.62 The theory implies that the delusion should exclusively concern individuals who have a close relationship with the patient, whereas this is not invariably the case. Also, Capgras patients usually express no fears about the genuineness of voices heard over the telephone, suggesting that some difficulty in processing visual information is a crucial element of the disorder. Most importantly of all, about one third of Capgras patients are known to have suffered from some kind of brain damage that predates the appearance of their delusion.
With Andy Young, a psychologist at the University of York, Ellis has argued that the Capgras syndrome is caused by damage to brain systems responsible for recognizing faces.63 When we encounter someone who is familiar to us, we normally experience a brief emotional response. This reaction, which is experienced as the feeling of familiarity, has considerable survival value, as it tells us whether we are meeting a friend or a foe. Not surprisingly, in most people it occurs automatically and very efficiently. (We may recognize a person as familiar after meeting him or her just once, many years ago.) In comparison, the skill of recalling specific information about a person, for example their name and occupation, is much less reliable. For this reason, most of us have had the embarrassing experience of meeting people who seem familiar to us without our being able to remember precisely who they are.
These two types of recognition seem to involve different pathways in the brain.64 A pathway through the visual cortex to the limbic system is involved in the specific identification of individuals. A second pathway, which passes through the inferior parietal lobe, appears to be responsible for the feeling of familiarity. Capgras patients show subtle deficits when they are tested for their ability to recognize faces, indicating that they do not always gain a sense of familiarity on encountering people they know well, so that the second pathway is disrupted. The delusion seems to be an attempt to explain this experience.
Ellis and Young predicted that Capgras patients would show no emotional response when meeting people well known to them. Normally, when we greet someone familiar, our emotional reaction (however slight) results in a slight increase in sweat, leading to a transient change in the electrical conductivity of our skin (the electrodermal response) which can easily be detected from electrodes placed on the back of the hand. Ellis, Young and their colleagues tested five Capgras patients and found this response to be absent, just as they had predicted.65
It is not clear whether an anomalous face recognition experience on its own is enough to create a Capgras delusion. Some brain-damaged patients who suffer from impairment in their ability to recognize faces retain insight into their difficulties, know that they have a perceptual problem, and do not develop bizarre explanations of their experiences. On the other hand, as Andy Young points out, patients like Mme M very often entertain grandiose or paranoid ideas, indicating that some kind psychological disturbance may also be required for a full-blown Capgras delusion to develop.66
The evidence we have considered so far concerns the impact of perceptual deficits on delusional thinking. An interesting but nearly overlooked possibility is that delusions might arise as a consequence of perceptual skills that are over-developed. This possibility was tested in a strange experiment reported in 1978 by Canadian psychologist Lucrezia LaRusso. LaRusso persuaded a group of normal volunteers to allow her to film their facial expressions as she gave them brief electric shocks. She also filmed them as they pretended to receive the shocks. Showing the films to paranoid patients and to a second group of normal volunteers, she found that the paranoid patients were better able to tell the differences between the genuine emotional responses and the pretend responses.67
Similar findings have recently been reported by Penelope Davis and Melissa Gibson of Griffith University in Australia, who studied schizophrenia patients’ ability to recognize genuine and posed facial expressions. While most of their patients were poor at recognizing posed expressions, their paranoid patients were better at recognizing genuine expressions of surprise and negative emotion than ordinary people.68 These observations raise the intriguing possibility that over-sensitivity to other people’s disingenuous communications might provoke paranoid thinking.
It is difficult to determine the extent to which the anomalous perception theory applies to delusions in general. One complication is that what counts as an anomalous experience may vary from person to person. Three American psychologists, William Johnson, James Ross and Marie Mastria of the University of Mississippi Medical Center, reported an extraordinary case study that illustrates this problem. Their patient complained that he was being sexually molested at night by ‘warm forms’. A detailed history revealed that, from the age of 12, he had had numerous sexual partners and had never learned to masturbate. The warm forms appeared during a long period of sexual abstinence, occasioned by his contracting of a venereal disease. During this period, he experienced spontaneous episodes of sexual arousal at night-time, which seemed anomalous to him, but which would have seemed quite normal to anyone with a less colourful history. Because, previously, sexual arousal had always occurred in the presence of a partner, the patient had concluded that an invisible sexual partner (a warm form) was responsible for these experiences. The treatment, which was brief and successful, is left to the imagination of the reader.69
There have been few systematic attempts to address the extent to which delusions are products of anomalous experiences. In one study, Loren and Jean Chapman of the University of Wisconsin carefully interviewed students who scored very high on their questionnaire measures of schizotypy. They found that some expressed delusional ideas but reported no anomalous experiences, and that others reported anomalous experiences in the absence of delusional ideas. In only a few cases did there appear to be an obvious causal connection between anomalous experiences and unusual beliefs and, even in these, the steps leading from the anomalous experience to the delusion were often judged not to be ‘reasonable’.70 In a more recent three-year prospective study of eighty children who heard voices, Sondra Escher, Marius Romme and their colleagues found that only 9 per cent developed delusions during the follow-up period.71
On this evidence, at least, Maher’s theory probably holds for only a minority of patients.
The role of attention
Despite the limited evidence for Maher’s theory, it is possible that perceptual and especially attentional processes play a role in maintaining delusional ideas, once they have been formed. Our brains constantly scan the environment for information that is relevant to us and ignore information that is not. In the late 1960s, two American psychologists, Leonard Ullmann and Leonard Krasner, suggested that this process would lead deluded patients to notice selectively evidence that supported their beliefs.72
Sue Kaney and I used the emotional Stroop technique to investigate this hypothesis. (Recall that, in this type of test, described in detail in the last chapter (pp. 281–2), participants are required to look at words written in different ink colours and name the colours while ignoring the words. The difficulty they have in doing this indicates the extent to which they are selectively attending to the words.) We used four types of stimuli: paranoia-related words (for example, ‘deceit’, ‘follow’), depression-related words (for example, ‘defeat’, ‘failure’), neutral words (for example, ‘bud’, ‘recipe’) and meaningless strings of letters. The words and letter-strings were in five different ink colours and were printed on cards, five words to a line, in random order. As we had expected, paranoid patients took much longer to colour-name the paranoia-related words than the neutral words, indicating that the meaning of the words was grabbing their attention and interfering with their ability to name the ink colours. This effect was not found for a group of ordinary people or for a control group of depressed patients.73 Perhaps this finding should be regarded as unremarkable. As we have
already seen, people will show a Stroop effect for words relating to any topic that causes them concern.74 Deluded patients are no exception to this rule.
In an interesting variant of this experiment, Kate Leafhead, Andy Young and Krystyna Szulecka studied a woman who was suffering from the Cotard delusion (she believed that she was dead). Their patient also suffered from a Capgras delusion that her brother and mother were impostors. She was tested on three occasions using words relating to death (for example, ‘coffin’, ‘dead’), depression-related words, words relating to duplicates (for example, ‘copy’, ‘double’), paranoia-related words, anxiety-related words and emotionally neutral words (see Figure 12.3). When very ill, on the first occasion that she was tested, the woman was slow to colour-name both the death-related and the duplicate-related words. On a second occasion, when she was beginning to improve, only the duplicate-related words caused her difficulty. By the third test, when she was nearly recovered, colour-naming was not impaired for any of the words.75
If people attend excessively to information relating to their delusions they should also recall this kind of information particularly easily. In several studies, Sue Kaney and I found that this was indeed the case. For example, after listening to a series of stories describing threatening and non-threatening social interactions, paranoid patients specifically recalled the threatening elements from the stories, but tended not to remember emotionally neutral elements.76
Interestingly, deluded patients do not seem to spend excessive time looking at stimuli relating to their delusions. Recently, Mary Phillips and Tony David at the Institute of Psychiatry in London studied the eye movements of patients as they looked at various pictures, some of which showed people making threatening gestures. Using sophisticated equipment to track the direction of gaze, they measured the amount of time spent visually scanning different parts of the pictures. Patients with paranoid delusions, compared with patients with other symptoms, spent less time looking at the threatening parts. Quick to identify stimuli in their environment that could be threatening, it seems they move rapidly to checking whether threats can be found elsewhere.77
Figure 12.3 Extent to which a patient suffering both a Cotard delusion and a Capgras delusion had difficulty colour-naming death-related words, depression-related words, duplicate-related words, paranoia-related words and anxiety-related words when tested while ill (Test 1) and during recovery (Tests 2 and 3). Scores indicate the extent to which colour-naming for these words is slowed in comparison with colour-naming neutral words. From Leafhead, Young and Szulecka (1996).
What Happens in the Inference Box?
The third component of the back-of-an-envelope model concerns the process of drawing inferences from observations. Two lines of research have focused on this process in deluded patients. One was initiated by British psychologist Chris Frith,78 and the other (which we will come to shortly) was started by myself.
Inferences about other minds
Frith has argued that a deficit in theory of mind (ToM; the ability to infer mental states in others, see Chapter 8), which is less severe than that found in autistic patients, and which is only present during episodes of psychotic illness, could lead to paranoid symptoms. According to this theory, paranoid patients suddenly discover that they have lost the ability to understand the thoughts and feelings of other people, and then explain this experience by assuming that other people are trying to hide their intentions. Together with fellow psychologist Rhiannon Corcoran, Frith has used a number of strategies to test this theory.
One approach has involved testing the ability of patients to comprehend the beliefs and actions of people portrayed in simple stories.79 For example, consider the following tale:
Burglar Bill has just robbed a bank and is running away from the police when he meets his brother Bob. Bill says to Bob, ‘Don’t let the police find me. Don’t let them find me!’ Then he runs off and hides in the churchyard. The police have looked everywhere for Bill except the churchyard and the park. When they come across Bob they were going to ask him: ‘Where is Bill? Is he in the park or the churchyard?’ But the police recognize Bob and they realize that he will try to save his brother. They expect him to lie and wherever he tells them, they will go and look in the other place. But Bob, who is very clever, and does want to save his brother, knows that the police don’t trust him.*
The person being tested is asked ‘Where is Bill really hiding?’ to ensure that he can remember the story. If he answers correctly, he is then asked: ‘Where will Bob tell the police to look for Bill – in the churchyard or in the park? Why?’ Answering this second question requires the ability to make complex inferences about Bob’s beliefs and intentions.
Frith and Corcoran found that two groups of schizophrenia patients experience problems with questions of this sort: paranoid patients and patients with mainly negative symptoms. In the latter group, ToM performance correlated strongly with IQ, suggesting that their poor performance was probably caused by general cognitive deficits. In the paranoid patients, however, poor performance on the ToM questions seemed to be unrelated to IQ. Interestingly, as Frith had predicted, recovered paranoid patients performed normally in these studies, indicating that ToM deficits are present only in episodes of illness.
Further experiments conducted by Frith and Corcoran involved other ingenious methods of assessing ToM skills. For example, in one study they investigated the ability of patients to recognize simple hints (as when a husband in a hurry to leave for work glances at the ironing board and laments to his wife that none of his shirts is ironed). In this study, patients with persecutory delusions were poor at inferring intentions from hints, but schizophrenia patients whose symptoms were in remission performed the task as well as the ordinary people.80
Frith and Corcoran’s ideas sparked off a series of studies by other researchers. Unfortunately, although these generally found that ToM skills are impaired in currently ill psychotic patients, the evidence that this kind of deficit is specifically associated with paranoid symptoms proved to be unclear. In studies carried out in France, Yves Sarfati and his colleagues have found strong associations only between ToM dysfunction and thought disorder,81 while Robyn Langdon and Max Coltheart, working in Australia, have found evidence of ToM problems in groups of patients with negative symptoms but not in people with persecutory delusions.82 Using a series of ToM tests, Val Drury and others in Birmingham found evidence of deficits in currently ill schizophrenia patients suffering from multiple positive and negative symptoms, but when they compared patients with persecutory delusions and those without they found no difference.83 Some studies have even found that some paranoid patients perform better on ToM tasks than other groups of schizophrenia patients.84 Overall, then, the evidence pretty decisively shows that ToM deficits are not specific to paranoia. Indeed, as one of my own postgraduate students, Natalie Kerr, found that both bipolar-depressed patients and bipolar-manic patients performed poorly on ToM tasks,85 it seems likely that this kind of disability is present in acutely ill psychotic patients whatever their diagnosis.
Despite this negative evidence, I don’t think we should give up on Chris Frith’s theory. Indeed, there is obviously something theory-of- mind-ish about persecutory delusions, which inevitably involve mistaken assumptions about the intentions of other people. It is as if the paranoid person can make inferences about the beliefs and attitudes of other people but, for some reason, reaches the wrong conclusions about what those beliefs and attitudes are. I will return to this idea later.
Inferences about causes
When I obtained my priming grant to study delusions in the late 1980s, an immediate difficulty was my embarrassing lack of hypotheses worth testing. After a few weeks thinking through this problem, I hit on the strategy of measuring the same processes that had been studied in depression, particularly attributions. After all, I speculated vaguely, Abramson and Seligman’s attributional model of depression seemed to imply that dysphoric mood springs from some kind o
f delusional interpretation of events. Moreover, it seemed obvious just from listening to paranoid patients that many of their beliefs were, in fact, unusual causal statements. This intuition was reinforced by an almost trivial incident that occurred at about this time. One of my patients, Gerald, arrived half an hour late for an appointment and, as another patient was expected, I had to tell him that I would be unable to see him. Embarrassed, Gerald explained that a malicious neighbour had used supernatural powers to enter his mind and mess with his memory of the appointment. This explanation appeared to have all the characteristics of an external attribution, as the cause of Gerald’s lateness was attributed to an external agent over which he had no control.
Readers will recall that the attributional model of depression was first tested by giving patients the Attributional Style Questionnaire (ASQ), which asks the person completing it to think of likely explanations for hypothetical positive and negative events (see Table 10.1, p. 242). In one of our first studies of paranoid thinking, Sue Kaney and I used this questionnaire to compare paranoid patients, depressed patients and ordinary people.86 We found that paranoid patients, like depressed patients, tended to make excessively global and stable explanations for negative events. (In other words, when things went wrong they tended to assume that the cause would affect all aspects of
Madness Explained Page 37
