Psychopathologists have studied three different kinds of non-optimal relationships between children and their families. The first involves some kind of disruption of the emotional bond that is usually formed between parent and infant at the beginning of the infant’s life. This kind of bond, known as an attachment relationship, has been the subject of intense investigation by psychoanalysts, psychologists and animal behaviourists. The second kind concerns the emotional climate that develops in families as the child grows older, and is reflected in the concepts of expressed emotion and affective style, which we encountered in Chapter16. Finally, as we have already seen, some research has focused on the concept of communication deviance, the idea that persistently vague, fragmented or contradictory communications from parents can lead to cognitive confusion in the child, increasing the probability that he will eventually become thought-disordered.
Attachment relationships
Following the British psychiatrist John Bowlby’s claim that secure attachment to a parental figure is necessary for healthy psychological development,8 opinions about the impact of early relationships have varied widely. In part, this reflects the difficulty of studying influences across the human life span. Sceptics have argued that emotional deprivation at an early age can be compensated for by adequate emotional care later in childhood. Certainly, some highly resilient children survive appalling mistreatment by their parents and emerge relatively unscathed as adults. However, studies in which attachment relationships have been deliberately disrupted in young animals suggest that negative consequences often ensue. Infant rhesus monkeys who are separated from their parents and raised with their peers, even if nurtured adequately in other ways, develop quite marked behavioural and emotional difficulties in adolescence, and these difficulties are accompanied by biochemical changes in the brain (decreased levels of 5-hydroxyindolacetic acid, the primary metabolite of the neurotransmitter serotonin).9 Moreover, recent studies of insecurely attached human children have revealed that they are handicapped by subtle social-cognitive deficits that may not be obvious to the casual observer. Although intellectually the equal of securely attached children, at 2 years of age they are less persistent and enthusiastic when solving problems.10 At the age of 11 years, they are less able to recall specific incidents from earlier in life, and are less able to reflect on their own mental processes.11 Perhaps more importantly for our purposes, studies conducted by Peter Fonagy and his colleagues at University College London12 and by Elizabeth Meins at Durham University13 have shown that children between 4 and 6 years of age who are insecurely attached perform less well than securely attached children on tests designed to measure their ability to understand the mental states of other people (‘theory-of-mind’ skills).
The suspicion that insecure attachment relationships confer vulnerability to psychosis is fuelled by the results of the Copenhagen high-risk study, which found an association between positive symptoms and early separation from parents. However, more powerful prospective evidence of this effect has emerged from a large cohort study carried out in Northern Finland, which was initiated in 1966. Before the birth of the 11,000 children who were entered into the study, their mothers were questioned about whether their pregnancies had been planned and whether they had considered an abortion. Examining outcome data twenty-eight years later, a team of British and Finnish investigators found that unwanted pregnancies resulted in a four-fold increase in later psychosis, even when possible confounding socio-demographic and medical factors were taken into account.14 Although the researchers did not offer a psychological interpretation of their findings, the obvious implication is that parents’ emotional ambivalence towards their children increases the risk that they will grow up to suffer from positive symptoms.
If this idea is correct, we should be able to detect evidence of insecure attachment styles in adult patients who have become psychotic. The definition and measurement of attachment styles is a complex field, and there is not enough space here to describe this kind of work in detail.15 Suffice it to say that several different styles are commonly observed in normal children and adults. Furthermore, there is evidence of inter-generational transmission of these styles. Children who have secure relationships with their parents typically but not always grow up to form secure relationships with adult partners, and later become parents of securely attached children. Insecurely attached children, on the other hand, often encounter problems when attempting to form relationships in later life, and sometimes fail to bond adequately with their own children.
In a series of studies carried out in the USA, Mary Dozier used questionnaires and interviews to assess psychiatric patients’ attachments to their parents. It is important to note that Dozier’s aim was to study attachment representations– the patients’ understanding of their relationships with their parents, rather than actual relationships. She accepted that her participants’ accounts might not be an accurate reflection of what actually happened in their families many years before. In an initial study, she found that patients with a diagnosis of schizophrenia, in comparison with patients with affective disorders, often had the kind of insecure attachment style that is described as dismissing or avoidant.16 Most people feel that their relationships are very important to them, however troublesome they may be. However, the dismissing-avoidant person devalues the importance of attachments and often fails to recall specific details about his relationship with his parents during childhood. It is as if, as a consequence of early emotional disappointment, the growing child develops a strategy that allows him to avoid the emotional hazards of mature relationships. In a later study, Dozier found that patients with this attachment style were especially likely to be deluded, hallucinated and suspicious.17 Perhaps this is unsurprising, as the emotional theme underlying the dismissing style is lack of trust.
Evidence supporting Dozier’s findings has emerged from community surveys of attachment styles undertaken by social psychologist Philip Shaver and his colleagues in the USA. In two studies, large representative samples of adults and adolescents were administered questionnaires measuring attachments to others. Several measures of psychopathology were also administered. In the adult sample of over 8000 people, schizophrenia as defined by DSM-III-R criteria was associated with an insecure and especially dismissing attachment style.18 In the adolescent sample of over 1500 individuals, high levels of psychoticism and paranoia were associated with both the dismissing style and a second type of insecure style, known as anxious-ambivalent.19 As the name suggests, people with this attachment style desperately want to have relationships, but feel in their hearts that no one will ever want to get close to them.
Of course, it might be argued that these styles are consequences rather than causes of paranoid thinking. If this were so, we might expect recovered paranoid patients to report warm and supportive relationships with their families. However, in a recent study conducted by Peter Rankin, myself and former colleagues at the University of Liverpool, we found that both currently ill and recovered paranoid patients report extremely difficult relationships with their parents.20
The idea that negative attitudes towards parents is a consequence of psychosis is also difficult to square with some of the prospective evidence we considered earlier – for example, from the Finnish cohort study and the Copenhagen high-risk study. In fact, early in the Copenhagen high-risk study, the high-risk children were interviewed about their relationships with their parents. Many years later, it was found that those who reported a negative relationship with their parents were especially likely to develop schizophrenia symptoms.21
Expressed emotion and communication deviance
We have already considered evidence that communication deviance in parents can contribute to the later development of thought disorder in their children. Perhaps the best indication that this is partly an environmental effect has emerged from Pekka Tienari’s adoption study, which found that both a genetic predisposition and exposure to vague and fragmented parental communicati
ons are necessary conditions for the development of this complaint.22
The possible causal role of expressed emotion in psychosis has hardly been investigated, perhaps because most researchers have decided to reject this possibility from the outset. In support of this prejudice, some have pointed out that studies of the relatives of patients experiencing their first episode of psychosis have usually found lower rates of high expressed emotion than studies carried out with the families of patients who have been ill for some time.23 This observation certainly supports the idea that expressed emotion grows during interactions between psychotic people and their parents, so that the emotional climate in families tends to deteriorate as patients become more disturbed and less able to look after themselves. However, it does not preclude the possibility that excessive criticism and parental over-control can play a role in causing symptoms in the first place. The only adequate way of addressing this possibility is by studying the effects of expressed emotion prospectively.
The University of California at Los Angeles (UCLA) Family Project was initiated by psychologist Michael Goldstein and his colleagues in 1965.24 The sixty-five families who agreed to participate had all sought help from the UCLA Psychology Department’s outpatient clinic because of difficulties they were experiencing with an adolescent child. At the time at which the project started, none of these adolescents had experienced psychotic symptoms. Two measures of family relationships were employed: a measure of communication deviance derived from the work of Singer and Wynne, and a measure of emotional climate (the Affective Style Index, based on the concept of expressed emotion, but derived from observations of the family members attempting to discuss a problem). Five and fifteen years later, the children were graded on a seven-point scale of psychopathology ranging from normal, through neuroses, more severe personality problems, to borderline, probable and definite schizophrenia. As in the other high-risk studies we have considered, the proportion of children developing frank psychotic symptoms was quite small (only four out of the fifty-two followed up met the criteria for probable or definite schizophrenia). Nonetheless, there was evidence that both communication deviance and affective style predicted future psychosis. These two types of interactions between parents and their children appeared to be independent influences, so that the combination of high communication deviance and a negative affective style appeared to be particularly dangerous to the children’s mental health. Moreover, there was evidence that the harmful effects of communication deviance were most marked in families with a high genetic risk of psychosis (that is, with other psychotic people in the family).25 This finding of an interaction between family processes and genetic risk is, of course, consistent with the results from Tienari’s adoption study.
How families affect their children
A complete account of the way that families affect the development of madness would explain how different family characteristics influence the psychological processes responsible for particular complaints. Fortunately, clues about these influences can be discerned from research carried out by developmental psychologists, whose efforts to map out the pathways to ordinary adulthood have too often been ignored by psychopathologists. For example, we have already seen that secure attachment relationships facilitate the development of ‘theory-of-mind’ skills, an observation of some importance given that these skills seem to be compromised in some psychotic patients.
Similar progress has been made in mapping the ways in which parents influence children’s beliefs about themselves, and the kinds of explanations that they make for important events. Martin Seligman found that the attributional styles of 11-year-old children correlate strongly with the styles of their mothers but not their fathers.26 Although this observation seems to indicate that children somehow ‘copy’ their mothers’ characteristic way of explaining things, other studies have shown that parents can influence their children’s attributions in less direct ways.27 For example, if a parent makes attributions for a child’s poor academic performance that are internal to herself (‘It’s my fault because I did not bring up my child properly’), the child is likely to make attributions for the failure that are external to himself (‘It’s not my fault that I keep failing my exams’). Conversely, if a parent criticizes a child’s performance, the criticism may be embraced by the child and later repeated as an internal attribution (‘I failed because I’m stupid’). These kinds of influences may be important even in adolescence and earlyadulthood. In a recent study, Judy Garber at Vanderbilt University in Nashville found that explanations for positive and negative events given by adolescents tend to match the explanations given by their mothers.28
Comparable findings have been obtained from studies of children and young adults judged to be at high risk of depression. Constance Hammen at the University of California found that the children of depressed mothers have more pessimistic styles of thinking, and more negative beliefs about themselves, than the children of non-depressed mothers.29 In the recent Temple–Wisconsin Cognitive Vulnerability to Depression (CVD) Project (see Chapter 10, pp. 245–6), interviews conducted with parents of students participating in the study revealed that the parents of the high-risk students were less accepting of their children and more likely to be ‘negatively controlling’ or emotionally abusive than the parents of the low-risk students.30 Moreover, the parents of the high-risk students were especially prone to explain hypothetical negative events happening to their children in terms of causes that were internal to the child (for example, deciding that a child was not invited to a party because she was unpopular).31
These findings strongly suggest that parents may inadvertently teach their children a depressive style of thinking. As attributions also play an important role in persecutory delusions, it is reasonable to ask whether paranoid thinking is similarly influenced by early relationships with caregivers. Studies have yet to be carried out to test this possibility…
To summarize, there seems to be quite good evidence that family relationships affect the risk that the developing child will eventually become psychotic. Families have this effect because they influence the child’s beliefs about himself, his thoughts about the causes of things that happen to him, and his ability to understand the behaviour of other people. At the risk of being tedious, it is worth repeating that this conclusion does not imply that every psychotic patient is a victim of an inadequate family, that parents choose to confuse or criticize their vulnerable children, or that the child plays no role in influencing these processes. It is also important to remember, as attachment theorists have been at pains to note, that parents are products of their relationships with their own parents. As we saw in Chapter 16, parents who are unable to provide an optimal emotional environment for their children have often themselves been raised by parents who were unable to provide them with adequate emotional support.32
This intergenerational spiral of affects was brilliantly captured by the poet Philip Larkin. The less remembered second verse of his most famous and irreverent poem, which I quoted earlier in this chapter, encourages us not judge our parents too harshly.
A World not of our own Making
The research literature contains several hints that long-term exposure to a stressful social environment can contribute to the development of madness. One hint comes from the over-representation of people with paranoid or manic symptoms among immigrant populations. As I explained in Chapter 6, this has been most extensively investigated in black Afro-Caribbean immigrants to Britain.33 Although the reasons for this over-representation have been the subject of much debate, careful epidemiological studies indicate that the increased risk is real, and cannot be explained away as just the product of culturally insensitive diagnostic practices. Moreover, obvious biological factors (for example, genetic influences) appear to be ruled out, or at least diminished in likelihood, by the finding that incidence rates of psychosis in the Caribbean are no higher than in other countries,34 by the observation that the children of Afro-Caribbean immigrants
are especially vulnerable,35 and by the recent discovery that immigrant groups in other parts of the world are similarly affected – for example, Surinamese immigrants to Holland,36 East African immigrants to Sweden,37 migrants to Germany,38 and even Afro-Caribbeans returning to Jamaica after a period of living in Britain.39 Further evidence that the critical factor is something to do with different ethnic groups mixing closely has emerged from a recent study carried out in London, in which the researchers analysed the incidence of psychotic illnesses in different neighbourhoods. It was found that non-white people living in white neighbourhoods are more likely to become psychotic than non-white people living in predominantly non-white neighbour-hoods.40 Exposure to racial tension, it seems, can drive people mad.
It is easy to see how exposure to overt discrimination and institutional racism, perhaps coupled with experiences of alienation and isolation, might affect psychological processes (especially attributions and self-representations) that appear to play an important role in paranoia and mania. It is perhaps less obvious why the children of immigrants should be particularly affected in this way. One possibility is that the practical and economic problems associated with migration make it difficult for families to provide their children with optimum child-rearing experiences that would protect them from these influences. Evidence that this might be the case was found in a study of Afro-Caribbean schizophrenia patients living in London carried out by psychiatrist Dinesh Bhugra and his colleagues. Of 38 Afro-Caribbean patients studied, 12 (34 per cent) had suffered separation from their mothers for a period of four years or longer during childhood, and 19 (53 per cent) had suffered a similar period of separation from their fathers.41
Madness Explained Page 54
