The main cognitive models of depression were presented earlier in Chapter 4 because Seligman's learned helplessness and Beck's cognitive therapy models have also been applied to other psychological disorders. Rather than repeat these presentations, therefore, we will simply highlight some of the conclusions that we reached.
First, there is no convincing evidence that attributional style is a vulnerability factor for the onset of depression, whatever its role in maintenance and recovery might be (Bieling & Segal, 2004). For example, although researchers have been at pains to demonstrate that depression can result from the impact of independent or uncontrollable life events (e.g., Brown & Harris, 1978), as Champion (1990; Champion, Goodall, & Rutter, 1995) has clearly shown, there is normally a more complex interaction such that vulnerable individuals may experience more of both uncontrollable (independent) and controllable (dependent) events. In sum, uncontrollability of events is not a necessary feature for the onset of depression.
Second, the cognitive models of Beck (1976) and Bower (1981) sent researchers off in pursuit of global cognitive biases in a range of cognitive domains. As in Lewis Carroll's classic tale of such pursuits, the Snark turned out to be something of a Boojum. The failure to find global biases led some critics to reject the cognitive approach (Coyne & Gotlib, 1983), it led others to consider a more refined approach to the question of cognitive bias (e.g., Williams, et al., 1997), and at the other extreme, yet others claimed that bias was a feature of being normal whereas "realism" was a feature of depression (Alloy & Abramson, 1979). We pitch our tent somewhere in the middle ground on this issue, which has now become the more widely held view (e.g., Abramson et al., 2002; Clark & Beck, 1999) and argue that bias is as much of a feature of normal thinking as it is of depressed thinking, although the biases may operate in different ways. The question of bias, however, remains an important one, so we will return to it later in the chapter.
Third, in Chapter 4 we also presented some of the key points from the integrative social-cognitive theories of depression (Champion & Power, 1995; Gotlib & Hammen, 1992; Oatley & Bolton, 1985). At the forefront of these theories have been some of the puzzling epidemiological findings mentioned in the previous section. These and other puzzles have contributed to the belief that an adequate model of depression will require the integration of a range of social, psychological, and biological factors, and that no theory pitched at a single level is likely to be successful (cf. Weissman, Markowitz, & Klerman, 2000). We will therefore examine in more detail the account of depression presented by Champion and Power (1995), which attempts to integrate a range of social and psychological factors.
The key points from the Champion and Power (1995) model are presented in Figure 7.5. Along with a number of social and social-cognitive accounts of depression, one of the key components of the model relates to the roles and goals that are available to the individual. Unlike purely social accounts however (e.g., Becker, 1964; Thoits, 1986), the theory focuses on the perceived value of roles and goals and not simply their availability or mere existence. It is proposed that vulnerable individuals typically overinvest in one role or goal, but underinvest and do not value other areas of their lives. When things go wrong therefore—for example, when a relationship with "the only person in the world who understands" breaks up, or when the adored child proves fallible, or when the longed-for promotion does not occur—the matching negative event leads to a breakdown in the defensive processes that normally protect the individual during successful pursuit of the role or goal (see also the discussion in Chapter 5 of roles and goals in different domains). Lam and Power (1991) reported supporting evidence for this proposal using the Roles and Goals Questionnaire, which was in part designed to test the model. They found as predicted that the proposed pattern of over- and underinvestment was more typical of depressed than non-depressed adult and elderly community samples. However, the studies were cross-sectional so it was unclear whether the pattern was a cause or a consequence of depression. Lam et al. (1994, 1996) have subsequently reported evidence in a longi-
(A) VULNERABLE INDIVIDUALS
(B) NON-VULNERABLE INDIVIDUALS
OVERVALUED ROLE OR GOAL
MATCHING NEGATIVE EVENT
BREAKDOWN OF MOOD/ ESTEEM REPAIR PROCESSES
MOOD/ESTEEM REPAIR PROCESSES
DOMINANCE BY NEGATIVE PART OF AMBIVALENT SELF
TRANSIENT NEGATIVE EMOTIONS
REPLACEMENT OF LOST ROLE OR GOAL
Figure 7.5 A summary of the model of depression presented by Champion and Power (1995).
tudinal study of outpatient depressive patients that the further occurrence of negative events in the most invested domain delays recovery from depression or increases the likelihood of relapse in someone who has already recovered. Additional evidence in support of this part of the model is presented in Champion and Power (1995).
A second key component of the model presented in Figure 7.5 is highlighted in the contrast between vulnerable and non-vulnerable individuals and concerns the self-protective function that the overvalued role or goal provides. The proposal is that while the overvalued role or goal is being successfully pursued, the vulnerable individual may stave off self-negativity. However, if the role or goal is threatened or subsequently lost, the consequent combination of sadness and self-negativity (primarily, that is, the predominance of self-disgust) will lead to depression. Although, as highlighted in Figure 7.5, non-vulnerable individuals also experience a range of emotions at the loss of important roles and goals (see the earlier section on grief), such individuals are less likely to experience self-disgust or self-negativity and are more likely to replace the lost role or goal from a range of other valued roles and goals.
John was a 35-year-old man who had become depressed for the first time when his current girlfriend, whom he had known for a year, suddenly and unexpectedly walked out on him one morning. Never before had he experienced such loathing of himself and a constant preoccupation with wanting to kill himself.
He remembered little of his childhood before the age of 8, only that he had a constant longing to be an adult so as to escape a feeling of desolation. His first clear memories were of being sent away to boarding school, because, as he thought at the time, his parents wanted him out of the way. Unlike other boys who got upset and ran away from school, he knew there was no point in giving in to such "weak emotions". However, he did little work at school and left at 16 without qualifications to go and work in a factory. Even now he took pleasure in the fact that he had been such a disappointment to his parents and turned out "the opposite of what they wanted".
His mother died after a long illness a few years later, but his father had never allowed anyone to talk of her illness, nor to express any emotion after her death. Shortly afterwards John married. It was an "ideal match", they were "completely suited to each other" and during the 8 or so years of the marriage John gradually "got his life together" and gained some qualifications and eventually they set up their own small business together. Unfortunately, however, for John, this idyllic period in his life was suddenly brought to an end when his wife was tragically killed in a car accident. The news, he said, sent him into a not-unpleasant "numbed state" which lasted for 3 months. Again, he had little or no memory for what happened at that time, and when he looked back at his diary for that time it suddenly went from being crammed full to being completely empty.
John is now recovering from his first episode of depression, he says, "because his girlfriend unexpectedly returned one day". His main worry now though is how to avoid "ever feeling so bleak and suicidal again" and how to forget all that has happened to him.
John's case draws together much of what we have said in this chapter about both grief and depression. The familial pressure not to express emotion had been internalised to the extent that he had been unable to grieve the earlier deaths of his mother and of his wife. Instead, these losses had left John in a state of numbness in which not only did he not experience emotion, he was unable to achieve much else either. In relation to the SPAARS model, one possibility in John's case was that the direct emotion route was being continually reactivated in the types of emotion cycles that we considered earlier in the chapter (see also Chapter 5), but that the emotion generated via this route was under such strong inhibition that the inhibition affected most other things in his life as well and produced a state of numbness. It was only the subsequent event of his new girlfriend walking out on him that led to him finally experiencing both depression and grief, a sequence that has been observed in other inhibited grief reactions (e.g., Parkes, 1993).
There are two further points that we should make about the loss of a valued role or goal. First, the loss may of course occur late in life: the individual may have pursued a highly successful career or have been involved in a long and satisfying relationship, before tragedy in one form or another strikes. Indeed, the very fortunate though vulnerable individual may skate on thin ice for a lifetime and be lucky enough to escape depression. Nevertheless, there is a substantial amount of first-onset depression in later life (e.g., Laidlaw, 2004; Murphy, 1982), which can come as a particular shock to an individual who may have taken pride in a belief in his or her emotional strength.
George was the son of a docker and, indeed, had himself worked as a docker all his life. He had never had time for "soppiness", that was women's stuff, as his father used to say. He couldn't ever remember having cried throughout his adult years, that is until the last few months when, at the age of 64, he had become "deeply depressed" for the first time in his life and now couldn't stop crying. He felt totally ashamed of himself at this weakness and, even though his father had been dead for over 30 years, he could hear him laughing at him and making fun of him for being "such a sissie".
George had worked hard and been a devoted, if somewhat reserved husband. His wife had been "his first real love" and they had met and married when they were both 18. Their dream had been to retire early and buy a small cottage by the sea, where he and his wife would blissfully pass their twilight years. Indeed, he had just taken early retirement and they were about to move into a cottage they had just bought when his wife had a stroke and was rushed into hospital. She died 2 days later.
George had put the cottage up for sale and hidden himself away from everyone "for fear that someone might see him in the dreadful state that he was in".
The second point that we would like to make is that the apparent "loss" of a role or goal can also occur because of its successful completion (e.g., Champion, 2000) in addition to the more commonplace meaning of loss that we have used until now. For example, the famous painter and sculptor Michelangelo Buonarroti (1474-1564) became depressed and often lost interest in his work as it was nearing completion, because of which many of his most famous sculptures remained incomplete. The philosopher John Stuart Mill reported in his autobiography (1853) on an episode of depression that occurred when he was already in a low mood, but he then imagined the following:
In the frame of mind it occurred to me to put the question directly to myself, "Suppose that all your objects in life were realized; that all the changes in institutions and opinions which you are looking forward to, could be completely effected at this very instant: would this be a great joy and happiness to you?" And an irrepressible self-consciousness distinctly answered, "No"! At this my heart sank within me: the whole foundation on which my life was constructed fell down. (1853/1991, p. 75)
In Mill's case merely imagining the successful completion of his goals and plans was sufficient to lead to a severe episode of depression. Although the evidence for so-called "success depression" is primarily anecdotal and clinical, and its existence has been disputed on theoretical grounds by behavioural theorists (Eastman, 1976), one area deserving of further research is that of the "Golden Boy" and "Golden Girl" syndrome identified by Seligman (1975) in his classic book on learned helplessness (see Chapter 4). Seligman's proposal was that so-called Golden Boys and Golden Girls had experienced non-contingent positive reinforcement throughout their lives, because their parents had indulged them completely, but without ever requiring anything of them. When they then entered the "real world", however, they were unable to cope with the fact that rewards were no longer non-contingent, but instead were very much contingent on their own actions, and as a consequence they became depressed. However, an alternative interpretation is suggested by Lorna Champion's proposals about the importance of lifestage transitions in relation to roles and goals (Champion et al., 1995; Champion & Power, 1995; Maughan & Champion, 1990); namely that the transition from adored-child-cum-adolescent may require the individual to relinquish the role in which they starred and, hence, to abandon the role that gave them most satisfaction. Many such individuals may therefore attempt to delay the transition rather than face up to it, rather like Hilde Bruch's "golden cage" (1978) interpretation of anorexia as an attempt to prevent or delay the onset of maturity and adulthood. The development of interpersonal psychotherapy adapted for adolescence takes this transition as a key focus (e.g., Markowitz, 2004).
To return, though, to the possibility of "success depression" we recount the story of Martha. We hope that her story provides at least some clinical support for the proposal that the pursuit of an overvalued role or goal may provide a crucial self-protective or defensive function for the individual, a proposal, therefore, that runs counter to any straightforward cognitive account of emotion which equates happiness with the achievement of goals and sadness with their loss (e.g., Oatley & Johnson-Laird, 1987); occasionally, the reverse may be true if it is the pursuit of the goal rather than the goal itself that is more important to the individual.
Martha was an extremely successful and competitive individual. She had until shortly before entering therapy been the Deputy Director of a well-known charity. Her rise through the ranks had been fast and predictable and had followed an equally successful time as a student in which she had done a degree in Russian "simply as a challenge".
She lived alone, though occasionally picked up men at parties to satisfy her sexual needs. If ever any such man attempted to get close, she ended the relationship. She very much resented these times of weakness when she needed other people and wished she could be completely self-sufficient.
The reason for her depression was that she had been promoted to Acting Director and, subsequently, to her extreme surprise, actually been appointed Director of the charity. However, the circumstances in which these promotions took place proved to be excessively disturbing to her mental calm. She had come to suspect the previous Director of the charity to be purloining money that was donated to the charity. He realised that she was on his trail and had tried to get her sacked by the Trustees of the charity on the grounds of incompetence. She had confided her concerns to one of the Trustees and, eventually, was completely vindicated.
Martha had no comprehension of why she felt so depressed having reached the height of her ambition.
Martha's story also illustrates a third key part of the Champion and Power (1995; Champion, 2000) model: the role of self-ambivalence in depression. Some of the characteristics of Martha's self-ambivalence are presented in Figure 7.6. The figure illustrates that as long as Martha remained in pursuit of a work- or achievement-based goal this maintained the positive sense of self and the avoidance of negativity
Success in school In work
Avoidance of negativity re self and personal relationships
Failure or loss of work goal
Awareness of negativity re self and personal relationships
Figure 7.6 An example of the ambivalent self based on Martha.
in relation to the self and significant others. In addition to the ambition to run the charity, Martha thought for example that one day she would write a great novel that would make her famous, although she had not yet embarked on this task. However, she had achieved her current main goal, which was to head the charity, and her success had left her feeling "completely empty". Perhaps, too, the fact that this success had been marred by personal antipathy in which she had competed with and subsequently triumphed over a man in authority had also helped to prime some of the more personal relationship issues that were held at bay during the blinkered pursuit of her dominant goal. Whatever the truth, by the time she entered therapy those aspects of her self-concept that she experienced as negative were now completely dominant; in contrast to how she had been, she now felt unable to put any energy or effort into her work, and was worried that the considerable amount of time that she was off sick would lead to her dismissal.
Moreover, three basic emotions were now predominant: the emotions of sadness and disgust as shown in Figure 7.6, but also considerable amounts of anger that she did not normally experience (the opposite in fact of retroflective anger). She felt sadness because her life was empty and meaningless. She lived alone and had no close relationships with either sex; weekends were particularly painful because she now did not have the pretence of too much work to have time to see people. She felt self-disgust because of her physical appearance, because of her weakness in needing other people, particularly men, and because of her current "wallowing in emotion". She felt angry with everybody, but especially with her mother who never allowed Martha to have any needs, because her mother's needs were always greater than everybody else's. It was clear from Martha's history that she was only acceptable to her mother when she was good and did not need anything; her negative self was unacceptable and remained unintegrated with her positive self. The negative self therefore contained everything that was disgusting and unacceptable, everything about herself that Martha now wanted to get rid of, and that her mother had originally instructed her to get rid of.
The attempt to "eliminate" emotions and needs that are labelled as "weak" and as "negative" is a common feature of depression and other emotional disorders, as revealed, for example, by the endorsement of such attitudes on the "Self-Control" subscale of the 24-item Dysfunctional Attitude Scale (Power et al., 1994). Therefore the expression of disgust by significant others towards aspects of the child's self, needs, and expression of emotion seems a crucial area for research in relation to disorder. Indeed, in extremely vulnerable adults such expression by significant others has been shown to lead to relapse in both depression and schizophrenia; thus, in the work on expressed emotion (EE) Vaughn and Leff (1976) originally reported that the occurrence of "critical comments" by significant others greatly increased the chance of relapse in both depression and schizophrenia. Although most of the subsequent work has focused on schizophrenia (e.g., Lam, 1991), the important point is that many of the critical comments are expressions of disgust by a significant other directed at the vulnerable person.
On a more general level it should also be noted that a positive self that is unfettered by negativity is as pathological as a negative self unfettered by positivity. Such extreme positive selves may be seen for example in some manic conditions in which the self can become grandiose, powerful, and invulnerable (see Chapter 10). Although there is a psychoanalytic clinical tradition (e.g., Lewin, 1951) and some empirical data (Bentall & Kinderman, 1999; Winters & Neale, 1985) that such states are a "manic defence" against depression, the argument here is that, sure, the positive state inhibits the negative state, but the obverse is also true in that the negative state inhibits the positive state. Both the positive and the negative states have "validity" therefore, but they inhibit the expression of the other state because they are unintegrated and work antagonistically, rather than the positive state being simply a defence against the "true" state of depression.
In contrast to grief, therefore, in which we argued for the key roles of the basic emotions of sadness and anger, we argue that depression may be derivable primarily from the basic emotions of sadness and disgust. This proposal offers an alternative to that suggested for example by Freud in Mourning and Melancholia (1917), in which mourning was derived from sadness, whereas depression was derived from sadness plus anger that was turned against the self. The key rejection of Freud's proposal for retroflective anger occurred in Bibring's (1953) classic ego-psychoanalytic reanalysis of depression, a paper that anticipated all of the major cognitive approaches to the disorder (see Chapter 4). Nevertheless, our proposal does hold some similarities to Freud's in that we emphasise self-condemnation and guilt as defining characteristics of depression, the crucial difference being that we derive self-condemnation and guilt from the basic emotion of disgust rather than from anger. We would also point out that since Bibring's paper, most cognitive models have focused on the role of low self-esteem in depression; thus, in both Beck's (1976) cognitive therapy and in Abramson et al.'s (1978) learned helplessness reformulation, the self is seen as culpable for negative events, and is considered to be worthless, failed, or bad. All of these aspects of the self can be derived from the turning of the basic emotion of disgust against the self, such that aspects of the self are seen as bad and have to be eliminated or rejected from the self.
Another basic emotion that is frequently observed along with depression is that of anxiety; thus, self-report measures of depression and anxiety typically correlate at about 0.7 across a range of populations (Clark & Watson, 1991; Goldberg & Goodyer, 2005). Indeed, the so-called "tripartite model" proposed by Clark and Watson (1991; Clark, 2000) argues that there is a common core of "negative affect" that forms the major component of a range of emotions including depression and anxiety. While there is much to be commended in such analyses, we take issue with the basic underlying model: first, because so-called "negative affects" are not necessarily experienced as negative, as we have argued earlier; second, because most of the results are based on student populations or, even in their tests of the model, patient groups such as those with drug problems that do not directly test the model (Watson et al., 1995a, 1995b); and, third, because individuals may show less anxiety rather than more as they become increasingly depressed (Peterson, Maier, & Seligman, 1993). We would therefore concur for once with the DSM-IV (APA, 1994) with the decision that more evidence is necessary before the putative category of mixed anxiety-depression can be introduced.
These points do not in any way deny the high comorbidity of depression and anxiety, particularly for less severe depression, and indeed there is every likelihood that the coupling of the basic emotions involved in depression together with anxiety will undoubtedly lead to a prolongation of this distressing state. The proposal is, however, that anxiety is not a defining feature of depression, nor depression of anxiety. What we do wish to emphasise is that severe life events often unfold over time rather than occurring suddenly and out of the blue—they often occur in the context of long-term related difficulties (e.g., Brown, Harris, & Hepworth, 1995) and, as we shall subsequently provide evidence, in depression-prone individuals they can occur in an overinvested domain about which there is already considerable worry. In addition, the threat of loss may subsequently turn into an actual loss (Finlay-Jones & Brown, 1981), and so a state of anxiety in which the individual remains hopeful may turn into a state of depression in which the individual feels hopeless (Alloy, Abramson, Safford, & Gibb, 2006; Peterson et al., 1993).
The key coupling of the basic emotions of sadness and disgust that we argue is the basis of some presentations of depression is shown in Figure 7.7. So why have we given disgust such a central role? For this choice we offer a number of reasons, some of which are more speculative than others, although we accept that the overall proposal requires direct empirical testing. Nevertheless, we are persuaded by ideas from a number of different areas that the role of disgust has largely been unrecognised in the development of emotional and other disorders (see Chapter 9).
In support of the proposal we should note first that the basic emotions of sadness, disgust, and happiness appear within the first 3 months of infancy, with sadness appearing for example at the withdrawal of positive stimuli, disgust in relation to unpleasant foodstuffs, and happiness in reaction to familiar faces (e.g., Fischer, Shaver, & Carnochan, 1990; Lewis, 2000). As Fischer et al. (1990) argue, the early appearance of the basic emotions in the first year of life indicates that they provide important guiding principles around which development is organised, in
particular the development of the self and interpersonal relationships. However, it is a mistake to focus on disgust primarily as a reaction to food or excrement in the way that some analyses of the term suggest (e.g., Rozin & Fallon, 1987; see Chapter 9). Disgust-based disapproval is used by significant others to socialise the developing child in a range of permissible expressions and activities, not simply those related to food and the contents of nappies (Miller, 2003). The gradual development of self-disgust in relation to bodily activities and emotional expression provides the basis for a number of disorders in addition to depression (see Chapter 9).
The second point in support of this proposal is that there are a number of complex emotions such as shame, guilt, and embarrassment whose role in psycho-pathology has been highlighted in the psychoanalytic literature, all of which may be derived from the basic emotion of disgust (e.g., Johnson-Laird & Oatley, 1989). Although Freud (1910) proposed that guilt did not appear until 5 or 6 years of age following the resolution of the Oedipus Complex, it is now clear that even by age 2 children show considerable individual differences in the expression of shame and guilt; for example, in Western cultures girls are more likely to show shame and boys to show guilt in relation to an apparent transgression (Barrett et al., 1993). Shame was defined by Barrett and her colleagues to be disgust towards the self, and guilt to be disgust towards some action carried out by the self rather than towards the self per se. Because guilt led to more active attempts to obviate the transgression ("Amenders" in the researchers' terms) whereas shame led to more passive withdrawal ("Avoiders"), one can identify even at 2 years of age strategies that are successful in repairing mood and self-esteem in contrast to strategies that maintain a negative self-state. Indeed, one is reminded of the distinction made by Nolen-Hoeksema (1990, 2002) between "Ruminators" and "Distractors", the former individuals being more likely to maintain a negative self-state through a comparatively passive preoccupation with that state and the events that led to it, the latter being more active and using other activities to escape from the aversive feelings. Moreover, Nolen-Hoeksema has found that Ruminators are both more likely to be women and more likely to become depressed.
Other evidence that shame and humiliation may provide important factors in the onset of depression has come from two separate studies of the same population in Islington, north London. In the first study, Brown et al. (1995) reported that life events that led to feelings of humiliation and entrapment in addition to loss or danger were more predictive of depression onset than loss or danger events that did not include humiliation and entrapment. In the second study, Andrews (1995) found that a strong link between childhood physical and sexual abuse and the later occurrence of depression was mediated by the experience of bodily shame that typically developed in the teenage years, presumably in response to the teenagers' own physical and sexual development. Andrews interpreted her findings as evidence for Gilbert's (1989) proposal that shame relates to the experience of defeat and inferiority, implicit in the experience of abuse (see Gilbert, 2004, for a recent summary). A recent replication of the Islington studies was carried out by Kendler, Hettema, Butera, Gardner, and Prescott (2003) using a large twin register in Virginia, who found that loss events linked to humiliation were the category of life events most likely to lead to major depression, thereby providing support for the proposed sadness-self-disgust (in the form of shame and humiliation) link proposed here.
Our third line of argument is a definitional one, although it also draws on the phenomenology of depression. It is clear that the phenomenological "taste" of dysphoric mood or of depression is not one of sadness, though sadness is there as a component. Nor, as we have argued above, is it anxiety that adds the missing ingredient. The additional ingredient, we suggest, is a loathing of the self, a loathing of relationships and ambitions, and a loathing of the state itself, which colours the world grey. In a state of dysphoria this additional ingredient is there just as a soupçon, just enough to add its unique taste, although in severe depression the contribution may be completely overwhelming. We suggest, therefore, that the core phenomenological state is a combination of sadness and self-disgust.
Was this article helpful?