Enriching practice of dialectic behaviour therapy with the dynamic maturational model of attachment and adaptation

Introduction

In this article, I describe the peculiarities associated with each of the Type A and Type C dimensions of attachment strategy. It is too much of a simplification to say that the under-controlled emotionally unstable adolescents have Type C strategies, and the over-controlled have Type A strategies, although there is a grain of truth in this. The detailed dynamic maturational model (DMM) classifications of attachment strategies of patients with emotional instability, together with the DMM system for clarifying strategy modifiers, will be used to illustrate the complexity of personality functioning which needs to be addressed in treatment. The different strategies pose complementary challenges for therapists to establish and retain their treatment alliances, and help explain why the elaborate team consultation and self-treatment in a dialectic behaviour therapy (DBT) programme is essential for success of treatment of this patient population. The demands made on clinicians (and other family members, close friends, etc.) by these patients are often confusing. DMM can bring some clarity to understanding the chaos, and so further development of the model at the heart of the DBT approach to understand the development of emotional instability. DMM adds complexity in a manageable way, and hopefully can therefore prove useful for holding out with this challenging patient population. It also provides signposts to integrate DBT and the more recent development, radically open DBT (RO-DBT; Lynch, 2014), which has been developed for those with over-controlled personality development.

Memory and danger

When we have experienced something seriously threatening, the memory sticks and intrudes when our memory is jogged by triggers in the present or imagined future. Maybe we do not see the links, but the kick of the past makes itself felt in the present and knocks us off balance. The signals stream in from our body to our ‘enraptured’ brains, and action follows according to safety first principles, coordinated action to be set in motion by brainy initiatives (not necessarily cortically guided, in fact more probably not). The goal is survival, with maximal comfort given the circumstances: that is, the evolutionary prerogative, which after puberty adds procreation to survival. And it can all happen without us ever being aware of what happened, before we are chastised for how we flew off the handle. A wise response lies buried under the debris of the past.

In order to assist adolescents to manage their past and respond with greater ease and with fewer costs in the future, new learning has to take precedence over prior learning. New memories have to become durable and dominant, which requires motivation and repetition, ¹ and to be associated with alternative action strategies. The past needs to be left behind, rather than turning up trumps to outwit the best intentions of a novice meditator becoming more mindful. Which principles govern learning in the different memory systems, and how do we understand the ways in which these experiences come together to bias our repertoire of responses from those we have used impulsively in the past? New responses need to be added to the repertoire, and subsequently enabled to take precedence over tried responses, which otherwise will regularly trump new learning. But as emotional learning (perceptuo-affective learning in implicit memory) never dies, being resurrected whenever there are doubts about the veracity of current perceptions being different from those of the past (perception is about 90% memory according to Gregory, 1998; see also Dell, 1985; Popper & Eccles, 1977), we are all vulnerable to repeat past ‘mistakes’. This happens especially if cortical control (executive function) is jeopardised, such as when drunk. As Peter Cook, who had problems with alcohol use, put it to Dudley Moore when asked what he had learnt from his past mistakes, ‘I could repeat them exactly’. Radical acceptance ² is necessary as an acknowledgement of the priority and effectiveness of earlier emotional learning to get us out of the holes we have been in, by applying our best bets from previous experience.

DMM has moved attachment theory away from Bowlby’s ‘working models’ towards a concept espoused by Popper and Eccles (1977), Damasio (1995) and Bronfenbrenner and Ceci (1994, in their bio-ecological model which expands the initial bio-social model adopted by DBT). We become ‘disposed’ to act in a particular way given our structure (but not predetermined to do so), based on the peculiarly personal representations engendered by our perceptions (Dell, 1985; Popper & Eccles, 1977), and activated in our various memory systems. DMM makes use of the concept of ‘dispositional representations’ (DR). Our emotional learning (LeDoux, 1998) biases our response repertoire towards what worked well-enough previously. Issues of survival, comfort and reproductive possibilities set the priority parameters for what subsequent room we can give our learning from explicit memory systems, what we have learnt by rote and can recollect from experience (source, semantic and episodic memories). And yet under perceived threat, our implicit procedural memory of defensive action patterns jumps into effect before we have had a chance to weigh up the pros and cons, and realise that the threat was not justified, just an echo from the past. Making use of our reflective function through integrating information from previous experience, episodic memory, and what others have told us, semantic memory, is diabolically slow when you face your personal ‘lion’. The general form to these defensive responses varies according to attachment strategy (Crittenden, 2002; see also familyrelationsinstitute.org and patcrittenden.com for easy access to articles covering DMM).

A typical DMM classification of attachment of a person who appears emotionally unstable

Crittenden and Newman (2010) presented a series of DMM evaluations of attachment in emotionally unstable people in connection with an article comparing DMM classificatory principles (Crittenden & Landini, 2011) with the Berkeley Adult Attachment Interview (AAI) system of classification associated with Main and Goldwyn (1984–1994; Main, Goldwyn, & Hesse, 2003). A Berkeley classification of UtrE3 (unresolved trauma and using an enmeshing strategy) looks simple, whereas the DMM classification is much more complicated to follow. Nevertheless, it is clinically more enriching. It provides us with an idea of the wide range of processes likely to be affecting the function of a patient (see **) with an emotional instability – and therefore a broader base from which to integrate the wisdom within DBT and a current development in attachment theory. The DMM version of attachment theory gives us a more detailed clinical handle to address interpersonal processes and structure therapy. ³ In what follows, I will look at this patient subset as illustrated by one DMM-informed classification, and not concentrate on DBT and DMM applications to eating disorders (see Lynch et al., 2013, for applying RO-DBT to anorexia nervosa), avoidant personality disorder or other diagnostic groups.

Here is a pretty typical classification from Crittenden and Newman’s article: Dp Utr(p&ds)_{PA, aban} l(p & ds)_many A3,4,5(8) C5-6 [ina]_h (the patient received a UtrE3 Berkeley system classification). This looks almost like computer programming, and it necessitates a translation. ⁴ First comes the ‘modifiers’ of the attachment strategy; then follows the strategy familiar as being A, B or C in childhood, and with the continued use of the same signifiers into adulthood in the DMM system (the Berkeley system uses Ds for dismissing – corresponding to Type A, E for enmeshed – corresponding to Type C, and F for balanced – corresponding to Type B). Finally, there comes a phenomenon which can cut across both the display of the strategies and their modifiers.

Starting with the modifiers of strategy, we find Dp standing for depressed, being used here in the sense of Seligman’s learnt helplessness concept (Abramson, Seligman, & Teasdale, 1978); the strategy has ceased to be effective in regulating close-relationships to modulate own dis-ease. As expected from the research literature and identified in the Berkeley system, the effects of trauma are found in the evaluation of attachment, written here as U (for unresolved), tr (for trauma) and also l (for unresolved loss). After the Utr and Ul comes the signifier for how the unresolved issue is being addressed in the interview. In this case with a clinically confusing mixture of p – preoccupying, and ds – dismissing processes, that is, the traumas and losses are both apparently set alight by tangentially related triggers, appearing unexpectedly and preoccupying attention, as well as them dropping out of awareness at times when it would have been natural for them to have appeared. Although this may seem paradoxical, the duality is nevertheless coherent with what we now know about the effects of trauma on memory. High levels of cortisol play havoc with brain cells, being associated with apoptosis in certain regions. Additionally, survival depends on identifying potential risk of trauma, even when triggers are only minimally similar to what one experienced previously, and hence associated with a natural perceptual preoccupation. The subscripts denote what the trauma and loss refer to, a kind of clinical aide-memoire – PA = physical abuse, aban = abandonment and the losses were multiple (nevertheless Ul is only used for potential attachment figures).

So far the success of DBT strategies in this patient group would, according to the above classification in DMM terms of one particular patient, appear successfully to address

This will be extra challenging because of the underlying strategies, which also reflect Type A and Type C processes (Crittenden & Landini, 2011 for information on identifying all modifiers and strategies when evaluating adult attachment classification according to DMM principles).

These patients are bound to confuse therapists and others who have stood by them. And it gets more confusing when we look at the details of the basic strategies. Type A and C dimensions of functioning can each be further classified into one of eight categories in the DMM system. Think of the categories more like ways of dividing clouds, than a pack of cards. It is as if Types A and C reflect different dimensions, a cold front coming in rather than a warm front, with the characteristics mentioned above. Within the general lines of information processing typical of the As and Cs, they divide into ‘lines’ typical for either the odd or even numbered within each A or C dimension (the A1, 3, 5 and 7, the A 2, 4, 6 and 8, etc.). The even numbered Type As tend both to exaggerate their own responsibility for rectifying situations, as all As, and also, if they do not succeed, to be preoccupied with self-blame, whereas the odd-numbered Type As will primarily be exonerating their VIPs. Within the Type C dimension, the even numbered Type Cs tend to display primarily their fear and need for comfort, whereas the odd-numbered Type Cs display primarily aggression (but each can switch over to the non-dominant display if they fail to elicit the desired response). This complexity of cloud systems and warm and cold fronts, with different pairings along each front, requires a complex dialectical synthesis.

Back to the specifics of the classification of the emotionally unstable person: the A3 signifies a ‘compulsive caregiving’ strategy, coherent with a tendency to idealise her VIP. This might be expressed in the way in which a patient is more concerned about the needs of powerful others around her, than identifying her own needs or being able to use others to help her predictably diminish her dis-ease. In this way, she maximises the chance that she will not be rejected and pushed-aside for another. The A4 ‘compulsive compliance’ will surprise some who work with these patients, whom we often experience as being very self-centred (a Type C characteristic) and not in the least compliant. The compliance addresses the need to maximise their lovability and maximise access to their caregivers, this time for doing the bidding of the VIPs. ⁵ It may also reflect a VIP who has primarily noticed her child’s performance, which has been fine-tuned to maximise praise and recognition, but not seen beyond this to her daughter’s subjective state of well-being. The A5 strategy becomes available in adolescence as the adolescent drive matures towards developing a network away from the home, and in which sexual favours can give a feeling of a compensatory retrieved intimacy, unavailable in the home base. Physical intimacy or social bravura can put a gloss on an emptiness, which has plagued them throughout childhood. Such contact can also ring hollow when important needs arise. We can expect developmental processes to have shaped their language for sharing their dis-ease (Wilkinson, 2003), with needs vaguely conveyed and hard to read. The fear of abandonment is lurking, but usually unacknowledged in this group. When A(8) is in brackets, it conveys that in an AAI there were not enough markers to be definitive about the presence of this strategy, and I will not discuss it further just here; see otherwise Crittenden and Landini (2011).

The challenge for DBT with these three characteristics within the Type A strategy involves the therapist

before the patient disappears into a cloud of social networking which goes well beyond Facebook contact – and potentially retreating from all contact with the therapist, in spite of from time to time following the staked-out path.

And we still need to get to the ‘preoccupied’ Type C strategy, here represented by C5 and C6. The Type C strategies are sometimes referred to as the obsessive strategies, in contrast to the compulsive Type A strategies. The C5 is characterised by a thirst for revenge, and the C6 for being rescued (see Crittenden & Landini, 2011, for the breadth of processes characterising these strategies). And this double whammy may well hit the DBT therapist head on;

When one strategy doesn’t force through what the patient was expecting, then the other is likely to hit in (C5) or there being a shift to demonstrative ‘helplessness’ (C6), with small digs to niggle what they hope to be a guilty conscience of the person whom they are ‘trying’ to manoeuvre. ⁶ The focus is on their point of view, and what they see as required – where the other person is expected to be the one doing the work to change the situation, and save them from being on their own.

In contrast, Type A are showing their adaptations to those who have wielded power in the past, and to the ‘powerful’ therapist in the present, with the patients making demands on themselves to change things. They easily become overwhelmed by their own silent guilt, and shame, at not meeting the standards of others. Confusion should reign for a novice entering the role of a DBT therapist. The structure provided in the treatment principles, and in the input from the team supervisor, proves essential to keep on track. Therapy strategies keep the balance between being involved and yet not enmeshed, involved and not rejecting. Loose the balance and the patient’s strategy will switch over from one Type C polarity to the other. The aim is synthesis, and I would hypothesise that one dimension for synthesis concerns the processes typical of Type A and Type C, and another a synthesis within Type C so that both the aggression and fears/need for comfort cease to be polarised. Flexible integration in a true synthesis gives the characteristics of the Type B strategy; and when previously A and C strategies have been employed, we describe the new synthesis as an ‘earned Type B strategy’.

DBT guidance, with supported reflection from the supervisor on following the treatment principles, is also needed if there is i ntrusion of the n egative a ffect (‘ina’): the final part of the classification which can cut across both strategy and modifiers. This would be the affect which was taboo and counterproductive under the contingencies with which the Type A strategies evolved. ⁷ In this case, the [ina]‘h’ conveys that there was only h istorical evidence that this had occurred in the past. There was no direct evidence of an intrusion, for example, a scatological outburst or aggressive rant, out of keeping with the otherwise pleasing and satisfying-of-others characteristic of these Type A strategies. It takes experience to distinguish this from the aggressive display, demanding notice and in character, of the C5. As the ‘ina’ classification was only in the history, I will not focus on this, preferring to illustrate the more central issues about how DBT principles can effectively address the main points connected with Dp, Utr and Ul, and Type A and C strategies.

Implicit learning, transformations of information and behavioural learning principles

Implicit learning can be demonstrated in all animals and appears to be present in humans from before birth. This learning occurs below the level of the cortex. It occurs according to classical learning theory principles. Schore (2012) postulates that psychotherapy should primarily be geared up to address the implicit learning of patients – their emotions and habits, seeing the focus on implicit emotional and procedural learning to be the next advance in treatment focus after the cognitive behavioural therapy (CBT) movement and mentalisation. In CBT patient, thoughts have had primacy. The modifications in second and third generation CBT seem to draw CBT more towards addressing implicit learning, so that when we get to DBT there is already a definite major focus on emotions and habits. Implicit learning is not open to reflection, but is mobilised automatically to inform working memory and prime dispositions to action (DR), in response to conditioned reflexes to environmental contingencies. It needs extreme effort of will to stop and think before acting – and it may well not be possible without intensive training. The more severe the danger to which the patient has been exposed, the less likely it is that they can risk time to think, until after they have already ‘blown it’. Compassion for the patient’s dilemma is easier when one understands the mechanisms of the implicit memory systems (see e.g. LeDoux, 1998), and then radical acceptance comes naturally, knowing that the patient really is doing his/her best – and yet at the same time has to do things differently, a fundamental dialectic in DBT.

The validity of the classification drawn from a scoring of a DMM-modified AAI interview (Crittenden & Landini, 2011, pp. 358–360) depends on the same pattern being found in at least three memory systems, for example, from two implicit memory systems, procedural for motor habits and/or the perceptuo-affective emotional memory system, and from the semantic and/or episodic explicit memory systems. Associated with the different strategies are different transformations of information, different twists to the way in which perceptions are transformed into meaningful contributions to the brain’s work in survival-maintenance and preparing the ground for procreation. These ensure we take action fast enough to protect ourselves, and at the same time optimise feeling comfortable, when exposed to everything from daily hassles to life-threatening events – and to sexual opportunity. Crittenden’s (2002) hypothesis, which informs her DMM model of adaptation, is that the Type A strategy arises in the context of predictable parental behaviour. It is just that the full range of affective signals of emotion do not facilitate intimacy, with external regulation of the emotion by the parent, or even identification of the emotion as such. For example, a child’s signs of fury at her frustration with her situation distance parents, rather than enabling them to help solve their child’s dilemmas and soothe her. Emotions fail to be labelled, which would have given the child a feeling label for her dis-ease (see Wilkinson, 2003, for a description of these processes). For example, crying can lead to parental frustration and rejection of the crying child. The child experiences (implicit memory for a process occurring before conscious awareness) that keeping a distance and disguising spontaneous affective displays of distress, including aggressive demands, enable them to remain reasonably close. Spontaneous sharing of dis-ease disappears, and temporary states give way to enduring traits through repetition and anchoring in developing brain structure. If a naturally occurring fear smile becomes shaped into a display of false positive affect (a characteristic of a Type A strategy; see an ethological analysis of smiles in Hinde, 1982, pp. 213–215), then it may be that the parent and professionals are seduced into thinking that the child is enjoying himself/herself. This has mislaid professionals as noted in diverse media coverage of children killed by their parents, when children have been seen in hospital or by child care services in the period before the murder (e.g. Baby P, Victoria Climbie and Daniel Pelka). Displaying her false positive affect and defusing parental rejection of her negative affects leads to the child avoiding the pain of rejection, but she has started a journey towards a false-self, and a disparity between how he feels subjectively (wordlessly as the emotions have not transited into feelings) and how she seems to others. When the process is marked, the emptiness described as a characteristic of emotionally unstable people is a natural consequence, just as the fear of abandonment.

The transformations of information which need to be addressed in therapy to facilitate synthesis of the several dialectics include for Type A, by adolescence, the omission of negative affect and the falsification of positive affect. By that age, they can distort their memories of causal sequences in such a way that they can retain an idealisation of powerful others (the odd-numbered strategies), or make it appear that their own failings were the driving force behind what happened (those even numbered strategies). They need help to identify the real chains of behaviour and vulnerability factors. Asking them why they did something only provides the adolescent’s distorted rationale which likely adds to our confusion.

The transformations are complementary to those arising with the Type C developmental dynamic. Early on, their experience with unpredictable responsiveness leads them to ignore information in temporal sequences as meaningless; they go on to develop skills in splitting their negative affects so that either their aggression is displayed exaggeratedly, or their fear or helplessness and need for comforting. They appear to rely on one strategy. Nevertheless, the affect not being displayed lurks, ready to be applied if the primary affective display does achieve its purpose, such as aggression leading to too much distance or apparently overwhelming need for care creating stifling closeness. In order to retain their certainty that the solution to their difficulties lies aside themselves, they more often creatively develop alternative causal sequences based on snippets of misunderstood perceptions.

DBT style chain analyses, in great detail going back over a period of time, are clearly the way to go. Yet the transformations which need to be illuminated differ depending on whether Type A or C transformations are operating. They pose differing challenges to the therapist. The patient protects himself from having to doubt his own experience-biased perceptions, which would otherwise leave him feeling as if he has landed in quick sand. The treatment team needs to appreciate this, and enable the therapist to think flexibly, not bound to one set of expectations about what might work.

DBT works towards accurate experience and expression of emotions. Type A’s hidden affects are to be validated (and in the process get named as feelings) and expressed in appropriate doses. Type C needs the hidden complementary affect to be validated enough (the affect which is not being displayed – aggression for the fearful/needing comfort, the fears and need for comfort for the aggressive), and for the exaggerated display to be balanced, yet not denied. Type C is the classic emotional reactivity characterised by apparent under-control, which is a Level 1 focus in DBT. I would also like to place a claim on the need to have as an equal priority the emotionally over-controlled underactivity of Type A. The reason for giving this as high a priority is because the explosive ‘ina’ phenomena, often perceived as psychotic ⁸ when they come out of the blue against the tide of apparent compliance and emotional shut-down, arise primarily associated with the high Type A–A5-8, although also reported as occurring with Type A4 strategies.

DMM-informed hierarchy of treatment principles

The implications of DMM for a treatment hierarchy leads to a modified version of Bowlby’s (1980/1988, see also ** for how these principles are applied to running an inpatient unit which addresses ‘off-beat’ personality development as part of adolescent clinical phenomenology).

But this doesn’t really help us know how to achieve each step in a durable way, and maintain reflection and thoughtful action in the face of past personal demons. It does point out the need for safety in the present, before work can start. The stress on patterns could usefully have been extended to specify directly the releasing contingencies, to which DBT pays great attention when carrying out a chain analysis. That is both those in the past and those which maintain current functioning. And does re-evaluating your view of yourself really lead to change when old triggers revive memories of danger? You can maybe sleep soundly when safe from real and imagined ghosts from the past, but triggers are bound to arise in the course of daily-life. You need skills to handle them. Similarly these general principles seem only vaguely related to the specifics highlighted in the detailed DMM classification.

DBT treatment principles

Conclusion

‘Wise-mind’ represents an integration and flexible use of cognitive and affective ways of living. In DMM terminology, a Type A strategy is described as ‘cognitive’, used in a specific way to convey that the reliability of affective information is low and instead they rely predominantly on their information about causal sequences and contingencies. The Type C strategy is complimentary in the sense that DMM describes it as an ‘affective’ strategy, used to convey a primary reliance on their own affects, with the information available in temporal ordering of events being unsound in their experience, and so overseen. Wise-mind can then be seen as flexible integration of the two modus operandi, which corresponds to the criteria for a Type B attachment strategy: not a shifting between, with an inability to choose the most helpful approach at any given moment, but a functionality which enables best choices given the circumstances, which depends on accurate reading of those circumstances.

Through highlighting the differences between preoccupying and dismissing modes, I believe that DBT therapists can find it easier to be compassionate, accept, validate, hold out, and manage their therapeutic relationships.