Thursday, February 24, 2011

Attachment theory in clinical supervision: Past, present, and future

This paper was written for my class on supervision and consultation class - part of a section our learning team did on attachment theory and clinical supervision. I'd like to add links and good stuff like that, but I don't have time - full references are listed at the end.

* * * * *

Attachment theory in clinical supervision

When John Bowlby and Mary Ainsworth (Ainsworth & Bowlby, 1991) began developing the attachment model for which both are now credited in revolutionizing psychotherapy, their focus was specific to the mother-child bond. Ainsworth joined Bowlby’s team in 1950 and left three years later to pursue another job (and because her husband had finished his Ph.D.), but they continued to share research and ideas over the rest of their careers. In some ways it was the perfect relationship in that he focused more on theoretical approaches while she preferred to work directly with the mother-child dyads and eventually developed the “strange situation” scenario, first reported in 1969 (Ainsworth & Bowlby, p. 339), to test Bowlby’s ideas.

In those early years, it likely was not conceivable that their work would be extended into nearly every other form of relationship, including adult romantic relationships (although they later helped develop this aspect of the model as well, see Ainsworth, 1989). If important new ideas are slow to gain acceptance, then attachment theory was a truly great idea—it has taken 40 years or more for their work to become widely accepted, but in the 1990s and early 21st century interest in all forms of attachment has exploded. Attachment theory has even hit the mainstream of relationship books with the recent publication of Attached: The new science of adult attachment and how it can help you find—and keep—love (2010) being only the most recent (and talked about) example, most of which claim attachment as a “new” science.

Over the past twenty-five years, as attachment theory has slowly infiltrated the therapeutic community and become a central element in many of the psychodynamic therapies, the processes of attachment bonding have also become a prominent element of some supervision strategies. Working from John Bowlby’s definition of attachment as “any form of behavior that results in a person attaining or retaining proximity to some other differentiated and preferred individual, who is usually conceived as stronger and/or wiser” (1977, p. 203), his perspective is also often true of the supervisor-supervisee relationship.

Pistole and Watkins (1995) offer a more in-depth definition of attachment in a paper that has become a standard reference for the attachment model in supervision:

[A]ttachment theory is specific to (a) people's biologically-based, normative tendency for proximity in an emotionally important relationship that provides protection, that is, a safe haven and felt security, as well as an anchor for exploration, and (b) the normative emotional distress reaction that ensues when the tolerable limits of the proximity are exceeded. (p. 458)

The biological basis of attachment, mentioned in this passage, is a central issue in more recent work on attachment and is a topic that will be addressed in more detail below. However, it is important to point out that Bowlby conceived of the attachment function as innate, as a “cybernetic behavioral system, biologically imprinted before birth” (Neswald-McCalip, 2001, p. 18). Our need to be relational, to bond with others, is hard-wired into the infant at birth from this point-of-view. More recent evidence seems to support that position (see Siegel, 1999; Schore, 1994).

Pistole and Watkins apply Bowlby’s model (and his definition of attachment provided above, in which the supervisor becomes the “preferred individual”) to clinical supervision by focusing on three fundamental areas: “developmental unfolding, the secure base, and pathological attachment styles” (p. 468). While Pistole and Watkins note that an actual attachment bond can form between the supervisor and supervisee, this generally is not the situation (however, other authors disagree, as will be seen below). More often the supervision takes on elements of the attachment bond, for example becoming the “secure base” or offering the element of emotional “holding.”

A. Developmental unfolding

The attachment model is essentially developmental in nature, seeking to understand the ways in which the parent-child relationship shapes affect regulation (Schore, 1994), brain development (Siegel, 1999), and any number of other maturational processes. Likewise, the supervision process is also developmentally oriented, regardless of which supervision model is employed. Pistole and Watkins observe the similarities in the relationships:

The attachment between counselor and supervisor would initially reflect close involvement and monitoring. Supervisees at the start of their training are often more in need, have to rely more so on their supervisors for help, guidance, and assistance. (p. 469)

As the supervisee matures and gains skills, s/he is no longer as dependent on the supervisor for guidance, yet there is a sense of comfort (haven of safety) in knowing that person is available when needed for support or assistance. The supervisory relationship functions as safe developmental space for the supervisee, promoting growth and maturation.

B. The secure base

The supervisor, as well as acting a “safe haven,” also serves as a secure base from which the supervisee can explore new skills and competencies. In this sense, a secure base is defined as a supervisory relationship that is able to “ground or hold the supervisee in secure fashion” (Pistole & Watkins, p. 469). The authors expand on this idea in the following way:

The secure base can be seen as serving a protective function, letting supervisees know that (a) they are not alone in their counseling efforts, (b) their work will be monitored and reviewed across clients, and (c) they have a ready resource or beacon—the supervisor—who will be available to them in times of need. (p. 469)

When the secure base is present, the supervisee feels more grounded in the learning process.

As the supervisee develops within the cocoon of the relationship (the safe haven) and with the guidance and support of the supervisor (the safe base), s/he is more apt to try new skills and find a unique voice as a therapist. The supervisee may try on different therapeutic models (cognitive behavior or psychoanalytic), attempt different interventions (somatic awareness or voice dialogue), and explore different ways to engage clients (role playing or a creative expression with art). This process is essential for the supervisee to find her or his own unique way of working with clients, conceptualizing cases, and to learn what does not work.

So how does one become a secure base as a supervisor? Pistole and Watkins identify a handful of qualities that the supervisor can embody to be a secure base: “availability, consistency, responsiveness, and judicious intervention” (p. 470). They add several more qualities identified by other researchers that are also important:

Empathic sensitivity and flexibility would also seem critical (cf. Osofsky, 1988; Watkins, in press-b). We, too, believe that appropriate structuring, setting of, and agreement on goals and tasks (see Bordin, 1983), and having set regular meeting times and meeting place contribute to the formation of a secure base. (p. 470)

What makes the base secure is that it is reliable, constant, nurturing, and is a place (person) to which one may return as needed. There are likely other factors involved, as well, that are unique to each relationship and each individual.

C. Problematic attachment styles

In his own work, Bowlby (1977, 1978) identified two basic versions of problematic attachment (no attachment style other than disorganized is pathological, only adaptive)—anxious attachment and compulsive self-reliance; he also added a variation, compulsive care-giving. Pistole and Watkins look at each those three pathological styles in turn (p. 471-473), however those are not the adult attachment patterns with which most people are familiar. Mary Main co-developed the Adult Attachment Inventory (AAI; George, Kaplan, & Main, 1984), an assessment tool that allows therapists or researchers to work directly with an individual rather than having to observe a dyad. The terminology builds on the original infant attachment styles (secure, avoidant, ambivalent-resistant, and [added after Ainsworth initial work] disorganized):

Like the avoidant infant, the adult who is dismissing of attachment minimizes the importance of attachment relationships. Analogous to the secure infant, the autonomous adult values intimacy and freely expresses her- or himself with respect to attachment. Similar to the ambivalent-resistant infant, the preoccupied adult is engrossed in attachment relationships, but cannot modulate stress through them. Finally, like the disorganized-disoriented infant, the unresolved adult experiences the periodic collapse of his or her predominant attachment strategy. (Atkinson & Goldberg, 2007, p. 8)

In looking at adult romantic attachments, which had been assumed to be the best indicator of attachment styles, Hazan and Shaver (1987) found that approximately 60% of adults self-identify as secure, 20% self-identify as avoidant, and around 20% self-identify as anxious-resistant (the outcomes tend to differ on the AAI). Most of the mainstream approaches to attachment (i.e., relationship books such as the previously mentioned Attached [Levine & Heller, 2010]) focus on these three simple attachment styles, confusing the terminology and, in general, losing the connection to the childhood patterns.

Neswald-McCalip (2001), however, also uses these three basic attachment patterns in her paper in support of Pistole and Watkins, and she does so without losing the complexity of the patterns. She identifies the secure pattern (seeing the attachment figure as available and supportive), those who tend to embrace challenges, enjoy exploration, and often ask for help when needed; anxious resistant (seeing the attachment figure as inconsistent or undependable), those who experience anxiety in new situations or when faced with challenges, who can seem needy and dependent, or who seem to always be in crisis; and anxious avoidant (see the attachment figure as inaccessible when needed, a pattern that can look a lot like learned helplessness), those who believe they are on their own, with no expectation of help when needed, who expect to be ignored, and who tend to be overly self-reliant (p. 20). A much more detailed model is available for clinical use from Brennan, Clark, and Shaver (1998), in which they generate a model based on two axes, low anxiety-high anxiety and low avoidance-high avoidance (offering four attachment styles, see appendix A).

Critical incidents and attachment supervision

As far as supervision work is concerned, there are basically secure and insecure (anxious) attachment styles in the supervisee. Taking this perspective as a general approach, Pistole and Fitch (2008) have articulated a supervisory model to employ when supervisees encounter a “critical incident” that triggers attachment patterns. The “important person” in attachment theory, the caregiver (supervisor), is the focus of their model, but they emphasize awareness of the basic terminology, including the care-giving bond (supervisory relationship), the safe haven and secure base (both the supervisor and the supervisory relationship), and exploratory behavior (trying new skills, and so on). They observe:

For a supervisee, the supervisor may be the preferred caregiver when stresses or anxieties are related to counseling sessions and training experiences.

The critical incident experience activates the supervisee’s attachment system because of the experience’s novelty (Bowlby, 1969) and the challenge that is associated with emotions, such as being drained, exhausted (Ellis, 1988), pressured, or defensive (Haferkamp, 1988). (p. 196)

When the supervisee presents as triggered and exhibiting attachment behaviors (largely variations on anxiety in this context), the supervisor is advised to operate from the care-giving bond, act as a secure base, and make the supervision session a safe haven. The goal is to deactivate the attachment system so that the supervisee can return to exploratory behaviors that will help them resolve the presenting issue.

Neuroscience of attachment and the mindful supervisor

Alan Schore (1994, 2003a, 2003b) has spent a couple of decades exploring and detailing the neuroscience behind attachment and how this model leads essentially to what is known as affect regulation. Also involved in similar research, but much more active in disseminating his work in books for the general reader (Schore’s book are dense with references and really aimed at professionals), Daniel Siegel has taken the findings of attachment theory and applied them to brain development (1999), the therapeutic process (2010), and to personal transformation (2007, 2010). The central goal of Siegel’s model, which he has termed interpersonal neurobiology, is to offer a “definition of the mind and of mental well-being that can be used by a wide range of professionals concerned with human development” (2006, p. 1).

In the same way that the attachment process helps wire the growing and developing infant’s brain and mind, so too does the interpersonal relationship between the therapist and client rewire the client’s brain, and likewise, by extension, so does the supervision relationship rewire the supervisee’s brain. In fact, all meaningful (i.e., emotionally important) relationships have this impact on the brain through a variety of mechanisms, including mirror neurons, neural plasticity, and modeling (Siegel, 2006; Schore & Shore, 2008).

Siegel employs an embodied cognition approach in suggestion that attachment relationships offer vertical integration (as well horizontal integration, but the focus here is on vertical processes) of brain functions, a process that mirrors at least seven of the nine basic elements of how mindfulness practice rewires the brain (see The Mindful Brain, 2007).

Linking the basic somatic regulatory functions of the brainstem with the limbic circuits’ generation of affective states, motivational drives, attachment, and appraisal of meaning and laying down of memory is a first layer of vertical integration.

Above the limbic circuitry emerged the neocortex, or “outer bark” of our evolving brains. The cortex, unlike the brainstem, is quite underdeveloped at birth and is shaped by both genetics and especially by experiences out in the world. In general, the posterior regions of the cortex are specialized for perception of the physical world (our first five senses) and the body itself is registered in the more forward aspects of this posterior region. (2006, p. 5)

He goes on to detail the involvement of the frontal lobe of the cortex, the prefrontal cortex, the dorsolateral prefrontal region, the orbital frontal area, the medial prefrontal cortex, and the anterior cingulate. Also involved are the insular cortex, and several other medial areas collectively referred to as the “middle prefrontal cortex” (p. 6).

This is the important and interesting piece—it is possible to see in attachment processes the same brain changes that result from mindfulness practice (a hot topic in psychological research at present). Siegel identifies nine qualities that are integrated within the middle prefrontal cortex (all of this is based on independent research by others in the field, including Schore): body regulation, attuned communication, emotional balance, response flexibility, empathy, insight, fear extinction, intuition, and morality (p. 6). Most of us are not fully integrated in this way, which makes mindfulness practice or a therapeutic relationship, or even a healthy, securely attached romantic relationship, necessary to achieve this level of integration.

A full explanation of how this works requires a longer citation from Siegel’s paper, but to put it in simplest terms, the vertical integration of body, brain stem, limbic system, and higher brain (the cerebral cortex) is the eventual outcome of healthy, secure attachment.

It is relevant to note that these nine middle prefrontal functions can be seen to emerge not only with mindful awareness practices, but at least the first seven are also be associated with the outcome of secure attachment between child and caregiver (13). This finding may suggest that experiences of “mental attunement” – interpersonal in the case of attachment or internal in the practice of mindful awareness – may be at the heart of developing an integrated brain and well-being. Healthy self-regulation, through relationships and self-reflective observation, may depend on the development of the integrated circuits of these prefrontal regions (12, 14, 15). (p. 7)

He goes on to explain how horizontal integration (bi-lateral integration of brain hemispheres) also plays a role in this process. The right hemisphere develops first after birth, which is why Schore has spent so much time writing about the importance of this early period in the development of affect regulation. In addition, the right brain (this is not a strict either/or, as both hemispheres share most skills) includes a wide range of functions that are important to healthy attachment patterns, including “stress response, an integrated map of the whole body, raw and spontaneous emotion, autobiographical memory, a dominance for the non-verbal aspects of empathy,” as well as a comfort with ambiguity, or “analogic” thinking, “meaning it perceives a wide spectrum of meaning,” not simply rational logic (p. 8).


All of this research grew out of Bowlby’s and Ainswoth’s (as well as many others) early research into parent-child attachment. The application to clinical practice came first, but it was not long before the interpersonal and intersubjective value of attachment theory was applied to the supervisory relationship. One need not know about all of this to be using it and benefitting from the ways attachment theory can support personal growth in the supervisee—but knowing about it does make it easier to apply the interventions of secure base, safe haven, and so on to deactivate the attachment patterns in the supervisee who has been triggered and to return him or her to an exploratory mode of learning and development.

Finally, as is apparent from the research of Schore and Siegel, among others, that there is much more going on in these supervision sessions than we are consciously aware. As the neuroscience continues to develop, it is likely we will learn new ways to mentor and supervise, ways that can rewire the supervisee’s brain (through emotional attunement) and allow them to be more effective therapists. It may even require that we rethink the separation of supervision and therapy—the two may be happening together even while we are not aware.


Ainsworth, M. D. (1989). Attachment beyond infancy. American Psychologist, 49, 709-716.

Ainsworth, M. D., & Bowlby, J. (1991, April). An ethological approach to personality development. American Psychologist, 46, 333-341. Retrieved from

Atkinson, L. & Goldberg, S., Eds. (2004). Attachment issues in psychopathology and intervention. Mahwah, NJ: Lawrence Erlbaum Associates.

Bowlby, J. (1977). The making and breaking of affectional bonds. I. Aetiology and psychopathology in the light of attachment theory. British Journal of Psychiatry, 130, 201–210.

Bowlby, J. (1978). Attachment theory and its therapeutic implications. In S. C. Feinstein & P. L. Giovachini (Eds.), Adolescent psychiatry. Vol. VI: Development and clinical studies (pp. 5-33). Chicago: University of Chicago Press.

Brennan, K., Clark, C. & Shaver, P. (1998). Self-report measurement of adult romantic attachment: An integrative overview. In J. A. Simpson & W. S. Rholes (Eds.), Attachment theory and close relationships (pp. 46-76). New York: Guilford Press.

George, C., Kaplan, N., & Main, M. (1984). Adult Attachment Interview Protocol (1st ed.). Unpublished manuscript, University of California at Berkeley.

Hazan, C. & Shaver, P. (1987). Romantic love conceptualized as an attachment process. Journal of Personality and Social Psychology, 52(3), 511-524.

Levine, A., & Heller, R. (2010). Attached: The new science of adult attachment and how it can help you find—and keep—love. New York: Tarcher/Penguin.

Neswald-McCalip, R. (2001, September). Development of the secure counselor: Case examples supporting Pistole & Watkins’s (1995) discussion of attachment theory in counseling supervision. Counselor Education & Supervision, 41, 18-27.

Pistole, M.C. & Fitch, J.C. (2008, March). Attachment theory in supervision: A critical incident experience. Counselor Education & Supervision, Vol. 47, 193-205.

Pistole, M. C., & Watkins, C. E., Jr. (1995, July). Attachment theory, counseling process, and supervision. The Counseling Psychologist, 23, 457-478. doi: 10.1177/0011000095233004

Schore, A. N. (1994). Affect regulation and the origin of the self: The neurobiology of emotional development. Hillsdale, NJ: Lawrence Erlbaum Associates, Publishers.

Schore, A. N. (2003a). Affect dysregulation and disorders of the self. New York: W. W. Norton.

Schore, A. N. (2003b). Affect regulation and the repair of the self. New York: W. W. Norton.

Schore, J.N. & Schore, A.N. (2008). Modern attachment theory: The central role of affect regulation in development and treatment. Journal of Clinical Social Work, 36:9–20. DOI 10.1007/s10615-007-0111-7

Siegel, D. J. (1999). The developing mind: How relationships and the brain interact to shape who we are. New York: Guilford Press.

Siegel, D.J. (2006, April/May). An interpersonal neurobiology approach to psychotherapy. Psychiatric Annals, Vol. 36, Number 4, 248-256.

Siegel, D.J. (2007). The mindful brain: Reflection and attunement in the cultivation of well-being. NY: W.W. Norton.

Siegel, D.J. (2010). Mindsight: The new science of personal transformation. NY: W.W. Norton.

Siegel, D.J. (2010). The mindful therapist: A clinician's guide to Mindsight and neural integration. NY: W.W. Norton.

Appendix A

Brennan, Clark, & Shaver (1998) offered this useful diagram in looking at adult attachment issues in individuals.

Figure 1. The two-dimensional model of individual differences in adult attachment.

No comments: