Web-only Feature

Web-only Feature

Internet Editor’s Note: Dr. Mallinckrodt and his colleague recently published an article titled “Meta-analysis of client attachment to therapist: Associations with working alliance and client pretherapy attachment” in Psychotherapy.

If you’re a member of the Society for the Advancement of Psychotherapy you can access the Psychotherapy article via your APA member page.

Not a member? Purchase the Psychotherapy article for $11.95 here.

Or, Join the Society for $40 a year and receive access to more than 50 years of articles.


In 1988 John Bowlby published a groundbreaking collection of his lectures and essays. He inspired a generation of researchers by asserting that the therapist-client relationship has key features in common with parent-child attachments. Roughly coinciding with the 25th anniversary of Bowlby’s book, four meta-analyses have recently been published. These articles and other summaries take stock of the many insights that adult attachment theory offers about the psychotherapy relationship and generate suggestions for how the research findings can be put to work in the therapy hour. Here a few of the highlights.

Psychotherapy Relationships through an Attachment Theory Lens

Five distinct features of healthy parent-child attachment have been identified, each with a parallel in many therapist-client relationships (Mallinckrodt, 2010). First, clients tend to view their therapist as “stronger and wiser” – someone who offers knowledge and a relationship that promises a degree of protection and security. Many clients seek emotional proximity to their therapist, view the therapist as a safe haven when they feel distressed, and derive a sense of security from their therapist — who then serves as a secure base for exploring threatening material. (“Save haven” and “secure base” are terms popularized by Bowlby.) Finally, some clients also experience intense ambivalence, separation anxiety, and fear of termination.

Not all therapy relationships activate each of these features. Some approaches do not emphasize the psychotherapy relationship as a vehicle for change, and even those that do can be constrained by session limits with insufficient time for all five attachment features to develop. However, a third limitation stems from clients themselves. Many experience interpersonal problems that limit their capacity, at least initially, to establish a secure attachment with their therapist. Therapists too experience challenges in forming productive attachments, especially with particular clients. Thus, attachment theory provides a framework for a broadened conceptualization of transference and countertransference.

Secure and Less-Than-Secure Psychotherapy Attachments

When we perceive a significant threat in our environment – whether it be tangible like developing flu symptoms, or psychological such as being criticized by one’s boss – we tend to seek comfort from a secure attachment figure. Adults vary in the social competencies they have acquired in childhood for recruiting social support, with important consequences for individual and group psychotherapy (Mallinckrodt, 2000). But what if a secure attachment figure is not available?

Mikulincer and Shaver (2007) describe how persons blocked in their efforts to seek comfort from a secure attachment engage in one of two “secondary strategies.” Persons who rely on a hyperactivating strategy magnify their expression of distress, closely monitor attachment figures for signs of abandonment, and attempt to establish very close proximity to a potential source of comfort. The “proximity” could be actual physical distance, but more often takes the form of close emotional dependency. In contrast, persons who rely on a deactivating secondary strategy expend great effort to divert their attention from both distress-evoking stimuli and attachment-related thoughts and feelings. These individuals typically avoid emotional intimacy.

These strategies correspond generally to the dimensions of attachment anxiety and attachment avoidance assessed by the self-report, Experiences in Close Relationships Scale (ECRS, Brennan, Clark, & Shaver, 1998). The dimensions are relatively orthogonal. Low scores on both subscales are thought to reflect relatively secure romantic attachment in adults from Western cultures. But note that due to cultural differences in how couples express affection and provide support to one another, the ECRS appears to overestimate attachment insecurity in persons from Taiwanese and perhaps other East Asian cultures (Wang & Mallinckrodt, 2006).

Therapists observe the hyperactivating strategy when their clients greatly intensify each of the five basic attachment features. For example, clients see their therapist as far more capable then themselves, become highly dependent on the therapy relationship (i.e., the seek close emotional “proximity”), and are largely unable to comfort themselves. In contrast, some clients use a deactivating attachment strategy in therapy, just as they do in other close relationships. They resist emotional engagement with their therapist, divert attention from distress evoking memories, disclose little, and downplay the importance of the therapy relationship.

Interpersonal Consequences of Insecure Attachment

A study based on a mostly female sample of graduate student therapists-in-training who participated in 12 interpersonal growth groups collected self-ratings of attachment anxiety and avoidance. Attachment anxiety was associated with self-reported “too friendly” interpersonal circumplex problems, whereas attachment avoidance was associated with hostile-dominant problems (Chen & Mallinckrodt, 2002). The same student therapists were asked to provide round robin interpersonal ratings of all other group members. These data were portioned into perceiver and target variance. Evidence of transference was inferred when an individual tended to systematically perceive target group members differently than the group consensus of the same targets (Mallinckrodt & Chen, 2004). For example, variance in attributions of hostility in other group members was associated with the perceivers’ memories of parents as intrusive and controlling.

Thus, a capacity for secure attachment allows adults to effectively recruit social support, and also to avoid (or quickly terminate) maladaptive relationships that add to stressful life events (Mallinckrodt, 2000). However, the problems of persons with even relatively mild insecure attachment tendencies – like these student therapists – are compounded because insecurity tends to be related to interpersonal problems that distance others, and to systematic misperceptions of others that compound this distancing.

Meta-analyses of Psychotherapy and Attachment

The Client Attachment to Therapist Scale (CATS, Mallinckrodt, Gantt, & Coble, 1995) was based on a factor analysis of over 130 client responses to a large item pool. Three subscales emerged: Secure, Avoidant, and Preoccupied. Studies that used the CATS were the focus of the most recent meta-analyses (Mallinckrodt & Jeong, 2015). Conclusions of all four meta-analyses must be tentative because the literature on client attachment in psychotherapy is not extensive, even after more than 25 years. (The four meta-analyses included only14-24 studies.) Nevertheless, here are some key points that emerge:

  • Client pre-therapy attachment security (i.e. low avoidance and low anxiety) is associated with stronger working alliances (Brenecker, Levy, & Ellison, 2014; Diener & Moore, 2011).
  • Client attachment anxiety but not avoidance is negatively associated with therapy outcome (Levy, Ellison, Scott & Bernecker, 2011).
  • Clients’ pre-therapy attachment security (i.e., low ECRS Anxiety and Avoidance subscale scores) is strongly associated with secure attachment to therapist after the first 3-6 sessions of therapy (Mallinckrodt & Jeong, 2015).
  • However, an exclusive one-to-one correspondence between pre-therapy attachment avoidance and CATS-Avoidant; or pre-therapy attachment anxiety and CATS-Preoccupied was not found (Mallinckrodt & Jeong, 2015).

Thus, when it comes to adult attachment security and the psychotherapy relationship or outcomes, apparently “the rich get richer.”   Clients with a capacity for secure attachment before therapy tend to develop more secure attachments and productive working alliances with their therapist. Both anxiety and avoidance appear to interfere with the psychotherapy relationship, but attachment anxiety may be especially detrimental to therapy outcome. Finally, the meta-analysis recently published in Psychotherapy (Mallinckrodt & Jeong, 2015) suggests that therapists can expect clients who are securely attached to others to form a secure psychotherapy attachment to them. But they should not expect a one-to-one correspondence between outside forms of insecurity and insecure dimensions of therapy attachment. One reason for the complexity may be that client insecure attachment sometimes takes the form of superficial “pseudo-secure” attachment to therapist (Mallinckrodt et al., 2015).

Summary and Suggestions for Psychotherapy

Two studies suggest that secure attachment to therapists provides a “secure base” as Bowlby predicted that allows clients to explore more deeply during the middle phase of therapy (Mallinckrodt, Porter, & Kivlighan, 2005; Romano, Fitzpatrick & Janzen, 2008). Evidence is mounting to suggest that clients’ pre-therapy attachment insecurity interferes with productive working alliances (Brenecker et al., 2014; Deiner & Moore, 2011) as well as secure attachment to therapist (Mallinckrodt & Jeong, 2015, and limits favorable therapy outcomes (Levy et al., 2011).

How then are therapists to overcome these challenges? From the perspective of interpersonal psychotherapy, the process of building a secure attachment and productive alliance is the central goal of the therapeutic work with clients who begin with a limited capacity for this type of relationship (Teyber & McClure, 2011). For these clients, a fully secure attachment to therapist with all five critical features in place is not a starting point for the work, but rather a marker that it is nearing its conclusion (Mallinckrodt, 2010).

A qualitative study of experienced therapists presented a detailed picture of how they approach working with hyperactivating and deactivating clients (Daly & Mallinckrodt, 2010). The crux of their approach was to deliberately regulate therapeutic distance. The optimal distance to create a corrective emotional experience differed for each of the two types of clients, and differed through the three therapeutic phases of engagement, working through, and termination. Readers may find this article especially useful because, with remarkable generosity, 10 of the 12 experts allowed the full transcript of their interview to be published through an online link in the article.

A new analysis of all twelve transcripts resulted in development of the Therapeutic Distance Scale (Mallinckrodt, Choi, & Daly, 2015). The 40-item TDS is a self-report measure designed to assess clients’ experience of the therapists’ attempt to regulate therapeutic distance. The measure operationalizes a theory derived from Daly and Mallickrodt’s (2010) interviews. According to this model, clients with hyperactivating tendencies are helped by gradually increasing therapeutic distance — to the extent that the client’s frustration will allow, in order to create a corrective emotional experience of growing autonomy. Clients with deactivating tendencies can be helped when the therapist insists on gradually less therapeutic distance — to the extent that the client’s anxiety will allow, in order to create a corrective emotional experience of growing engagement. Preliminary results appear to support this model of attachment and corrective emotional experience (Mallinckrodt et al., 2015). I am seeking collaborators interested in helping me advance the work. If you are interested, please contact me.

Cite This Article

Mallinckrodt, B. (2015, May). Attachment theory and the psychotherapy relationship – Summarizing what we know. [Web article]. Retrieved from www.societyforpsychotherapy.org/attachment-theory-and-the-psychotherapy-relationship-summarizing-what-we-know


Bowlby, J. (1988). A secure base: Parent-child attachment and healthy human development. London: Routledge.

Bernecker, S. L., Levy, K. N., & Ellison, W. D. (2014). A meta-analysis of the relation between patient adult attachment style and the working alliance. Psychotherapy Research, 24, 12-24, 10.1080/10503307.2013.809561

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

Chen, E. C., & Mallinckrodt, B. (2002). Attachment, group attraction, and self-other agreement in interpersonal circumplex problems and perceptions of group members. Group Dynamics, 6, 311-324. doi: 10.1037/1089-2699.6.4.311

Daly, K. D., & Mallinckrodt, B. (2009). Expert therapists’ approaches to psychotherapy with adult clients who present with attachment avoidance or anxiety. Journal of Counseling Psychology, 56, 549-563. doi: 10.1037/a0016695

Diener, M. J., & Monroe, J. M. (2011). The relationship between adult attachment style and therapeutic alliance in individual psychotherapy: A meta-analytic review. Psychotherapy, 48, 237–248. doi:10.1037/a0022425

Levy, K. N., Ellison, W. D., Scott, L. N., & Bernecker, S. L. (2011). Attachment style. Journal of Clinical Psychology: In session, 67,193-203. doi: 10.1002/jclp.20756.

Mallinckrodt, B. (2000). Attachment, social competencies, social support and interpersonal process in psychotherapy. Psychotherapy Research, 10, 239-266. doi: 10.1093/ptr/10.3.239

Mallinckrodt, B. (2010). The psychotherapy relationship as attachment: Evidence and implications. Journal of Personal and Social Relationships, 27, 262-270. doi: 10.1177/0265407509360905

Mallinckrodt, B., & Chen, E. C. (2004). Attachment and interpersonal impact perceptions of group members: A Social Relations Model analysis of transference. Psychotherapy Research, 14, 210-230. doi: 10.1093/ptr/kph018

Mallinckrodt, B., Choi, G., & Daly, K. (in press). Pilot test of a measure to assess therapeutic distance and its association with client attachment, and corrective experience in therapy. Psychotherapy Research. doi: 10.1080/10503307.2014.928755

Mallinckrodt, B., Gantt, D. L., & Coble, H. M. (1995). Attachment patterns in the psychotherapy relationship: Development of the Client Attachment to Therapist Scale. Journal of Counseling Psychology, 42(3), 307-317. doi: 10.1037/0022-0167.42.3.307

Mallinckrodt, B., & Jeong, J.-S. (2015). Meta-analysis of client attachment to therapist: Associations with working alliance and client pre-therapy attachment. Psychotherapy, 52, 134-139. doi: 10.1037/a0036890.supp

Mallinckrodt, B., Petrowski, K., Sauer, E. M., Tishby, O., Wiseman, H., & Levy, K. N. (June, 2015). Pseudo-secure vs. individuated-secure client attachment to therapist: Implication for therapy process and outcome. In B. Strauss (Chair), Studies related to the construct of adult attachment and its assessment. Symposium presented at the annual international convention of the Society for Psychotherapy Research, Philadelphia.

Mallinckrodt, B., Porter, M. J., & Kivlighan, D. M. Jr. (2005). Client attachment to therapist, depth of in-session exploration, and object relations in brief psychotherapy. Psychotherapy: Theory, Research, Practice, and Training, 42, 85-100. doi: 10.1037/0033-3204.42.1.85

Mikulincer, M., & Shaver, P.R. (2007). Attachment in adulthood: Structure, dynamics, and change. New York: Guilford Press.

Romano, V., Fitzpatrick, M., & Janzen, J. (2008).The secure-base hypothesis: Global attachment, attachment to counselor, and session exploration in psychotherapy. Journal of Counseling Psychology, 55, 495–504. doi: 10.1037/a0013721

Teyber, E., & Mcclure F. H. (2011). Interpersonal process in therapy: An integrative model (6th ed.). Belmont, CA: Brooks/Cole.

Wang, C. C., & Mallinckrodt, B. (2006). Differences between Taiwanese and U.S. cultural beliefs about ideal adult attachment. Journal of Counseling Psychology, 53, 192–204. doi:10.1037/0022-0167.53.2.192


  1. LS

    I have to disagree with some of this. As someone who is in therapy for C-PTSD, I have found myself a strange and confusing mixture of avoidant and anxious. Avoidant as in: wanting to have DIY therapy and remain distrustful for my own protection, and anxious as in: fearing termination for being too much of a difficult client.
    My therapist has responded by adapting sessions as needed, and if he hadn’t done that I think I’d have just given up and left. If he has decided to respond to my anxious part by creating distance I would have found that rejecting and damaging. The more he gave the more I relaxed and could drop my guard.

    • Ellie Dupont

      I completely agree with this comment! I was diagnosed with DID two years ago and apparently have a disorganised attachment. Therefore also swaying between avoidant and fearful constantly. My therapist has also had to adapt accordingly to different parr’s attachment needs. It’s a delicate balance of helping anxious parts grow in autonomy and avoidant parts develop more willingness to connect. This article could do with discussion on disorganised attachment, where a client oscillates between different styles in more extreme ways.

  2. Charlotte Attwood

    It seems that your therapist did what the article says is most useful for avoidant types. As in, “Clients with deactivating tendencies can be helped when the therapist insists on gradually less therapeutic distance” – as in, they close the distance for avoidant types.

    I’d say even the idea that you are a difficult client is an example of avoidant attachment- it’s hard to tell but it shows insecurity and that you placed your own needs quite low on the scale. So, perhaps, you didn’t really attach, you saw yourself as quite separate? Having anxiety and anxious attachment appear to me to be two different things.

  3. Charlotte Attwood

    “Attachment anxiety was associated with self-reported “too friendly” interpersonal circumplex problems,”

  4. Siobhan

    interpersonal psychotherapy which states “ The process of building a secure attachment is the central goal of the therapeutic work. For clients with attachment issues, when they have developed a fully secure attachment to the therapist, this is a marker that therapy is nearing its conclusion”

    Wow! So the client develops their first healthy and secure attachment and then loses the relationship with the therapist!

    So when looking at this from an Internal Family Systems  perspective, for me it would look like this.

    1. Client enters therapy with insecure attachment and no sense of Self.

    2. Therapist begins work to develop a secure attachment to their own “Therapist Self “

    3. Secure attachment to the “Therapist Self” is achieved. No work has been done to develop the clients sense of Self or attachment to Self.

    4. Therapist sees that a secure attachment to their “Therapist Self” has been achieved and therapist Pats themselves on the back.

    5. Therapist initiates the wind down of therapy.

    6. Client loses the one secure attachment that they may have built in their life.

    7. Therapy ends, and the client is left without any Therapist Self to remain stable and has no sense of Self within their own bodies.

    8. Client uses coping methods that they have learnt in therapy to maintain stability, but still feels that something is wrong. Client feels grief over the loss of the “Therapist Self” This is because they have not developed an attachment to The Self .

    9. Client returns to therapy. The cycle begins again.

    There has to be a better way!

    • Charlie

      I agree. I quite honestly don’t see the any benefit in withdrawing from the therapeutic relationship once a secure attachment has finally been formed. This would only reinforce the clients distrust of people and inability to form any healthy relationships in the future.


Submit a Comment

Your email address will not be published. Required fields are marked *