When Attachment Styles Meet in Therapy

This article explores how therapist and client attachment styles interact in therapy – and what it means for rupture, repair, and clinical practice.

Related articles: Parallel Process: Burden, Blessing, or Clinical Signal?, Ten Common Mistakes Therapists Make (And What To Do Instead).

Introduction

Attachment theory has long guided our understanding of client presentation, relational expectations, and therapeutic engagement. Yet, in clinical practice, attachment is not a one-person phenomenon. It unfolds dynamically between two nervous systems – each bringing its own history of safety, threat, and connection into the room.

While clinicians are typically trained to recognise attachment patterns in clients, far less attention is given to how therapists’ own attachment styles influence the therapeutic relationship. This omission is not trivial. Emerging research suggests that therapist attachment patterns can shape alliance formation, responsiveness to rupture, and even treatment outcomes (Degnan et al., 2016; Mikulincer & Shaver, 2019).

In practice, this means that what unfolds between therapist and client is not simply “transference” or “countertransference” (more about these here), but a co-created attachment system – one that can either facilitate corrective emotional experiences or quietly reinforce longstanding relational patterns.

In this article, we explore how therapist and client attachment patterns interact in the therapy room, how these dynamics contribute to rupture and repair, and what clinicians can do to work with – rather than against – these powerful relational forces.

Why attachment matters for both sides of the room

Attachment theory, originally articulated by Bowlby (1969), describes how early relational experiences shape internal working models of self and others. These models influence expectations of care, trust, and responsiveness across the lifespan. In adulthood, attachment patterns are often described along dimensions of anxiety and avoidance, shaping how individuals seek proximity, regulate emotion, and respond to perceived relational threat (Mikulincer & Shaver, 2017).

In therapy, clients’ attachment styles are frequently conceptualised as drivers of engagement, resistance, and relational testing. However, the therapeutic relationship itself functions as an attachment bond, particularly in longer-term or emotionally focused work (Mallinckrodt et al., 2017). Clients may experience the therapist as a secure base – or as unpredictably available, intrusive, or distant – depending not only on their own attachment expectations, but also on the therapist’s relational stance.

Crucially, therapists are not neutral observers. Their own attachment histories shape how they:

• Respond to emotional intensity
• Tolerate dependency
• Manage distance and closeness and
• Interpret relational cues.

Research indicates that therapist attachment security is associated with stronger alliances and more effective repair following relational ruptures (Degnan et al., 2016). Conversely, insecure therapist attachment may subtly bias clinical responses – for example, withdrawing from highly dependent clients or over-engaging with those who seek reassurance.

Therapist attachment patterns: The invisible variable

Therapist attachment patterns are rarely assessed explicitly in training or supervision, yet they function as a continuous, often implicit influence on clinical work. Rather than being static traits, these patterns reflect organised strategies for regulating closeness, affect, and relational threat, which can be activated within the therapeutic relationship itself (Mikulincer & Shaver, 2019). In this sense, therapist attachment is not simply background context – it is an active ingredient in how therapy unfolds moment to moment.

Research suggests that therapists’ attachment orientations are associated with differences in empathy, alliance formation, and responsiveness to rupture (Degnan et al., 2016; Eubanks et al., 2018). These differences are often subtle, expressed not through overt behaviour but through micro-level shifts in attention, timing, and emotional availability.

A securely attached therapist tends to demonstrate flexibility in regulating proximity and distance. They can remain emotionally present without becoming over-identified, tolerate uncertainty without rushing to resolution, and engage openly in relational repair. This capacity allows them to function as a secure base, from which clients can explore distressing material while maintaining a sense of relational safety (Mallinckrodt et al., 2017). Importantly, security here is less about perfection and more about responsiveness over time – including the ability to recognise and repair inevitable misattunements.

By contrast, therapists with anxious attachment tendencies may exhibit heightened sensitivity to signs of disconnection or dissatisfaction. This can manifest as over-attunement to the alliance, excessive reassurance-seeking (sometimes subtly enacted through clinical questioning), or difficulty tolerating client withdrawal. While this orientation often brings warmth and investment, it may also lead to over-functioning, where the therapist takes disproportionate responsibility for maintaining the relationship. In such cases, the therapist’s efforts to secure connection can inadvertently amplify pressure within the dyad, particularly for clients who experience closeness as overwhelming.

Therapists with avoidant attachment tendencies may prioritise autonomy, cognitive processing, and emotional containment. These qualities can support clear boundaries and thoughtful formulation; however, they may also be associated with deactivation strategies, such as minimising affect, shifting prematurely to problem-solving, or maintaining relational distance when emotional intensity increases (Mikulincer & Shaver, 2019). Clients with high attachment needs may experience this stance as disengagement or lack of responsiveness, even when the therapist’s intention is to remain grounded and professional.

It is important to emphasise that these patterns are not inherently problematic. Each reflects an adaptive strategy that can be clinically useful in certain contexts. The key variable is not the presence of a particular attachment tendency, but the therapist’s capacity for reflective awareness and regulation. Therapists who can recognise when their own attachment system is activated – and adjust their responses accordingly – are better positioned to offer a consistent, attuned, and reparative relational experience (Degnan et al., 2016).

Case vignette 

Helene, a clinical psychologist, notices a familiar tension arising in her work with Geoff, a client who frequently cancels sessions and appears emotionally distant when he attends. After several weeks, Helene begins to feel a creeping sense of inadequacy – wondering whether she is “failing” him.

In session, she finds herself working harder: offering more interpretations, checking in repeatedly, and subtly seeking signs of engagement. When Geoff responds minimally, Helene feels deflated and questions her effectiveness.

Upon reflection in supervision, Helene recognises a pattern. Her own anxious attachment tendencies are being activated by Geoff’s avoidant presentation. His distance amplifies her drive to pursue connection, which, in turn, may reinforce his withdrawal.

This realisation shifts her stance. Rather than intensifying her efforts, Helene begins to tolerate the relational space more deliberately, offering consistent presence without overreaching. Over time, Geoff’s engagement increases – tentatively, but meaningfully.

When patterns meet: Common therapist-client pairings 

Attachment dynamics in therapy are not simply the sum of two individual styles; they emerge through interactional patterns that are co-created and continuously reinforced (or reshaped) within the therapeutic relationship. These patterns often unfold outside of conscious awareness, expressed through subtle shifts in pacing, emotional tone, and relational positioning. Over time, they can stabilise into predictable cycles that either facilitate or constrain therapeutic progress (Mallinckrodt et al., 2017).

Research on interpersonal complementarity and attachment suggests that certain pairings tend to amplify attachment strategies, particularly under conditions of emotional activation or relational strain (Mikulincer & Shaver, 2019; Degnan et al., 2016). While these configurations are not deterministic, they offer a useful lens for anticipating and navigating common clinical challenges.

One frequently observed dynamic is the anxious therapist–avoidant client pairing, which can give rise to a classic pursue–withdraw cycle. The therapist, sensitive to cues of disconnection, may increase efforts to engage – asking more questions, offering interpretations, or leaning forward emotionally. For the avoidantly oriented client, these moves may be experienced as intrusive or overwhelming, prompting further withdrawal. The resulting pattern can leave both participants feeling ineffective: the therapist experiences rejection or inadequacy, while the client experiences pressure or loss of autonomy.

In contrast, an avoidant therapist–anxious client pairing often centres on unmet attachment needs. The client may seek reassurance, validation, or increased relational closeness, while the therapist – whether consciously or not – maintains distance through brevity, cognitive focus, or emotional restraint. From the client’s perspective, this can confirm expectations of unavailability or rejection, intensifying attachment anxiety. Meanwhile, the therapist may feel overwhelmed or subtly criticised, reinforcing their own tendency to step back.

When both therapist and client exhibit anxious attachment tendencies, the relationship can become highly engaged but also emotionally volatile. There may be strong alliance formation initially, characterised by warmth and mutual investment; however, this can shift into over-involvement or boundary diffusion, particularly when the therapist feels responsible for maintaining connection. Ruptures in this context may escalate quickly, with both parties experiencing heightened sensitivity to perceived slights or disconnection.

An avoidant–avoidant pairing, by contrast, may appear stable on the surface. Sessions can proceed smoothly, with a focus on problem-solving, insight, or behavioural strategies. However, emotional depth may remain limited, and opportunities for corrective relational experience may be missed. Both therapist and client may collude – implicitly – in maintaining a low-affect, low-risk relational space, which can be experienced as “comfortable but not transformative.”

Case vignette

Liam, a therapist who tends toward emotional restraint, is working with Sophie, a client who presents with high relational anxiety. Sophie frequently seeks reassurance: “Are we making progress?” “Do you think I’m getting better?” Liam responds with thoughtful but measured reflections, often redirecting to Sophie’s own perspective.

Over time, Sophie becomes increasingly distressed, expressing that she feels “alone” in the therapy. Liam, in turn, notices a growing sense of pressure and begins to feel that Sophie is “too dependent.” He shortens his responses and leans more heavily on structured interventions.

In supervision, Liam recognises the emerging pattern: his avoidant tendencies are being activated by Sophie’s anxious bids for connection, while Sophie’s anxiety is intensifying in response to his perceived distance. The work shifts as Liam experiments with more explicit emotional validation, while maintaining clear boundaries. Sophie begins to settle, experiencing the relationship as both responsive and containing.

Clinical pearl: Attachment pairings don’t cause problems – unexamined pairings do. Awareness creates choice.

Across these pairings, the central clinical task is not to “correct” the dynamic directly, but to recognise, tolerate, and work within it. Therapists who can track these patterns in real time are better able to avoid rigid enactments and instead offer responses that introduce flexibility into the system – often by doing something slightly unexpected, yet attuned.

For example, this might involve:

• Slowing down rather than pursuing in a pursue–withdraw cycle,
• Leaning into emotional validation when the impulse is to distance, or
• Gently introducing structure when the relationship becomes overly enmeshed.

These shifts, though small, can disrupt entrenched patterns and create space for new relational experiences. Over time, such experiences contribute to the development of earned security, both within and beyond the therapy room (Mikulincer & Shaver, 2019).

Rupture, repair and attachment activation in session 

Relational ruptures – moments of tension, disconnection, or misunderstanding – are not only inevitable in therapy; they are central mechanisms of change. From an attachment perspective, ruptures can be understood as activations of the attachment system, in which perceived threat to the relationship triggers protective strategies organised around anxiety or avoidance (Mikulincer & Shaver, 2019; Eubanks et al., 2018).

Contemporary psychotherapy research distinguishes between two broad categories of rupture: withdrawal ruptures and confrontation ruptures (Safran & Muran, 2000; Eubanks et al., 2018). These rupture styles often map closely onto attachment strategies and can be observed in subtle, moment-to-moment shifts within sessions.

Withdrawal ruptures are characterised by movement away from the therapist or the therapeutic task. Clients may become quiet, change the subject, intellectualise, or comply superficially without genuine engagement. From an attachment lens, these behaviours reflect deactivating strategies, commonly associated with avoidant attachment, where emotional distance serves to reduce perceived relational threat.

Confrontation ruptures, by contrast, involve movement against the therapist or therapy. Clients may express dissatisfaction, challenge the therapist, or reject interventions. These responses are often linked to hyperactivating strategies, associated with anxious attachment, where heightened emotional expression functions to secure attention and responsiveness.

Importantly, therapists are not immune to these processes. Their own attachment patterns can shape how ruptures are perceived and managed. An anxiously oriented therapist may experience withdrawal ruptures as personal rejection, prompting increased efforts to re-engage, while an avoidantly oriented therapist may experience confrontation ruptures as overwhelming or critical, leading to subtle distancing or defensiveness.

Dialogue snippet (withdrawal rupture):

• Client: “I don’t really know what to say today… everything’s fine, I guess.”
• Therapist: “Part of you sounds unsure about being here right now. I’m wondering if something in our work has felt difficult or off lately?”
• Client: (shrugs) “Maybe… I just didn’t want to get into it.”
• Therapist: “That makes sense – sometimes getting into it can feel like a lot. We can go at a pace that feels manageable, but I also don’t want us to miss something important between us.”

In this exchange, the therapist gently names the withdrawal without forcing engagement, maintaining connection while respecting the client’s defensive need for distance.

Dialogue snippet (confrontation rupture):

• Client: “Honestly, I don’t think this is helping. You keep saying the same things.”
• Therapist: “I’m really glad you said that. It sounds like something about how I’ve been responding hasn’t felt useful.”
• Client: “Yeah… it feels like you’re not really listening.”
• Therapist: “That’s important for me to understand. Can you help me see what I might be missing?”

Here, the therapist resists defensiveness and instead moves toward collaborative repair, signalling openness and valuing the client’s experience.

Clinical pearl: Ruptures are not interruptions to the work – they are the work, especially from an attachment perspective.

Effective repair involves more than resolving the immediate tension; it requires the therapist to provide a corrective relational experience. This often includes:

• Acknowledging the rupture explicitly,
• Validating the client’s subjective experience,
• Taking appropriate responsibility (without over-apologising or collapsing), and
• Collaboratively exploring the meaning of the interaction.

Meta-analytic evidence indicates that successful rupture repair is strongly associated with improved therapeutic outcomes across modalities (Eubanks et al., 2018). From an attachment standpoint, repair processes offer clients an opportunity to experience something fundamentally different from past relationships: that disconnection can be recognised, tolerated, and repaired without abandonment or escalation.

For therapists, this work often requires tolerating their own attachment activation. Moments of rupture can evoke feelings of inadequacy, rejection, or overwhelm – particularly when they resonate with the therapist’s own relational history. The capacity to remain reflective under these conditions – to notice one’s internal responses without acting on them reflexively – is a hallmark of secure therapeutic functioning.

In practice, this may involve pausing, slowing the interaction, and shifting from content to process awareness:

• What just happened between us?
• How did each of us experience that moment?
• What does this pattern feel like for the client, and how might it connect to their broader relational world?

These questions move the work from surface-level problem-solving into deeper relational territory, where attachment patterns can be experienced, understood, and gradually transformed.

Clinical applications: Working with the dynamic 

Recognising attachment dynamics is only the first step. The clinical task is to translate this awareness into intentional, moment-to-moment practice.

Therapist self-reflection framework 

Before or after sessions, therapists can reflect on:

1. Emotional activation. What did I feel during the session (e.g., urgency, withdrawal, irritation)? 
2. Behavioural tendencies. Did I move toward or away from the client? Did I over-function or disengage?
3. Relational meaning. What might this response reflect about my own attachment tendencies?
4. Clinical choice. What would a more secure, flexible response look like here? 

This process supports earned security in clinical practice – where awareness enables intentional shifts in relational stance (Mikulincer & Shaver, 2019).

In-session interventions 

Consider these:

• Name the process: Gently bring attention to relational patterns (“I notice it can feel hard for us to stay connected when things get intense”). 
• Pace connection: Match the client’s tolerance for closeness without abandoning attunement. 
• Model repair: Acknowledge misattunements openly and non-defensively. 
• Maintain consistency: Predictability in presence supports a sense of safety, particularly for insecurely attached clients. 

Cultural and ethical Considerations in attachment work

Attachment theory, while widely applied, is grounded in Western conceptualisations of autonomy and caregiving. Cultural variations in relational norms – such as interdependence, emotional expression, and family structure – must be considered when interpreting attachment behaviours (Talia et al., 2019).

Therapists must also remain mindful of power dynamics within the therapeutic relationship. Framing client behaviour through an attachment lens should not obscure systemic factors such as trauma, marginalisation, or socio-economic stressors.

Ethically, therapists are responsible for monitoring how their own attachment patterns influence clinical boundaries, decision-making, and responsiveness. Ongoing supervision and reflective practice are essential safeguards.

Key takeaways 

• Attachment dynamics in therapy are co-created between therapist and client, not solely driven by the client. 
• Therapist attachment patterns influence alliance formation, emotional attunement, and rupture repair. 
• Common therapist–client pairings can produce predictable relational cycles (e.g., pursue–withdraw). 
• Ruptures often signal attachment activation and provide opportunities for corrective relational experiences. 
• Therapist self-awareness is a critical clinical skill, enabling more flexible and secure responses. 
• Reflective practices can help therapists recognise and regulate their own attachment-driven tendencies. 
• Effective repair requires curiosity, openness, and tolerance of relational discomfort. 
• Cultural and systemic factors must be integrated into attachment-informed formulations. 
• Secure therapeutic relationships are built through consistency, responsiveness, and repair – not perfection. 

Conclusion 

Attachment patterns do not remain at the therapy room door. They enter quietly, shaping expectations, reactions, and relational possibilities on both sides of the dyad. When left unexamined, these patterns can constrain the therapeutic process, reinforcing familiar cycles of disconnection or over-engagement.

However, when brought into awareness, attachment dynamics become a powerful clinical resource. By recognising how our own relational histories intersect with those of our clients, we can respond with greater flexibility, intentionality, and care.

In doing so, therapy becomes more than a site of insight – it becomes a lived experience of relationship that can challenge and transform long-standing attachment patterns.

Questions therapists often ask

Q. How can I tell if my own attachment pattern is being activated in session?

A. One of the most reliable indicators is a shift in your internal state that feels disproportionate or persistent. You might notice a sense of urgency (“I need to fix this”), withdrawal (“I don’t want to go there”), or self-doubt (“I’m not doing this well”). These responses often emerge alongside subtle behavioural changes – working harder, pulling back, over-explaining, or avoiding certain topics.

Rather than viewing these reactions as problems, it can be helpful to treat them as clinical data. Ask yourself: What is being evoked in me right now, and how might that relate to what the client is bringing? Over time, this reflective stance allows you to distinguish between your own attachment activation and the client’s relational needs, creating space for more intentional responding.

Q. What should I do if I notice a pursue–withdraw pattern developing with a client?

A. The instinct to “correct” the pattern directly is understandable – but often counterproductive. In a pursue–withdraw dynamic, increasing effort (on the therapist’s side) can intensify withdrawal (on the client’s side).

A more effective approach is to introduce relational flexibility. This might mean:

• Slowing the pace rather than pursuing,
• Tolerating moments of silence or distance,
• And gently naming the pattern in a non-judgmental way.

For example: “I notice when things feel a bit closer between us, it can get harder to stay in the conversation. I wonder what that’s like for you?”

This kind of intervention reduces pressure while maintaining connection, allowing the client to engage without feeling overwhelmed.

Q. How do I balance being emotionally attuned without becoming over-involved?

A. This is a central tension in attachment-informed work. Emotional attunement requires presence, responsiveness, and openness – but not fusion. The distinction often lies in whether your responses are guided by the client’s needs or by your own discomfort with disconnection.

A helpful anchor is the idea of “regulated availability.” This involves staying emotionally present while maintaining awareness of your own internal boundaries. If you notice yourself over-functioning – doing more work than the client, seeking reassurance, or struggling to step back – it may signal that your own attachment system is becoming over-activated.

Supervision and reflective practice are key here, helping you recalibrate toward a stance that is engaged but not enmeshed.

Q. What if a client experiences me as distant or unavailable, even when I feel present?

A. This is a common and clinically significant moment. From an attachment perspective, the client’s experience is not simply a misperception – it reflects their internal working model of relationships, which shapes how they interpret your behaviour.

Rather than correcting or defending your intent, the therapeutic opportunity lies in exploring the experience itself. You might respond with curiosity:

“It sounds like something about how I’ve been with you has felt distant. I’d really like to understand that better – can you tell me more about when you’ve felt that?”

This approach validates the client’s reality while opening space to examine the interaction together. Over time, these conversations can become powerful sites of corrective relational experience, where the client learns that perceived distance can be named, explored, and repaired.

Q. Do I need to resolve my own attachment issues to work effectively with clients?

A. Not at all. In fact, expecting complete “resolution” sets an unrealistic standard. What matters far more is your capacity for awareness, reflection, and regulation.

Therapists inevitably bring their own attachment histories into the room. The goal is not to eliminate these influences, but to develop what is sometimes described as “earned security” – the ability to notice when your attachment system is activated and to respond with flexibility rather than reflex.

In this sense, your own relational patterns can become a resource rather than a liability, enhancing your empathy, attunement, and capacity to engage meaningfully with clients’ experiences.

References

• Bowlby, J. (1969). Attachment and loss: Vol. 1. Attachment. Basic Books. 
• Degnan, K. A., Seymour-Hyde, A., Harris, A., & Berry, K. (2016). The role of therapist attachment in alliance and outcome: A systematic literature review. Clinical Psychology & Psychotherapy, 23(1), 47–65. PMID: 25445258 DOI: 10.1002/cpp.1937    
• Eubanks, C. F., Muran, J. C., & Safran, J. D. (2018). Alliance rupture repair: A meta-analysis. Psychotherapy, 55(4), 508–519. https://doi.org/10.1037/pst0000185
• Mallinckrodt, B., Anderson, M. Z., Choi, G., Levy, K. N., Petrowski, K., Sauer, E. M., Tishby, O., & Wiseman, H. (2017). Pseudosecure vs. individuated-secure client attachment to therapist: Implications for therapy process and outcome. Psychotherapy Research, 27(6), 677–691.
  https://doi.org/10.1080/10503307.2016.1152411
• Mikulincer, M., & Shaver, P. R. (2017). Attachment in adulthood: Structure, dynamics, and change (2nd ed.). Guilford Press.  ISBN 9781462533817
• Mikulincer, M., & Shaver, P. R. (2019). Attachment orientations and emotion regulation. Current Opinion in Psychology, 25, 6–10. https://doi.org/10.1016/j.copsyc.2018.02.006
• Safran, J. D., & Muran, J. C. (2000). Negotiating the therapeutic alliance: A relational treatment guide. Guilford Press.
• Talia, A., Taubner, S., & Miller-Bottome, M. (2019). Advances in research on attachment-related psychotherapy processes: Seven teaching points for trainees and supervisors.
  Research in Psychotherapy: Psychopathology, Process and Outcome, 22(3).
  https://doi.org/10.4081/ripppo.2019.405   PMCID: PMC7451314  PMID: 32913812.