EMDR Therapy Quarterly

It’s taken me some months to put the following together, but as the Association’s 2025 conference fades into the collective EMDR UK subconscious, and as our community prepares to meet in Bristol in March 2026, I would like to return to Matt Wesson’s 2025 Liverpool address – Why therapists are not delivering the EMDR evidence base and how to put it right – on the dangers of model drift.

His argument was uncompromising: EMDR’s credibility depends on fidelity to the research base. When clinicians drift from the procedures that generated our evidence, he warned, we risk diluting outcomes, confusing commissioners and undermining public trust.

In the subsequent issue of EMDR Therapy Quarterly, Jessica Woolliscroft’s thoughtful commentary underscored that same concern. She wrote with characteristic precision about therapist drift, the limits of good intentions and the need for clear supervision structures that ensure therapists ‘do what it says on the tin.’

Both voices – Wesson’s from the podium and Woolliscroft’s from the editorial page – speak to something essential: a shared wish to preserve what makes EMDR distinct, powerful and testable.

From that foundation, I would like to add a complementary perspective – one that shares their respect for fidelity but frames it through a clinical and relational lens. My aim is not to soften the evidence-based message but to ground it in the reality of what actually happens in the therapy room.

---

1)        Fidelity as mechanism, not script

Over the past two decades of teaching and practising what I have come to call attachment-informed EMDR (ai-EMDR) – read on for some thoughts on that term – I’d like to invite our community to see fidelity not primarily as a choreography of language and steps but as a fidelity to ‘mechanism.’

The Adaptive Information Processing (AIP) model remains our compass. We are helping the brain integrate fragmented experience so that memory networks, once isolated, become linked and flexible, and our clients can access a fuller range of healthy adult responses to the challenges of life.

The research base demonstrates that, provided the right formative material from the past is activated in the right state of mind, bilateral stimulation (of all kinds, not exclusively eye movements) facilitates this process (Shapiro, 2018). Everything else – the phrasing, metaphors and micro-sequencing – is there to serve that aim.

When fidelity is understood at the level of ‘mechanism,’ flexibility stops being a threat and becomes a precision tool. The question is never “Did I stick to the correct words and sequence word-for-word?” or “Did I stay out of the way as I’m taught?” But, “Did my intervention engage the correct network, maintain dual attention and permit adaptive resolution?”

If yes, we are still well within the evidence base.

---

2)        Starting in the present, bridging to the past

In EMDR with an explicitly attachment-informed perspective, wherever things need to go as the hour unfolds, the therapist begins each session in the client’s present. Good therapy is, after all, about changing how the person lives now: the past is only relevant where it’s compromising the present.

Of course, there are the bad things that have been done to and experienced by our clients which need to be addressed directly. This is the original heritage of EMDR, a model designed to identify and process discrete, identifiable memories of specific trauma that continue to intrude into the present. For these, the Standard Protocol as basically trained remains a reliable route to symptom reduction.

Yet we also know that many, indeed most, clients do not process memories cleanly unless the deeper attachment and developmental context that maintain dysregulation is also addressed, a pattern widely described in the trauma and attachment literature (Liotti, 2004; Solomon & Shapiro, 2008; van der Kolk, 2014). In an EMDR that works from an explicitly attachment-informed perspective, we’re always ready to bridge – through imagery, affect or spontaneous association – to the developmental roots of the current state. The landing memory is then processed as far as we can manage in that session, using what are in actual practice standard desensitisation and installation procedures, before we come ‘back across the bridge’ to map whatever shift has happened against the client’s current experience.

I know some colleagues make an emphatic case in their own trainings that bridging and ‘floatback’ (a term that I know I am not alone in finding problematic) are a distraction. I respectfully and very solidly disagree. Rather than deviating from the standard protocol, this, in my view and based on my experience, often actually tightens fidelity to the AIP model. By following the natural associative path of the client’s awareness, we reach the necessary targets faster and more efficiently. ‘Bridging’ (a better term than ‘floatback’) in this conscious, careful, appropriate and structured way keeps the work specific, embodied and anchored in the client’s lived reality, not in a spreadsheet of past events that may or may not speak to what they’ve come into therapy to change.

---

3)        The legitimate fear of drift

I share Wesson’s concern that many therapists, especially newer ones, drift away from trauma activation into relationship-only work or endless resourcing, or rush to add new protocols to their portfolio when they find, as they so often do with client complexity, that standard EMDR isn’t working for them.

The reasons are understandable: feeling stuck with the rigidities of scripts, fear of destabilising clients, limited supervision, and, for colleagues used to delivering talk therapy, the seductive comfort of supportive dialogue. But EMDR’s research strength lies precisely in its capacity safely to identify, activate and resolve a client’s most intense embedded narrative. If we avoid going there, or are centrally informed by a fear of ‘getting it wrong,’ EMDR therapy will fail.

---

4)        Defining fidelity at the level of phases

The eight phases (Harry Potter And Dumbledore In Big Cheesy Relationship, as Derek Farrell helpfully suggests these might be remembered, starting with H for history-taking through to R for Re-evaluation, and a framework every EMDR therapist should know by heart) are not ornamental; they are the very architecture of EMDR. Yet they are also spacious. Within each, there is room for informed adaptation that honours EMDR’s core understanding of the role of Adaptive Information Processing.

In ai-EMDR, we work with six practical principles that map neatly onto those eight phases:

1. Clear, attachment-informed case conceptualisation. More than history taking; we locate the wound in its current relational context.
2. Rich resourcing and preparation. Phase 2 as both safety and imaginal preparation for rewiring the dysfunctional past ‘as if for real.’
3. Target identification through developmental bridging. Finding, not assuming the core wound, often embedded in formative experience rather than obvious ‘trauma.’
4. Activation that privileges emotion and body over numbers. We orient in phase 3 to felt sense, with understandings of self that emerge at an emotional and embodied level.
5. Rewiring the past, not just desensitising it. Rather than primarily staying out of the way and trusting (hoping?) that AIP will do its thing, when attachment-informed we work closely – intensely, creatively – with the client’s present adult self (dual attention after all) to reveal, rewire and repair their formative developmental past.
6. A complete session arc. From present to past and back again, we end every session in re-evaluation and embodied, future-relevant change.

If an external observer – or video supervisor – can see those principles alive in the session, I would argue that fidelity is being upheld.

It may be more constructive to view attachment-informed EMDR not as a departure from the standard protocol but as its natural evolution – extending the same mechanisms of AIP into the developmental and relational domains that often give rise to complex presentations. In that sense, ai-EMDR is a close ally, a next-generation articulation of the same principles rather than a competing model.

---

5)        Supervision as both mirror and microscope

I share Matt’s and Jessica’s worry about supervisees who drift into relationship-only or muddily structured EMDR therapy, or avoid trauma activation for fear of dissociation. The antidote is to help therapists feel confident in appropriately and safely adjusting how they use EMDR’s standard procedural steps, not to make them feel bad or judged. What matters is target-selection logic, creative engagement with a client’s root formative experiences, and a firm focus on what needs to shift to render the present and the future different.

If someone uses bridging, it’s entirely appropriate to ask them for their decision tree. Why and how did they go with that starting point? Where did they land? How did they work with the client’s emerging narratives? Did the present trigger feel usefully different when they ‘returned to target’ in the here and now? And yes, a supervisee’s verbal account is very often insufficient. Video makes things transparent.

---

6)        Researching the flexible edges

If NICE and other regulators hesitate to broaden EMDR’s endorsement, one reason is the paucity of outcome research on the creative yet disciplined adaptations already happening in practice. Rather than shaming flexibility and disciplining those newly trained (in a mere seven days after all) not to stray from the basics of their initial, brief introduction to this therapy, we might instead distinguish flexibility that clarifies from flexibility that avoids.

Some obvious research questions arise:

• Does a present-anchored bridging strategy reduce overall session numbers compared with ‘start with worst trauma’ sequencing?
• Which interweaves are most effective at unsticking blocked processing, and under what conditions do they achieve this?
• How does fidelity to a phase structure correlate with relational attunement, therapist presence and dropout rates?
• To what extent is EMDR compromised, or not, by adjusting the sequence of questions in phase 3 and by leaving out, at this stage, the PC?

These are empirically testable within standard designs. Generating such data would not weaken EMDR’s scientific base; it would deepen and future-proof it.

---

7)        Beyond procedures: the culture of practice

Underneath the technical debate lies something cultural. Fidelity cannot thrive in an atmosphere of anxiety or moralism. The moment therapists fear censure for adapting EMDR appropriately to the story of the individual client, the moment they fear sharing in supervision what they’re actually doing in therapy, the community stops learning and EMDR’s original spirit – Shapiro’s relentless curiosity – risks ossifying into orthodoxy.

A healthy professional culture should distinguish between debate and dissent. It should welcome attachment-informed perspectives as expansions, not heresies. It should also make explicit that the goal is client transformation, not therapist compliance.

In my own teaching, I remind colleagues that EMDR was born as a challenge to orthodoxy and as a creative leap embedded in the evolutionary dynamics of our survival as a species. The task now is to carry that spirit forward without losing scientific discipline.

---

8)        The heart of the matter: what clients need

Clients rarely arrive asking for ‘protocol fidelity.’ They come asking for relief, coherence and reconnection. For them, the therapeutic experience that matters is not whether the therapist counted sets exactly or read the assessment script verbatim, but whether the work reached the embodied memory that drives their suffering – and whether they can now live differently.

Fidelity, then, must ultimately be judged by outcomes that matter to clients: reduced re-experiencing, restored capacity for relationships, increased present-moment safety. When attachment wounds are fully integrated, these outcomes emerge naturally. If we can show that attachment-informed targeting achieves those results efficiently and safely, we are being faithful to the AIP model’s essence.

---

9)         The two AIs: where attachment meets intelligence

There remains, of course, a challenge in the very term ‘ai-EMDR.’ The double meaning of the two letters A and I, AI/ai – ‘attachment-informed’ on the one hand, and ‘artificial intelligence’ on the other – risks confusion, of course, but also captures something of the tension within our profession itself between clinical depth and technological transformation. As I noted in my earlier ETQ article (Brayne, 2024), even ‘attachment-informed’, written in lowercase to differentiate, can suggest yet another protocol rather than a disciplined and relationally attuned way of applying the standard one.

We may eventually need a new name for what I would argue is, at heart, nothing more (and nothing less) than a fuller, systemically informed use of EMDR as originally conceived – rooted in the same evidence base, faithful to AIP – but drawing consciously on developmental science and relational understanding to enhance effectiveness. The point is not to brand another variant but to describe a way of practising EMDR that makes full use of the model’s reach.

At the same time, the other ‘AI’ – artificial intelligence – is reshaping how many of us think, write and even conduct our professional lives. In my own work, this technology has become an unexpected reflective partner: a way of testing ideas, refining language, and exploring clinical nuance with speed and precision that once took months of peer dialogue. Used well, it extends rather than replaces human judgement, offering mirrors that deepen rather than dilute our understanding.

The intersection between these two AIs – the attachment-informed and the artificial intelligence – may yet prove one of the most creative frontiers for EMDR in the years ahead, provided we stay grounded in what truly matters: real clients, real change and the disciplined curiosity that has always defined our field.

10)  Common ground

To conclude, there is far more common ground than division between the positions voiced in Liverpool and those of clinicians exploring attachment-informed adaptations. We all want EMDR to remain evidence-based, respected and effective.

The path forward may lie in a shared redefinition of fidelity – not as obedience to a single procedural dialect, but as disciplined allegiance to the mechanisms that make EMDR work. That means clearer training standards, more video supervision and less premature eclecticism – exactly as Wesson argues.

It also means encouraging skilled flexibility, contextual sensitivity and relational depth – the elements that keep EMDR human, as the best research consistently shows. If we can hold both sides of that equation – discipline and imagination, evidence and attunement – EMDR will continue to grow as a living, learning psychotherapy rather than a closed system. That, surely, is fidelity in its truest form.