How does EMDR really work?
Healthcare Counselling and Psychotherapy Journal, April 2026

Dr Yvonne Waft describes how bilateral stimulation can help people to reprocess painful memories and recover from trauma

Eye movement desensitisation and reprocessing (EMDR) therapy was developed in the USA in 1987 by Francine Shapiro, a former English teacher turned PhD psychology student, who was struggling with some personal health challenges.

Shapiro was walking in a park, thinking about a recent cancer diagnosis, when she noticed that her eyes were scanning the park from side to side, a movement that reduced her stress and anxiety. Shapiro investigated further, applying bilateral eye movements to volunteers, while they recalled traumatic experiences. By 1989, she had published the first research paper on the use of eye movement desensitisation (EMD) to treat post-traumatic stress disorder (PTSD).¹

Initially, Shapiro believed the eye movements simply helped to desensitise a memory. However, further research quickly showed that cognitive reappraisals of the memories were taking place, so they were being fully reprocessed. EMD thus became EMDR.

Shapiro put forward an adaptive information processing (AIP) model, proposing that trauma symptoms were caused by maladaptively stored memories, which created disturbance in our emotional, cognitive and physical experience.

Typically, when difficult things happen, we think about them, talk about them, dream about them and heal from them. It’s a natural process that Shapiro termed AIP. However, when events are overwhelming, this process is disrupted and the memory is stored in a fragmented form. The thoughts, images, emotions and sensations then remain as vivid and disturbing as they were when the event originally occurred. Maladaptively stored memories need some help in order for the normal healing process to take place.

With careful preparation and an attuned therapist, the bilateral eye movements in EMDR seem to accelerate the processing of the maladaptive memory. The person then no longer experiences disturbance when recalling the event. They also no longer suffer from flashbacks, nightmares or mood disturbances when new events trigger the old memory into awareness.

Practice-based evidence over many years has shown that other forms of bilateral stimulation, such as bilateral tapping and bilateral sounds, can also be effective, especially for those with visual impairment or eye pain.

The evidence

There has been considerable controversy and mistrust around EMDR, which continues in some quarters to this day. However, we now have more than 35 years of research evidence to support the view that EMDR is a valid, effective approach to treating not only PTSD but also a much wider range of mental health conditions, including depression,² anxiety disorders³ and phantom limb pain.⁴

EMDR has also proved as effective as trauma-focused cognitive behavioural therapy (TF-CBT) and prolonged exposure therapy (PET), often in a shorter time frame. A Cochrane Review in 2013⁵ found EMDR and TF-CBT to both be effective treatments for PTSD, and that results held up at four-month follow-up.

A systematic review in 2024 also reported overall support for EMDR in the treatment of PTSD, although it called for further research to understand the mechanism of action and use in wider populations.⁶ A further meta-analysis, published in December 2025, indicated strong support for EMDR in the treatment of adults with a history of childhood trauma.⁷

EMDR is also recommended as a treatment of choice for trauma by many international organisations, such as the World Health Organisation, the National Institute of Health and Care Excellence, the US Department of Veterans Affairs, the US Department of Defence and the American Psychological Association.

The use of EMDR has been popularised by celebrities and royalty, such as Prince Harry, the actress Jameela Jamil and the journalist Natasha Kaplinsky. The celebrity endorsement has raised awareness of EMDR – and the demand for EMDR therapists and supervisors is growing.

As a result of her research, Shapiro devised a detailed eight-phase EMDR protocol. The eight phases should be followed carefully, sticking closely to the manual wherever possible. Each phase is, to some degree, scripted, based on Shapiro’s research. So close adherence to the script is encouraged. There is, however, some flexibility within the model, and experienced clinicians often find their own ways of adapting to their clients’ individual needs.

Phase one (history taking) is a trauma-focused history taking. It is slightly different from a typical intake assessment, as it is looking at preparedness for processing, contraindications and number/order of trauma targets to be processed. It would normally take place after the initial intake assessment and once a decision has been made that a trauma-processing approach is required.

Phase two (preparation) expands on the history taking to identify any skills or resources required to support the person through trauma processing. For example, if there is social deprivation, we might want to focus first on encouraging social engagement or accessing statutory support. In addition, we can help to install inner resources, such as an imagined nurturing figure or safe place. We need the person to feel robust enough to bring up difficult material in the processing phase, without going into crisis. We also give general psychoeducation about trauma and detailed information about the EMDR process.

The calm or safe place exercise is not specific to EMDR. We ask the person to bring to mind somewhere real or imagined, where they can feel relaxed and at ease with themselves. We also ask for a sensory description of the place, and the feelings in their body as they imagine it. In EMDR, we then use slow bilateral eye movements or taps on the body to enhance the feeling of relaxation.

Phase three (target assessment) assesses the first target for processing, breaking it down into five elements:

an image of the worst part of the memory
a negative belief about the self that comes up when the person thinks of the memory
a positive belief they would prefer to hold about themselves
the emotions elicited when they think of the memory
where they are feeling this in the body right now.

We ask the person to rate the level of distress elicited in the here and now when they think about the memory, using a subjective units of disturbance (SUD) scale of 0 to 10, where 0 is no distress or neutral and 10 is the worst distress imaginable. We also ask them to rate their preferred, positive belief on a scale of 1 to 7, where 1 is not true at all and 7 is completely true, known as the validity of cognition (VOC) scale.

In phase four (desensitisation), we remind the client of the elements in phase three. We then ask them to let the memory go wherever it needs to go, as we guide them to apply bilateral stimulation (BLS), either with side-to-side eye movements, bilateral taps on the body or sounds in each ear alternately.

We check in with them after each set of BLS and ask what they are noticing, reminding them there is no right or wrong, while we resume the BLS. Typically, the disturbance will reach a peak, the person will experience a range of thoughts, feelings, images and body sensations as the memory processes, and they will gradually begin to connect with more positive imagery, thoughts and emotions. Once the SUD rating reaches 0, we move on to phase five.

Phase five (installation) is when we install the positive belief about the self, with the target memory in mind and continuing BLS. Once a VOC score of 7/7 is reached, we can move on to phase six.

Phase six (body scan) involves holding the target memory in mind and checking for any residual disturbance in the body. If there is any unusual tension, tightness or tingling, we apply more BLS until it resolves.

Phase seven (closure) involves a specific process for closing both complete and incomplete processing, to ensure safety and grounding before the person leaves the session.

In phase eight, we evaluate in each session how the target memory has changed, or not, over the week. We then either resume processing the target memory, if incomplete, or we move on to the next target memory.

With complex cases, phases one and two can be very lengthy, lasting months rather than weeks. Good supervision is essential to ensure safety and containment. Nonetheless, in my experience, EMDR typically resolves even complex cases far quicker than other models of therapy.

Accredited training and ongoing accredited supervision are essential in order to deliver EMDR safely and effectively. To find out more, I recommend the EMDR Association UK website.

Peter is sent to therapy by his case manager, as a result of a personal injury claim following a road traffic accident. Peter is experiencing nightmares, ongoing pain from his physical injuries and tries to avoid driving. He has waves of severe anxiety when he is in the car, either as a driver or passenger, and is struggling to attend his medical appointments. He would like to return to work, but cannot get there due to his travel anxiety.

Phase one. The initial assessment shows no previous trauma of note. Peter had a relatively stable upbringing, with loving and supportive parents. He has a wife and two young children, who are helping him through his recovery. Peter enjoys keeping fit and has been able to return to the gym and the pool. He enjoys art and music, which he finds soothing. Peter is not a big drinker and prefers not to take strong medication for pain. He reports no risk issues.

Phase two. The therapist feels Peter is an ideal candidate for EMDR, as there are no contraindications. Some psychoeducation and explanation of EMDR are given. Peter is taught some relaxation strategies, including a ‘calm place’ exercise and some breathing/mindfulness techniques.

Phase three. A target memory – the accident – is identified. Peter was driving to work and had stopped at a roundabout when a car hit him from behind at speed, pushing his car out onto the roundabout, where he was hit from the side by another vehicle. He suffered a broken arm, whiplash and cuts and bruises. Peter identified the worst image as the sight of the second car as it hit his own car. The negative belief was, ‘I am going to die’; the positive belief was, ‘I’m alive’ (rated 3/7 initially, with the image in mind). The emotion was terror and it was felt as tension throughout the body, with an initial subjective units of disturbance (SUD) score of 10/10.

Phase four. Peter used eye movements, following the therapist’s fingers moving rapidly back and forth across his field of vision. As the eye movements began, Peter tensed and his breathing became very rapid. After the first set of eye movements, Peter told the therapist that he was scared. After another set, he said: ‘I can see the car’. With the third set came the sound of the impact. Then, ‘Ouch, my arm really hurts!’. Peter then felt fear and anger, and he began to cry.

The therapist continued with the eye movements. Gradually, after about 20 minutes of processing, Peter said: ‘The paramedics came. I’m safe. I’m going to be OK. I’m calmer now.’ His breathing became noticeably slower and more regular, and he sank back into the chair. As Peter grew calmer, the therapist checked back in with the target memory and asked for a SUD rating. Peter gave the memory 3/10, adding that he was still angry about the accident and felt tension in his jaw. Some further sets of bilateral stimulation (BLS) cleared the anger and tension. Peter reported a SUD score of 0/10 when he next returned to the target memory.

Phase five. When asked about his positive belief, Peter’s validity of cognition (VOC) rating was 5/7. So, further sets of BLS were carried out. Within a few sets, the VOC score was 7/7.

Phase six. A body scan, while holding the memory in mind, revealed a little ongoing tension in Peter’s jaw. After three more sets of BLS, the tension was gone. Peter said: ‘I can hardly even see the memory now. It’s like a grainy old black and white movie.’ He had no disturbance left and felt very positive.

Phase seven. The therapist warned Peter that the memory processing might continue after the session, and he should keep a note of anything that came up, good or bad, over the next week. The session closed with a relaxation exercise.

Phase eight. On re-evaluation the following week, Peter reported that he was feeling really well, that the accident no longer troubled him and that he had tentatively been out in the car, with minimal anxiety. He was keen to drive more, return to work and maybe take the family on holiday. He could not believe how easily the memory had resolved.

People only rarely come to therapy with such a straightforward presentation as the fictitious Peter. But we can adapt the model when things are more complex, perhaps due to early childhood trauma, dissociation, multiple targets, neurodivergence or learning disability.

1 Shapiro F. Efficacy of the eye movement desensitisation procedure in the treatment of traumatic memories. Journal of Traumatic Stress 1989; 2(2): 199–223.
2 Sepehry AA, Lam K, Sheppard M, Guirguis-Younger M, Maglio AS. EMDR for depression: a metaanalysis and systematic review. Journal of EMDR Practice and Research 2021; 15(1): 2–17.
3 Yunitri N, Kai CC, Chu H et al. The effectiveness of eye movement desensitization and reprocessing toward anxiety disorder: a meta-analysis of randomized controlled trials. Journal of Psychiatric Research 2020; 123: 102–113.
4 Schneider J, Hofmann A, Rost C, Shapiro F. EMDR in the treatment of chronic phantom limb pain. Pain Medicine 2008; 9(1): 76–82.
5 Bisson JI, Roberts NP, Andrew M, Cooper R, Lewis C. Psychological therapies for chronic post-traumatic stress disorder (PTSD) in adults. https://tinyurl.com/2mwwfm4x (accessed February 2026).
6 Vereeken S, Corso G. Revisiting eye movement desensitisation and reprocessing therapy for post-traumatic stress disorder: a systematic review and discussion of the American Psychological Association’s 2017 recommendations. Cureus 2024; 16(4): e58767. https://tinyurl.com/ yc75x8fv (accessed February 2026).
7 Ajele JW, Idemudia ES. Long-term effects of eye movement desensitization and reprocessing on childhood trauma: a cross-national meta-analysis of adult well-being. Journal of EMDR Practice and Research 2025; 19.