TIRCIA · EMDR
The Adaptive Information Processing (AIP) model
How EMDR explains healing from trauma, and what the science says about it
By Marie-Agnès Thulliez · Author · Publishing Director13-minute read
Why does a traumatic memory remain so vivid, years later, when most memories fade? And why, after successful EMDR therapy, can a person think back to it without reliving the fear? To answer these questions, Francine Shapiro proposed an explanatory framework: the Adaptive Information Processing model. In English it is known as the AIP model; in French, it is called the TAI model (traitement adaptatif de l’information). Both acronyms refer to the same thing. This model guides the whole of EMDR practice. What exactly does it say, where does it come from, and how does it stand up to research?
In brief
- According to the AIP model, our brain naturally “digests” difficult experiences. Trauma can block this work: the memory remains stored as it was, with its original images, emotions and sensations.
- These poorly integrated memories are thought to underlie many current symptoms. EMDR seeks to restart their processing.
- The effectiveness of EMDR in post-traumatic stress disorder (PTSD) is well established and recognised by the major international guidelines.
- The AIP model itself remains a clinical hypothesis: it converges with other models of traumatic memory, but it has yet to be demonstrated as such. Working memory is currently the best-supported explanation for the effect of eye movements.
1. Where does the AIP model come from?
In 1987, Francine Shapiro, then a psychology student, noticed that disturbing thoughts lost some of their intensity when her eyes moved rapidly back and forth. From this she developed a procedure, initially called EMD (Eye Movement Desensitization), which she tested in a first study published in 1989 with people suffering from traumatic memories (Shapiro, 1989).
She soon observed that patients did more than “desensitise” the memory: new associations appeared, the way they saw themselves changed, and insights emerged spontaneously. She then added the “R” for Reprocessing and, in the 1990s, proposed a model to account for these observations, first called “Accelerated Information Processing”, then “Adaptive Information Processing”. This is the AIP model, presented and refined in the successive editions of her reference work (Shapiro, 2018).
This origin is worth bearing in mind: the AIP model was born of clinical observation, to make sense of what was being seen in sessions. It was not derived from laboratory experiments.
2. What the model says
The model starts from a simple idea: the brain has an information processing system that links each new experience to existing memories. These memories are organised into memory networks, that is, sets of images, thoughts, emotions and sensations that are associated with one another. An argument, a disappointment or a passing fear thus eventually finds its place: we keep what is useful from it, and the emotion settles.
When an event is too intense, this system is thought to be overwhelmed. The memory would then be stored in isolation, in a form close to that of the moment it was lived, with its original perceptions, emotions and bodily sensations. It would remain unconnected to the information that would allow it to be put into perspective: “it’s over”, “I survived”, “it wasn’t my fault”.
According to Shapiro, these dysfunctionally stored memories underlie a large share of current symptoms: intrusive images, negative beliefs about oneself (“I’m in danger”, “I’m worthless”), disproportionate emotional reactions, bodily sensations. Something in the present, a noise, a smell, a tone of voice, would reactivate the old network, and the person would react as if the past were still there.
Big T and small t
- “Big T” traumas: traumatic events as defined by medical classifications (assault, accidents, disasters, violence).
- “Small t” traumas: more ordinary experiences that are nonetheless felt as hurtful or humiliating (repeated mockery, rejection, failures, emotional neglect), which can also leave unintegrated memories.
This distinction, specific to the EMDR tradition, is not an official classification. It does, however, echo a research finding: in a general population study in the Netherlands, non-traumatic life events, such as divorce or job loss, led to as many post-traumatic-type symptoms as traumatic events, or even more (Mol et al., 2005).
3. What it predicts in clinical practice
| Without reprocessing | After reprocessing | |
|---|---|---|
| The memory | Remains “frozen”, vivid, charged with emotion and sensations | Keeps its content, but loses its emotional and sensory charge |
| The symptoms | Re-experiencing, nightmares, hypervigilance, avoidance | Diminish or disappear |
| The belief about oneself | “I’m still in danger”, “it’s my fault” | “It’s over, I’m safe now”, “I did what I could” |
| The lived experience | The patient knows the danger has passed, without feeling it | Knowing and feeling come together |
The model has an important practical consequence: it invites us to look, behind a current symptom, for the memories that feed it. This is what structures the so-called “three-pronged” protocol: we process the past memories at the root of the difficulties, then the present triggers, and finally prepare for future situations using an imagined scenario (Shapiro, 2018).
4. The eight-phase protocol
The AIP model translates into a standardised eight-phase protocol. Reprocessing memories is only one part of it:
| Phase | What happens |
|---|---|
| 1. History taking and treatment planning | The patient’s history, current symptoms, choice of memories to target. |
| 2. Preparation | Explaining the method, building the therapeutic alliance, emotional regulation tools. |
| 3. Assessment | The patient identifies the image of the memory, the associated negative belief, the desired positive belief, the emotions, the level of disturbance (SUD, from 0 to 10) and the bodily sensations. |
| 4. Desensitisation | The patient holds the memory in mind during short sets of bilateral stimulation (eye movements, alternating sounds or taps); between sets, they simply notice what emerges. |
| 5. Installation | Once the disturbance has subsided, ideally to 0, the positive belief is strengthened. |
| 6. Body scan | The patient checks whether any tension linked to the memory remains, which is reprocessed if needed. |
| 7. Closure | The session ends in a stable state, even if reprocessing is incomplete. |
| 8. Re-evaluation | At the next session, the therapist checks what has changed and what remains to be processed. |
On the role of the preparation phase, particularly in complex trauma, see our article on stabilisation.
5. A model that echoes other theories of traumatic memory
The AIP model is in line with other work. Other models, drawn from cognitive psychology and tested experimentally, also describe traumatic memory as a poorly integrated memory:
- Ehlers and Clark’s cognitive model: the traumatic memory is thought to be poorly elaborated and poorly linked to its context, which explains why it can be reactivated by cues in the present and experienced as a current threat (Ehlers and Clark, 2000).
- Brewin’s dual representation theory: re-experiencing is thought to stem from sensory representations of the memory that are insufficiently linked to a contextualised representation, situated in time and space (Brewin et al., 2010).
These models use a different vocabulary, but they converge with AIP on the essentials: the problem is thought to lie in the way the memory is stored and linked to the rest of memory, rather than in the memory itself. Authors close to EMDR have proposed bringing AIP more explicitly into line with this work and with the neuroscience of memory (Hase et al., 2017). This convergence makes the model plausible; on its own, it falls short of demonstrating it.
6. Why does EMDR work? The competing explanations
The AIP model describes what changes, but it leaves open precisely how. Several explanations have been proposed, and a systematic review of 87 studies examined them (Landin-Romero et al., 2018).
Working memory: the best-supported explanation
Holding a memory in mind while following an eye movement places heavy demands on working memory, whose capacity is limited. The memory then becomes less vivid and less emotionally charged, and is thought to be stored again in this attenuated form (van den Hout and Engelhard, 2012). This hypothesis has the advantage of being testable in the laboratory, and it has been confirmed by numerous experiments.
Do eye movements add anything?
This question has been debated since the early days of EMDR, and the answers have evolved. An early meta-analysis found no additional effect of eye movements compared with the same procedure without them (Davidson and Parker, 2001). A more recent meta-analysis, covering more studies, concludes on the contrary that the movements bring a moderate additional effect, in clinical trials as well as in laboratory studies (Lee and Cuijpers, 2013).
Neurobiological avenues
Imaging studies suggest changes in brain activity before and after EMDR therapy, particularly in regions linked to fear and its regulation. Another avenue involves reconsolidation: in animals, a reactivated memory becomes temporarily modifiable before being stored again (Nader et al., 2000). EMDR may take advantage of this window. But these findings rest on small samples or on extrapolations from animal research, and the authors of the review consider research on the mechanisms of EMDR to be “still in its infancy”.
Clinical vignette
A fictional case, drawn from common clinical situations.
Marc, 52, seeks help for anxiety attacks before every meeting at which he has to speak. He has experienced no “big” trauma. During history taking, the therapist invites him to let the earliest memory associated with this feeling of a tight throat come to mind. He recalls a scene from when he was ten: a teacher made him read aloud and the whole class laughed. Associated negative belief: “I’m ridiculous”. Disturbance: 8 out of 10.
Over the sets of bilateral stimulation, other images arise, then the memory transforms: Marc sees “a little boy who had stage fright, like everyone else”. The disturbance falls to 0, and the belief “I can speak, even when I’m moved” is installed. The following sessions address current triggers, then an imagined meeting scenario. In the reading offered by the AIP model, an isolated “small t” memory was feeding his present-day anxiety.
7. What we know, and what we do not yet know
What is established
- EMDR is effective in post-traumatic stress disorder (PTSD). It is recommended by the WHO (2013), NICE (2018) and the ISTSS (2019); the APA “suggests” it in its updated 2025 guideline. In France, Inserm’s collective expert review had already included it among the validated approaches for PTSD (Inserm, 2004).
- It is one of several such treatments. The ISTSS guidelines place other trauma-focused therapies at the same level: cognitive therapy, cognitive processing therapy, prolonged exposure, trauma-focused CBT.
What remains to be established
- For other conditions, such as phobias, anxiety, chronic pain or addictions, a meta-analysis of 76 trials concludes that the evidence is still insufficient to recommend EMDR (Cuijpers et al., 2020).
- Even in PTSD, these authors point out that few trials are of good methodological quality, which calls for caution.
- The AIP model itself has not been directly validated: the effectiveness of EMDR does not prove that its explanation is the right one. This is in fact the position of its authors, who speak of “potential mechanisms” (Solomon and Shapiro, 2008).
Frequently asked questions
Are TAI and AIP the same thing?
Yes. TAI (traitement adaptatif de l’information) is the French translation of AIP (Adaptive Information Processing). Both acronyms refer to the same model.
Do you need to have experienced a major trauma for EMDR to be useful?
The AIP model holds that more ordinary experiences can also leave unintegrated memories. But the robust scientific evidence concerns mainly PTSD; for other difficulties, EMDR remains an option to discuss with a trained professional.
If the model has not been demonstrated, can EMDR be trusted?
Yes, for PTSD: its effectiveness has been established by clinical trials, independently of the theoretical explanation. Many effective treatments were used before their mechanism was precisely understood.
Is the memory erased after EMDR?
No. The patient remembers what happened, but the memory no longer causes the same distress. This is one of the central predictions of the AIP model.
In a nutshell
The AIP model explains trauma symptoms by poorly integrated memories, which EMDR is thought to help “digest”.
It is a useful clinical framework that echoes other models of traumatic memory, but it remains a hypothesis: EMDR is effective in PTSD, while the way it works is still being debated.
References
- Brewin, C. R., Gregory, J. D., Lipton, M., & Burgess, N. (2010). Intrusive images in psychological disorders: Characteristics, neural mechanisms, and treatment implications. Psychological Review, 117(1), 210-232. Link
- Cuijpers, P., van Veen, S. C., Sijbrandij, M., Yoder, W., & Cristea, I. A. (2020). Eye movement desensitization and reprocessing for mental health problems: A systematic review and meta-analysis. Cognitive Behaviour Therapy, 49(3), 165-180. Link
- Davidson, P. R., & Parker, K. C. H. (2001). Eye movement desensitization and reprocessing (EMDR): A meta-analysis. Journal of Consulting and Clinical Psychology, 69(2), 305-316. Link
- Ehlers, A., & Clark, D. M. (2000). A cognitive model of posttraumatic stress disorder. Behaviour Research and Therapy, 38(4), 319-345. Link
- Hase, M., Balmaceda, U. M., Ostacoli, L., Liebermann, P., & Hofmann, A. (2017). The AIP model of EMDR therapy and pathogenic memories. Frontiers in Psychology, 8, 1578. Link
- Inserm (2004). Psychothérapie : trois approches évaluées [Psychotherapy: three approaches evaluated]. Expertise collective [Collective expert review]. Paris: Les éditions Inserm.
- International Society for Traumatic Stress Studies (2019). PTSD Prevention and Treatment Guidelines.
- Landin-Romero, R., Moreno-Alcazar, A., Pagani, M., & Amann, B. L. (2018). How does eye movement desensitization and reprocessing therapy work? A systematic review on suggested mechanisms of action. Frontiers in Psychology, 9, 1395. Link
- Lee, C. W., & Cuijpers, P. (2013). A meta-analysis of the contribution of eye movements in processing emotional memories. Journal of Behavior Therapy and Experimental Psychiatry, 44(2), 231-239. Link
- Mol, S. S. L., Arntz, A., Metsemakers, J. F. M., Dinant, G.-J., Vilters-van Montfort, P. A. P., & Knottnerus, J. A. (2005). Symptoms of post-traumatic stress disorder after non-traumatic events: Evidence from an open population study. British Journal of Psychiatry, 186, 494-499. Link
- Nader, K., Schafe, G. E., & LeDoux, J. E. (2000). Fear memories require protein synthesis in the amygdala for reconsolidation after retrieval. Nature, 406, 722-726.
- National Institute for Health and Care Excellence (2018). Post-traumatic stress disorder (NG116). Link
- World Health Organization (2013). Guidelines for the management of conditions specifically related to stress. Geneva: WHO.
- Shapiro, F. (1989). Efficacy of the eye movement desensitization procedure in the treatment of traumatic memories. Journal of Traumatic Stress, 2(2), 199-223.
- Shapiro, F. (2018). Eye Movement Desensitization and Reprocessing (EMDR) Therapy: Basic Principles, Protocols, and Procedures (3rd ed.). New York: Guilford Press.
- Solomon, R. M., & Shapiro, F. (2008). EMDR and the Adaptive Information Processing model: Potential mechanisms of change. Journal of EMDR Practice and Research, 2(4), 315-325. Link
- van den Hout, M. A., & Engelhard, I. M. (2012). How does EMDR work? Journal of Experimental Psychopathology, 3(5), 724-738. Link
This article summarises the scientific evidence available at the date of publication. It is no substitute for clinical training, supervision, or the advice of a health professional.
