What is EMDR and how does it work ?

Eye Movement and Desensitization and Reprocessing (EMDR) is a very special, sometimes somewhat mysterious treatment method for Post Traumatic Stress Disorder (PTSD). In recent years, there has been increasing evidence for the effectiveness of EMDR for other (anxiety) disorders, depression and even personality problems.

It all started with a psychiatrist Francine Shapiro who invented it in 1987. As the story goes: during a walk in the park. She discovered that her eye movements (and those of her clients) had a lot to do with becoming less sensitive to unpleasant memories. She devised and carried out a fixed procedure for her clients and initially called it Eye Movement Desensitization. Later she added the more cognitive term Reprocessing. Much later she developed an interesting model about emotion processing that she called Accelerated Information Processing. A nice concept, but purely hypothetical because it has never been proven and probably to prove it is not possible at all. How can you demonstrate in the brain that information is processed even faster than it is normally processed? She does not mean this neuropsychologically but more clinically psychologically: information is processed faster emotionally (whatever exactly that means).

Below I would first like to explain EMDR as a procedure. As a neuropsychologist, I then want to discuss several theories about the how and why of the effectiveness of EMDR. I compare these theories with my knowledge about the brain and my experience with EMDR, especially in adults, and seeing and hearing videos in which EMDR was used. Finally, I actually want to give the best possible explanation for why EMDR works. Of course, this is not the ultimate final explanation because we simply don't know much yet. But I do want to try to get as close as possible to the most likely explanation. I also want to inspire researchers and clinicians (in particular) to come up with brilliant ideas to further improve the method and the explanation for the effectiveness of EMDR.

EMDR: the procedure

In feite wordt de gehele procedure van EMDR 1.0 heel goed uitgelegd in een Youtube-filmpje van 6min 26 seconden. Daarom zet ik hier de link naar dit filmpje neer. Klik op het plaatje om het filmpje af te spelen.


Several very short examples of HOW to do EMDR. Internationally this is the EMDR 1.0 procedure. It does not differ that much from the Netherlands procedure.

The essential components of the EMDR protocol are:

  1. The focus on the worst memory NOW. Without this beginning there is no tension
  2. The tension must be high enough. If the tension is much too low in the beginning (e.g. lower than 5), then the client does not seem to have the right focus or openness.
  3. The distraction: usually these are the eye movements. This can be done with moving fingers, but nowadays also with a light bar. Headphones also have sounds and also vibrating devices that you can hold. Also self-tapping, as in this video, is used.
  4. Short, regular interruptions in which the client must briefly indicate what is on the client's mind.
  5. Tension control during these interruptions: this way you can monitor whether the tension drops or increases.
  6. The therapist's questions or comments in between (often during the interruptions, but also during the eye movements). This cognitive part is, in my opinion, the most important. More about that later.

This is the standard protocol, for convenience called EMDR 1.0. In the follow-up EMDR course, more techniques are used, especially when standard EMDR does not seem to work so well (because that happens regularly). In fact, these techniques also created a slightly different, much shorter and easier EMDR procedure, for convenience called EMDR 2.0.

And it is precisely this EMDR 2.0 that has given me much more insight into the correct working mechanism of EMDR and probably the reasons why it does or does not work. I have been exclusively using this EMDR 2.0 procedure for years, which is at least as effective as EMDR 1.0. In most cases I am done in 1 or 2 sessions! Only with multiple psychotraumata (i.e. several very unpleasant images/memories) it can sometimes take up to 4-5 sessions, but usually never more than that. That seems remarkable, but you will read here and ultimately understand why it can happen so quickly.

Just a quick look at the differences between EMDR 1.0 and 2.0. For EMDR 2.0 I give a different initial instruction and explanation:

"Try to retain the most unpleasant image in your head as much as possible, but at the same time also follow the light or the fingers and alternately beat the rhythm of the rock song 'We will rock you' by Queen with your hands. You will notice that all this is quite difficult, but that is precisely the intention. Try to focus on the unpleasant image."

Here the difference with EMDR 1.0 is that someone is not asked not to stay with the initial image. Everything that comes up (in your head or body) must happen automatically with EMDR 1.0. In practice it is true that if you notice that someone is very compulsive and is very capable of only thinking about and holding on to that unpleasant image - and therefore cannot properly follow the light or hit the rock song - then my experience is that it doesn't work. The intended distraction then does not work. A very important observation!

Another difference with EMDR 1.0 is that with each interruption I only ask 2 things: 1. What is your tension now from 0 to 10? (same as EMDR 1.0); 2. Where in your body do you feel that tension now? Then I immediately continue with the eye movements and tapping (rhythmically tapping on legs). By the way, I tap 'We will rock you' just as loudly, but on a chair or table or something that makes a nice full  sound. So I don't think at all about what comes to someone's mind, I don't have a conversation about it. In practice with EMDR 1.0, it often turned out that these conversations often reduced the client's tension and focus. That is why it was omitted and my experience is that it is indeed of little or no use. But more about that later.

A third difference is that I often use cognitive interweaves for even more distraction. Especially when I notice that someone is very focused on the unpleasant image or the task itself. You can't get a laugh at all and especially with my sometimes very crazy, spontaneous comments, someone sometimes breaks, gives me a sudden smile. My experience is that without such a (often relieving short) break, the tension does not always subside. This tension then seems to be stuck or even held onto by the client himself, who is very focused on 'wanting to do it well' but not on feeling or experiencing his own emotions or images. At that moment they seem to be stuck in a so-called tunnel vision, from which, in my opinion as a therapist, you really have to get them out of it, usually suddenly.

Finally: I will never again do an extensive positive ending like with EMDR 1.0. I do ask and talk to the client for a few minutes about what it was like and how special it is that the client can now often look at the first worst picture without too much tension. In most cases, the client's tension is no longer palpable, not even in his/her body. Then you already know (after 1 time) that the EMDR has worked.

The proposed theories about EMDR: 1. Eye movements

To understand the mechanism of action, I would first like to discuss the most important theories that have been put forward in recent years. I then hold the current data from what I know from science and clinical practice against these theories.

Eye movements: necessary or not ? 

In my opinion, based on several studies, eye movements are not necessary for the positive effect of EMDR. However, they do appear to have a more positive effect than other distraction techniques such as sounds or vibrations.

You could actually see this very quickly in the videos made of EMDR treatments with real clients. Particularly in children. For example, EMDR was performed on babies and often did not require eye movements, but it did require body touching (feet). But you also saw that young children were not at all focused on the therapist's fingers; in other words, their eyes often barely moved. They usually stared straight ahead for a moment, something that several clients do during EMDR. It then looks very much like they are thinking about something.

There is another clear observation that eye movements are not very important: Ad de Jongh, among others, uses very fast finger movements horizontally and sometimes in all directions. Such movements are so fast that it is impossible to actually follow everything. The eyes often move much less than you would expect based on the stimuli presented. If you were to actually measure that in a laboratory, I am sure that these eye movements are indeed much less present than you think. Moreover: every therapist has his own pace with eye movements and there are several ways that involve looking at fingers as well as at balls on a stick moving back and forth, but also e.g. a computer screen on which a ball really goes in all directions.

It doesn't seem to matter, not even the direction of the eye movements. So much for the original idea that eye movements had to run from left to right so that both hemispheres of the brain were stimulated alternately. A completely nonsensical idea because it had long been clear that both hemispheres of the brain were 'on' at the same time, or were stimulated during the EMDR procedure. There has never been any question of 'switching between hemispheres'. This would only be possible if you offered something in only 1 visual field in a controlled laboratory environment and then with only 1 eye!

In short: just a simple but good study of various videos of EMDR with clients shows that eye movements do have some effect in the procedure, but are certainly not necessary. It seems to be much more about a certain type of focus rather than eye movements. And then it gets interesting: eye movements seem to have a more positive effect in clinical practice than just other distractions. There also seems to be more evidence for this in scientific studies, but because many studies are not of very high quality, this is not 100% proven.

In any case, my experience is that with unpleasant memories that are mainly present in a client in the form of clear images, eye movements seem to do more than mere sounds or vibrations. It seems that the light bar with a point of light flashing back and forth horizontally, but also the moving fingers of the therapist, can blur an image. As if it is being erased from long-term memory. Many clients say that after a few eye movement sets, not only is the tension less, but the original image has also changed. Usually more 'blurry', clouded, less sharp. And of course: the less sharp, the less strongly the associated emotion (e.g. fear) is triggered. That is plain and simple memory theory: the less something resembles a trauma image, the less strongly it is activated in the associative memory and the less strongly the associated emotion is activated.

Scientific studies in a laboratory actually show that adding information to a certain evoked image changes the image and its storage. In fact, this has been known for a long time because that's just how memory works.

Every time we retrieve something from long-term memory, that memory is rebuilt. It is really not the case that a memory is taken out of the closet as a kind of solid matter and then viewed again in consciousness. That is the biggest misconception among people who understand little about the cognitive memory system. EVERY memory is always an active remaking process based on just a few fundamental elements of that memory.

In other words: every time you consciously remember something, you are in fact changing this memory again. You put it back in your long-term memory, as it were, in a slightly different way. Especially if you put more effort into retrieving a memory, you automatically make it more vivid and extensive (the 'imagination inflation effect', some researchers have argued). Eventually, after remembering something many times, this memory has often changed so much that it is usually no longer the same as the original memory. And therefore often not how it really was!

Therefore, memories of our human memory are in fact really unreliable. Studies into what eyewitnesses thought they remembered about, for example, an accident or a robbery, are notorious examples of our weak memory. And I haven't even discussed our attention system, which plays an essential role in HOW something is stored.

2. The working memory model

Eye movements therefore do not appear to be essential for a positive effect of EMDR. But...they do seem to have a more positive effect than if you don't use them. To explain this, the Working Memory Model has been proposed. I don't know who came up with it first, but this article by Marcel van den Hout and Iris Engelhard from 2011 explains it very clearly (in Dutch). The article can be DOWNLOADED HERE.

Working memory is seen here as central to the processing of trauma. And then very specifically: in the desensitization of unpleasant feelings that are linked to an unpleasant memory. They literally say: an unpleasant memory becomes less clear, less salient, and therefore less vivid within that working memory, and thus fewer negative emotions are triggered. This is the so-called 'imagination deflation' effect.

This may explain that unpleasant memories no longer have such a major impact on someone, as is the case with PTSD. After all, with PTSD the related emotions (usually fear) are still far too intense, causing the brain (and body) to actually be in a constant state of anxiety (hyperarousal). This hyperarousal in any case makes it worse to sleep (both falling and staying asleep is often disturbed), and the dreams also often include nightmares that are strongly related to the theme and emotions of the trauma.

The working memory model is a very strong model, well accepted within cognitive (neuro) sciences. In fact, this Working Memory is nothing more than our Attention System. This attention system has be known for many years under different names in the scientific literature. In my opinion these are 'Attention Director' (Shiffrin & Schneider, 1977), 'Supervisory Attentional Control' (Shallice, 1982), 'Working memory' (Baddeley, 1986), and ' Cognitive Control' (Miller & Cohen, 2001). They are in fact all different terms for one and the same mechanism: the predominantly conscious attention system. This system is probably mainly set up by the dorsolateral prefrontal lobe. But honestly, it's a huge oversimplification to say that such a complex system is only located there. The brain is a complex network system and the dorsolateral prefrontal lobe has many connections with other cortical areas as well as with subcortical areas such as the amygdala and the RAS (Reticular Activating System).

Theoretically it is assumed that within this Attention System, which is often called Working Memory, in fact all important processing of cognitive and emotional information takes place: the focus (attention) is on the processing of specific information (this focus is very limited, often 5 plus/minus 2 information units maximum), and with that focus, part of the long-term memory is temporarily active and there is also inhibition of other information (selection).

It is very important to realize that the capacity of this Working Memory is constantly changing; something that is often overlooked. The brain is a biodynamic system, built on electrophysiological and chemical processes. This means that factors such as the amount of glucose that is burned within cells, as well as the amount of oxygen, and then the different types of amounts of substances such as neurotransmitters and hormones (emotions) largely determine this Working Memory capacity. For example, when anxiety is high, there is much less working memory capacity (tunnel vision) than when the biological system is very relaxed.

This varying capacity of our Working Memory (or Attention System) also differs per person. For example, EMDR works better for people who have limited attention capacity and are therefore less able to multitask (can consciously process fewer things at the same time). After all, distraction here will much more quickly lead to working memory strain (working memory is smaller) and that is precisely what is necessary to desensitize an unpleasant memory. People who have a large working memory therefore have to be distracted much stronger with EMDR.

The working memory load model also easily explains why we have already seen in practice that eye movements work much better than sounds alone. In Van den Hout's article, experiments show that beeps are only 1/3 effective in desensitizing an unpleasant memory. They also show that installing positive cognitions should NOT be accompanied by eye movements. The idea is that positive cognitions are then less well stored.

3. The role of Self-image in EMDR

It is clear that the Working Memory Model can explain a very important part of the effects of EMDR. However, it is also clear that it is researchers and not clinicians who can properly substantiate this model. But in my opinion, they have not yet been able to explain all the effective elements of the EMDR procedure.

I see that something else is essential as well that is being overlooked by researchers, but probably not by clinicians. Successful EMDR also requires a cognitive restructuring, a changed meaning to the unpleasant event and especially in the self-image. If that does not happen, someone will continue to suffer too much from the trauma. You may have desensitized an unpleasant memory, which means nothing more than that the memory is less sharp and therefore the associated emotions are evoked less intensely, but feelings are a complex interplay of cognitions and emotions. If cognitions, i.e. the attribution of meaning, have not also changed, then not enough has changed. In my experience, it is even true that an essential change has taken place at a fundamental Self-image level after a successful EMDR treatment. And that is within the Self-Image, within the 2 main domains Validation and Control. These 2 dimensions are also very strongly linked to Sadness and Fear respectively.

How do I think it works now? First of all, the strongest emotions must be addressed and reduced through EMDR. With PTSD, a client is clearly in a state of hyperarousal: intense emotions are stored in both the brain and the body around the unpleasant event. We prefer to stay away from these intense emotions, but that is precisely the problem with PTSD. With high intensity emotion, normal emotional processing seems to be blocked. The attribution of meaning does not take place in a normal way: one cannot place the event in the long-term memory at all (almost literally).

Processing is a very cognitive process: the brain is a planning machine and constantly stores events in different drawers/boxes according to a certain logic. But if something happens that (often completely unexpectedly) really cannot be placed in a certain context, both because of being 'illogical' and because of too intense emotions, then this event actually continues to wander in our brain. The brain still tries to place it, especially in REM sleep, but if that produces too many intense emotions, this process of storing it logically will not work well.

So EMDR must of course reduce the strongest emotions and is well able to do this through the distraction of eye movements, the rhythmic beating of the legs to a rock song and through crazy distractions on my part. This distraction should not be too much, because - as the working memory theory predicts - then there is too little space to keep the memory itself within the working memory. In fact, the distraction serves to keep the capacity of the working memory large enough to actually contain both the memory and also add space for other information. Adding that other information is actually the process of cognitive restructuring, so that its meaning changes. And this change happens fascinatingly quickly: often in just a few milliseconds. You usually see it happen: the client suddenly has a different look in his/her eyes or suddenly puts up a different face. As if the 'insight' sunk in, suddenly.

In my opinion, the second effective aspect of EMDR is: cognitive restructuring that takes place very quickly. An insight that literally comes to someone. E.g. After, for example, a hostage situation with dead people, clients with PTSD often feel enormous guilt that they survived and others did not. They cannot forget that image of shooting one of the hostages dead, precisely because of this feeling of guilt. Rationally speaking, they know that they are not guilty, they have often heard that from others. But in their Self-image, this being not-guilty has not changed since the trauma. My idea is that this Self-Image Demon (a negative core cognition) has been changed so much by the intense emotion that many other ideas surrounding Self-Guilt have also increased in intensity (in just one take). The strong emotion about this seems to prevent rational cognitive restructuring. It seems as if they are in a loop, a tunnel vision. They know they are innocent, but they don't believe or feel it.

It is interesting that within the trauma memory, i.e. within their memory of that unpleasant event, they must both recall the associated emotion in their working memory and also add the corrective cognition at the same time in that working memory. Only then does the memory seem to be viewed differently. And I suspect that distracting shock effects can increase someone's working memory capacity just as quickly, so that the correct corrective information is then merged with the trauma memory and the link with the emotion is therefore a lot less.

Another interesting observation from practice that supports the hypothesis of cognitive restructuring is the equally great effectiveness of imaginary rescripting. Here too, unpleasant emotions and/or images must be considered. If a client succeeds, new information and/or an image is added to someone's imagination with the help of the therapist. For example, the suggestion is made that a stronger other, or the client himself who has become much stronger or bigger, comes into this unpleasant picture and helps the frightened client at that moment. As a result, this unpleasant memory usually changes very quickly and is stored differently in the long-term memory. This is also more self-empowerment: someone often realizes after such an imagination session that he/she is much stronger than he/she ever thought.

That negative thinking process or loop that someone often ends up in after a serious psychotrauma is in fact a completely normal brain process. In my opinion, clear parallels can be drawn with normal cognitive processes that involve such a loop.

For example, who hasn't experienced constantly reading a long piece of your own text, and yet still not seeing a few spelling errors? Only when you take a break or get into a different mood will you see the same spelling error after reading it again. In fact, during this frequent reading you were constantly in a cognitive loop, tunnel vision, that's how you could put it. It takes a change, often sudden, to get out of that loop. It is known that if someone takes a hot coffee in between and burns his tongue, the spelling error will also be noticed when reading.

Another well-known phenomenon of not seeing but knowing something is the concept of 'realization'. You only realize something, you only see or understand something when multiple cognitions (ideas) coincide very quickly. If that does not happen, then something will continue to be misunderstood. Everyone knows of a difficult piece of information that has been explained to you several times but it is still not understood. Until your teacher suddenly takes a completely different approach and the explanation becomes completely different. Then you may suddenly 'get it'. This insight comes suddenly and you are even surprised that you suddenly understand it and... that it is so simple that you do not understand why you did not see it much earlier.

Another example explains the power of just one thought very clearly in the process of cognitive restructuring or meaning formation. Take a look at the picture below. Those are all stains. But if you look closely it is a known thing. Do you see it yet?


As long as you look at the picture and your knowledge is not optimal, your interpretation of meaning falls short, then you do not see what others do see. Just try to look carefully, rack your brain and try to find in the picture which familiar something can be found. Keep staring, squinting and pissing yourself off why you're so stupid that you don't see it...

But if I tell you now that I can significantly change your brain here within a few milliseconds, in such a way that you see it and then think afterwards: "Gosh, why didn't I see this before?". Only then do you realize that seeing or understanding something is an active process that has everything to do with certain knowledge. You know and knew what this is, it had been in your brain for a long time, but you still didn't see it. In fact, this is the same phenomenon as the person with PTSD who feels very guilty after a hostage situation with dead people. He/she knows that he/she is innocent, but he/she does not realize it.

By the way... an animal can be seen in the picture, very clearly in fact.

Don't you see it yet?

It's a Dalmatian...

And once you have discovered it and you see it....now try to look at the picture and NOT see the dog. So you won't be able to do that anymore. Your brain has changed significantly, in a split second.

This often also happens during EMDR treatment. But only if a cognitive restructuring has actually taken place. When someone actually looks differently, gives a different meaning to the event and especially to themselves. A fundamental change has then taken place in the (unconscious) Self-image.

Go back to that YouTube video explaining EMDR. Listen carefully and realize that the biggest change after EMDR most often is that clients do not see themselves as powerless or as losers anymore.

One more thing: how can these changes happen so quickly? I suspect it is due to something that is hardly taught in Dutch neuropsychology: electrophysiology. The brain is a complex electrical network and the shifting of currents and patterns is constantly happening in the brain. The only reason such changes occur so quickly (in milliseconds) is that they are not biochemical processes (which are much slower) but electrophysiological ones. Someone who has described this very nicely is a Hungarian György Buszáki in Rhythms of the Brain (2011). A complicated subject but very worthwhile reading material.

In summary

The Working Memory Load Model can explain many effects of EMDR. But that is not everything. Fundamental changes are also needed in the Self-Image (the core of your personality), something we call cognitive restructuring. See also my page about Self-image and PTSD. Finally, the rapid changes in EMDR are likely due to electrophysiological changes, not neurochemical or tissue changes.


I will not take any responsibility for how the information on this website will affect you. It always remains your responsibility to handle all information with care and in case of medical or mental problems you should ALWAYS consult a professional in your neighbourhood!

Ik neem geen enkele verantwoordelijkheid voor hoe de informatie op deze site u zal beïnvloeden. Het blijft altijd uw verantwoordelijkheid om al deze informatie zorgvuldig te bekijken. In het geval van lichamelijke en/of mentale problemen dient u ALTIJD een professional in uw directe omgeving te waarschuwen!