What Does Research Actually Say About Eye Movements in EMDR?
EMDR has one feature that attracts enormous attention:
The eye movements.
They are unusual.
They are visible.
And they make EMDR look very different from traditional talk therapy.
So it is reasonable to ask:
Do the eye movements actually do anything?
Or is EMDR effective because the client is repeatedly remembering the trauma while sitting with a therapist?
This question has been debated for decades.
And the answer today is more interesting than either extreme.
Research does suggest that eye movements can change the subjective quality of distressing memories.
But that does not mean we have completely explained why EMDR works—or that eye movements alone explain the entire therapy.
First, EMDR and eye movements are not the same research question
This distinction is essential.
There are really two separate questions:
Does EMDR help people with trauma and PTSD?
What does the eye-movement component contribute to EMDR?
Those are not interchangeable.
EMDR is a complete psychotherapy protocol.
It includes history taking, preparation, assessment, memory reprocessing, cognition work, body scanning, closure, and reevaluation.
Eye movements occur within that larger treatment.
So evidence that EMDR works does not automatically tell us which component produced the effect.
To understand the eye movements themselves, researchers have had to isolate them.
How do researchers study eye movements?
One common experimental design is surprisingly simple.
Participants choose an emotional autobiographical memory.
They rate how:
vivid it feels,
emotionally intense it feels,
or unpleasant it feels.
Then they recall the memory under different conditions.
For example:
while making eye movements,
while performing another demanding task,
or while simply remembering without a second task.
Then they rate the memory again.
This allows researchers to ask whether eye movements change the experience of remembering.
What did a 2020 meta-analysis find?
A 2020 meta-analysis by Houben and colleagues examined 15 laboratory studies involving 942 participants.
The researchers compared:
eye movements with memory recall alone,
other dual tasks with memory recall alone,
and eye movements with alternative dual tasks.
The results were interesting.
Eye movements produced a moderate reduction in memory vividness compared with recall alone, with an average effect size of d = .59.
They also reduced emotionality, although the effect was smaller, at d = .28.
That suggests the most consistent immediate effect may be on how vivid the memory feels.
Eye movements seem to affect vividness more than emotion
This is a subtle but important finding.
People often imagine eye movements directly “turning off” emotional distress.
The research suggests something more nuanced.
The effect may begin with the memory itself becoming less sensory and immersive.
Less vivid.
Less sharp.
Less present.
Then emotional intensity may also decrease.
The 2020 meta-analysis found that both eye movements and other dual tasks affected vividness more strongly than emotionality.
That pattern fits well with working-memory theory.
What is working-memory theory?
Working memory is the brain's limited-capacity system for temporarily holding and manipulating information.
A vivid emotional memory uses working-memory resources.
So does visually tracking a moving target.
When you do both at the same time, they compete.
The working-memory account proposes that there are not enough cognitive resources available to maintain the memory with exactly the same vividness while also performing the eye-movement task.
The memory may therefore become temporarily degraded.
When it is recalled later, some of that reduced vividness may remain.
This is not just a theoretical idea
The working-memory account is appealing partly because it makes testable predictions.
If the theory is correct:
eye movements should reduce memory vividness more than recall alone,
other sufficiently demanding tasks should sometimes do the same thing,
the secondary task should occur while the memory is active,
and the amount of cognitive demand should matter.
Research has generally found support for several of these predictions.
That is one reason working-memory taxation has become one of the strongest current explanations for eye-movement effects.
But other tasks work too
This is where things get especially interesting.
The 2020 meta-analysis found that alternative dual tasks also reduced memory vividness and emotionality.
That included tasks other than eye movements.
In fact, alternative dual tasks produced:
a vividness effect of d = .49 compared with recall alone,
and an emotionality effect of d = .32.
So the effect may not depend on horizontal eye movements having a unique neurological property.
It may depend substantially on doing a second, cognitively demanding task while the memory is active.
Are eye movements better than other dual tasks?
Maybe slightly for vividness.
In that same meta-analysis, eye movements showed a small advantage over alternative dual tasks for reducing vividness, with d = .29.
But there was essentially no meaningful difference in emotionality between eye movements and alternative tasks.
The researchers also cautioned that the number of studies was limited and that the different tasks were not independently matched for how much they taxed working memory.
So this does not establish that eye movements are uniquely necessary.
It suggests they are one effective way to create the dual-task effect.
Earlier research found similar patterns
This line of research is not new.
A study published in 2001 asked healthy volunteers to recall positive and negative autobiographical memories while performing rapid eye movements, finger tapping, or no dual task.
The researchers found that memories became less vivid after eye movements, and negative memories became less negative afterward.
That kind of experimental result helped fuel later investigation into whether eye movements directly alter the subjective qualities of emotional memories.
What about EMDR with eye movements versus EMDR without them?
This is closer to the clinical question.
The 2020 meta-analysis summarized an earlier 2013 meta-analysis by Lee and Cuijpers that compared EMDR procedures with and without eye movements.
According to that earlier analysis, clinical studies found greater reductions in subjective distress when eye movements were included, with a medium effect size of d = .53.
They also found larger increases in positive-cognition ratings, with d = .72.
Laboratory studies in that earlier review also found larger reductions in memory vividness and emotionality when eye movements were included.
That supports the idea that eye movements contribute something beyond simple recall.
But it still does not mean they explain the whole treatment.
Why laboratory research is not the same as clinical treatment
This distinction matters enormously.
A laboratory study may ask a healthy participant to remember an upsetting event for a few minutes.
EMDR therapy involves:
clinical assessment,
target selection,
trauma history,
preparation,
therapeutic relationship,
reprocessing,
cognition work,
body sensations,
closure,
and reevaluation.
Those are very different environments.
The authors of the 2020 meta-analysis explicitly cautioned that their laboratory results cannot simply be generalized to clinical practice.
So when a lab study shows reduced memory vividness, that does not automatically prove how much PTSD symptoms will improve in therapy.
We still do not fully know whether reduced vividness drives symptom change
This may be the most important unanswered question.
If eye movements make a traumatic memory less vivid, does that directly cause PTSD symptoms to improve?
Possibly.
But that connection has not been definitively established.
The authors of the meta-analysis noted that convincing evidence linking eye-movement-induced changes in vividness and emotionality directly to reductions in clinical symptoms is still limited.
That means we have evidence for one mechanism-level effect without yet having a complete bridge from that effect to every part of clinical recovery.
Does this weaken EMDR?
Not necessarily.
It simply means mechanism research is still evolving.
A therapy can have evidence of effectiveness before every causal pathway is understood.
Medicine is full of treatments that were clinically useful before scientists fully understood their mechanisms.
The responsible position is not:
“We do not know exactly how EMDR works, therefore EMDR does not work.”
Nor is it:
“EMDR works, therefore every explanation therapists use for it must be correct.”
Those are both bad scientific reasoning.
What we know with more confidence
We can say with reasonable confidence that:
performing eye movements while recalling emotional memories can reduce their subsequent vividness,
emotional intensity may also decrease,
other demanding dual tasks can produce similar effects,
the eye-movement effect fits well with working-memory competition,
and eye movements appear to contribute something measurable beyond simple memory recall in at least some experimental and clinical comparisons.
That is already interesting.
What we do not know with the same confidence
We cannot yet say with equal confidence that:
eye movements are the sole active ingredient in EMDR,
bilateral left-right stimulation is uniquely necessary,
eye movements work primarily by synchronizing the hemispheres,
they recreate REM sleep,
they “reset” the amygdala,
or reductions in memory vividness fully explain why PTSD improves.
Those claims go beyond the evidence.
Does “bilateral” matter?
This is one of the questions I find most interesting.
If bilateral stimulation itself were the critical mechanism, we might expect alternating left-right stimulation to have a uniquely powerful effect.
But other cognitively demanding tasks can also reduce memory vividness.
That makes it plausible that dual-task competition matters more than bilaterality itself.
This does not mean alternating eye movements are irrelevant.
It means the mechanism may be less about “left brain, right brain” and more about what happens when a vivid memory competes with another attention-demanding task.
Why are eye movements so effective at creating that competition?
Vision uses substantial cognitive resources.
Following a moving target requires:
attention,
tracking,
spatial processing,
and motor coordination of the eyes.
At the same time, visual traumatic memories may themselves be highly image-based.
That combination could make eye movements particularly effective at competing with vivid visual imagery.
This may help explain why eye movements often show strong effects on memory vividness.
Could the effect be simple distraction?
Not exactly.
Distraction would imply that the person stops thinking about the memory.
But in EMDR, the client is intentionally asked to keep the memory active while performing the second task.
That overlap matters.
The brain is not simply being told:
“Think about something else.”
It is being asked:
“Hold this memory in mind while doing this too.”
That is the essence of the dual-task model.
What about dual attention?
Another concept often discussed in EMDR is dual attention.
Part of the client's attention is on the traumatic material.
Part remains oriented to the present.
The client is remembering the past while simultaneously tracking something happening now.
That may help create distance between:
“This happened then”
and
“This is happening now.”
Dual attention and working-memory taxation are not necessarily competing explanations.
Both could contribute.
What about physiological arousal?
Eye movements may also affect physiological arousal.
Some theories propose that repeatedly orienting toward a harmless moving stimulus may initially activate attention and then promote a calming response.
That is often called an orienting response account.
There is some research interest in this explanation, but current evidence does not establish it as the sole mechanism.
EMDR may involve several processes at once.
What about memory reconsolidation?
Memory reconsolidation is another frequently discussed possibility.
When a memory is retrieved, it may become temporarily modifiable before it is stored again.
If the memory is retrieved while its vividness, emotional intensity, meaning, or context changes, the subsequently stored representation may also change.
That could fit with what happens during EMDR.
But saying:
“Eye movements cause memory reconsolidation”
is stronger than current evidence justifies.
Reconsolidation remains a plausible contributing framework rather than a complete proven explanation.
Does EMDR reproduce REM sleep?
This explanation remains popular.
Both involve eye movements.
REM sleep is involved in memory processing.
So the analogy is understandable.
But similarity is not mechanism.
There is not enough evidence to say that therapeutic eye movements work because they recreate REM sleep while someone is awake.
It is an interesting hypothesis, not an established fact.
Does EMDR integrate the left and right hemispheres?
This is another explanation I would treat very cautiously.
The brain does not operate as a simple emotional-right-brain versus logical-left-brain system.
Memory, emotion, attention, and regulation involve distributed networks across both hemispheres.
There is not strong evidence that the essential therapeutic action of EMDR is alternating activation or synchronization of the left and right hemispheres.
It is a compelling soundbite.
It is not the strongest scientific explanation.
Why the eye-movement debate has lasted so long
Part of the reason is that EMDR contains several active elements at once.
A client:
activates a painful memory,
attends to emotional responses,
notices body sensations,
works with beliefs,
engages in repeated bilateral stimulation,
and does all of this within a therapeutic relationship.
Separating those elements experimentally is difficult.
And even if researchers isolate an eye-movement effect in the lab, that does not tell us exactly how that effect interacts with everything else happening in therapy.
Eye movements are probably neither magic nor meaningless
That may be the clearest takeaway.
The evidence does not support dismissing eye movements as irrelevant theater.
They appear capable of producing measurable changes in emotional memory.
At the same time, the evidence also does not support describing them as a mysterious neurological switch that “clears trauma.”
The reality is more sophisticated.
Eye movements appear to modify how emotional memories are experienced.
Working-memory competition provides a strong explanation for at least part of that effect.
But EMDR therapy involves more than eye movements.
Why this matters for ART too
This research matters beyond EMDR.
Accelerated Resolution Therapy, or ART, also uses visually guided eye movements while emotionally meaningful material is active.
That means working-memory research may help us think about what the eye movements contribute to ART as well.
But ART uses those eye movements within a different protocol involving more structured guidance, body sensations, and deliberate imagery procedures.
So we should not assume the mechanisms are identical.
The eye-movement component may overlap.
The therapies themselves do not.
Why I care about this distinction
I am trained in both ART and EMDR.
One of the things I find most compelling about studying both is how quickly simplistic explanations fall apart.
Both use eye movements.
But they do not feel the same.
They do not process material the same way.
They do not ask the client to do the same things between sets.
And they do not have identical research histories.
That makes the scientific question much more interesting than:
“Which eye-movement therapy is better?”
The better questions are:
What are the eye movements actually doing?
What does working-memory competition contribute?
What is happening to imagery?
What happens to emotional arousal?
How does memory updating occur?
Which components are shared?
Which are unique?
Those are much more useful questions.
Working with Laura Geftman, LCSW
I use EMDR and Accelerated Resolution Therapy as part of a broader trauma-focused practice.
My work may also incorporate IFS, Cognitive Processing Therapy, Prolonged Exposure, CBT, psychodynamic, interpersonal, attachment, and schema therapies depending on the client and clinical question.
I offer focused trauma therapy and therapy intensives for adults in Pennsylvania, New Jersey, New York, and Florida.
In-person therapy intensives are available in Ardmore, PA, on the Main Line near Philadelphia.
Virtual therapy may also be available when clinically appropriate.
Interested in EMDR, ART, or focused trauma therapy?
If you understand what happened but a particular memory, image, trigger, body response, or emotional reaction still feels stuck, EMDR, ART, or another focused trauma approach may be worth exploring.
The goal is not to choose therapy based on a neuroscience slogan.
The goal is to understand what kind of treatment fits the person, the target, and the clinical context.
Suggested Reading
If you are interested in EMDR, ART, eye movements, bilateral stimulation, or trauma memory, these related articles may be helpful:
How Do Eye Movements Work in Accelerated Resolution Therapy?
ART vs. EMDR: How Is Bilateral Stimulation Used Differently?
FAQ
Do eye movements actually matter in EMDR?
Research suggests they do contribute measurable effects. Laboratory studies show that eye movements during memory recall can reduce the vividness and emotionality of negative autobiographical memories compared with recall alone.
What does the strongest research show?
A 2020 meta-analysis of 15 laboratory studies involving 942 participants found that eye movements reduced memory vividness with an average effect size of d = .59 and emotionality with an effect size of d = .28 compared with recall alone.
Are eye movements better than other dual tasks?
Eye movements showed a small advantage over alternative dual tasks for reducing vividness in the 2020 meta-analysis, but alternative tasks also reduced vividness and emotionality. This supports the idea that working-memory competition may be important.
Does this prove that eye movements are what makes EMDR work?
No. EMDR is a complete psychotherapy protocol, and laboratory memory studies do not fully reproduce clinical treatment. Eye movements appear to contribute something, but they do not explain every aspect of EMDR's clinical effects.
Why do eye movements make memories less vivid?
One leading explanation is working-memory taxation. Recalling a vivid memory and making eye movements simultaneously compete for limited cognitive resources, which may weaken the vividness of the memory while it is being held in mind.
Does bilateral stimulation work by balancing the two hemispheres?
That is not an established explanation. The evidence fits better with working-memory, dual-attention, and memory-processing accounts than with a simple left-brain/right-brain balancing model.
Does EMDR work by recreating REM sleep?
That has been proposed, but it has not been established. The presence of eye movements in both REM sleep and EMDR does not demonstrate that the underlying processes are the same.
Is memory reconsolidation part of EMDR?
It may contribute, but it has not been established as the single mechanism responsible for EMDR. Researchers continue to study how memory retrieval, cognitive load, new information, and emotional processing interact.
Can research on EMDR eye movements tell us how ART works?
It can provide useful hypotheses because ART also combines emotional memory activation with visually guided eye movements. But ART uses those eye movements within a different protocol, so EMDR mechanism research cannot simply be assumed to apply identically to ART.
Is the science settled?
No. The evidence that eye movements affect emotional memory is meaningful, but researchers are still studying how those changes relate to full clinical recovery and which mechanisms account for EMDR's treatment effects.