Mirror therapy is a rehabilitation technique that uses the reflection of an unaffected arm or leg to create the visual impression that an affected or missing limb is moving normally. In stroke rehabilitation, the main goal is to support motor practice; for phantom limb pain, the same illusion may help resolve a painful mismatch between what the brain sees and what it senses. The tool is remarkably simple, often just a mirror or a mirror box, but the experience can be powerful enough that it deserves careful setup and professional guidance.
What mirror therapy is and what the mirror actually shows
During mirror therapy, a mirror stands along the body’s midline between the arms or legs. The person moves the unaffected limb while watching its reflection. The affected limb stays behind the mirror, still or moving as able, so the reflection appears to be the affected limb moving with normal range and timing.
The term mirror box therapy usually refers to the classic arrangement: a simple, low-cost box with a vertical mirror fixed in the middle. The unaffected arm goes on the reflective side and the affected arm sits out of sight on the other side. Mirror therapy is the broader principle; a mirror box is one practical way to deliver it. For a leg, a larger mirror can be placed lengthwise between the legs, with the affected leg hidden behind it and the unaffected leg in front.
The point is not to trick someone into believing an injured or absent limb is physically present. It is to give the brain a coherent visual signal during a purposeful movement. The person watches a hand open and close, turn palm up, or touch each fingertip with the thumb. In the mirror, that movement appears on the affected side. The image can make a difficult movement feel more imaginable and can change how the brain represents the limb.
How the movement illusion may change the brain’s map
Mirror therapy creates a visual-proprioceptive mismatch: the eyes report movement while joints and muscles on the affected side may report little movement, no movement, or no limb at all. One proposed response to that conflict is cortical reorganization, meaning the brain adjusts its representation of the body and movement.
Mirror neurons are often used to explain this idea, but the explanation needs restraint. In 1992, Giacomo Rizzolatti and colleagues found neurons in a premotor area of macaques that fired both when an action was performed and when a goal-directed action was observed. In humans, researchers have identified a mirror-neuron system involving the inferior frontal gyrus and inferior parietal lobe that maps observed and performed actions. The human evidence is indirect, however; there is no universally accepted single-cell proof of mirror neurons in humans. Mirror therapy does not depend on a single, settled neural explanation to be used thoughtfully.
For people with phantom limb pain, brain imaging findings offer one possible clue. In a 2014 study of 11 participants, four weeks of daily mirror training were associated with a 27% average reduction in phantom limb pain and lower activity in the inferior parietal cortex. A 2019 NIH imaging study in people with lower-limb amputations found that four weeks of therapy brought an overactive sensorimotor-cortex response to foot images down to the level seen in healthy controls. These findings describe changes observed in small groups; they do not predict an individual result.
Where mirror therapy fits: stroke, phantom pain, and CRPS
Mirror therapy is used most often for weakness on one side of the body after stroke, called hemiparesis, involving either the upper or lower limb. It has also been used for phantom limb pain after amputation and for complex regional pain syndrome (CRPS), including post-stroke CRPS or shoulder-hand syndrome.
Each condition poses a different problem, so the target of therapy differs too. After stroke, a therapist may use the mirror to support repeated movement practice when the affected hand or leg has limited voluntary movement. After amputation, the reflected intact limb can provide movement-related visual feedback for a limb that is no longer physically there. With CRPS, mirror therapy may be one physiotherapy option, not a stand-alone treatment plan.
For CRPS in particular, treatment needs perspective. CRPS commonly affects an arm or leg after an injury and involves pain that is disproportionate to the trigger. Most people with CRPS improve spontaneously within a year, and guidance describes four pillars of care: education, pain relief, physical rehabilitation, and psychological support. A 2024 review of graded motor imagery and mirror therapy reported an average 20-point reduction on the Neuropathic Pain Scale, but mirror therapy alone is not considered sufficient in most CRPS cases.
Does mirror therapy work for stroke rehabilitation?
The strongest summary evidence comes from a 2018 Cochrane review of 62 studies involving 1,982 participants. It found moderate evidence that mirror therapy improved motor function after stroke, with a standardized mean difference (SMD) of 0.47, and motor impairment, with an SMD of 0.49. An SMD is a way to express the average difference between groups on a common scale; these figures indicate a measurable average benefit, not a promise that every person will make the same gain.
The review also found moderate evidence of improvement in activities of daily living: SMD 0.48, based on 622 participants across 19 studies. That matters because rehabilitation is not only about an isolated movement. It is about the practical work that movement supports.
| Finding after stroke | What the review reported |
|---|---|
| Motor function | Moderate evidence of improvement; SMD 0.47 (95% CI 0.27–0.67) |
| Motor impairment | Moderate evidence of improvement; SMD 0.49 (95% CI 0.32–0.66) |
| Activities of daily living | Moderate evidence of improvement; SMD 0.48 (95% CI 0.30–0.65) |
| Typical study dose | Five sessions a week, 30 minutes per session, for four weeks |
| Range studied | Three to seven sessions weekly, 15–60 minutes, for two to eight weeks |
Mirror therapy has been studied in acute, subacute, and chronic stages, including more than six months after stroke. It has also shown effects in people with severe hemiparesis. Still, recovery is highly variable: only 20% of people with severe arm paresis achieve full upper-limb function, compared with 80% with mild paresis, and about half of people whose arm is completely paralyzed at first regain some motor function within six months. Those figures are broad recovery data, not a forecast for one patient or a measure of mirror therapy alone.
The evidence has important limits. Many trials had small samples, incomplete method reporting, or unclear connections between study components. The size of the benefit also changed with the comparison treatment: when mirror therapy was compared with unrestricted viewing rather than a covered mirror, the effect was no longer present. For post-stroke pain, the clearest benefit in the review was in CRPS; there was no clear effect for visual-spatial neglect.
Why a mirror may help with phantom limb pain
Phantom sensations are common after amputation and are not always painful. Phantom limb pain is also common: reported lifetime prevalence is 76% to 87%, and the one-year incidence after amputation is 82%. In the United States, an estimated 2.3 million people live with amputations, with a projection of 3.6 million by 2050. These numbers explain why a low-tech approach to a difficult pain problem continues to draw attention.
For phantom pain, the reflected intact limb appears to move where the missing limb would be. That visual movement may help resolve the mismatch between the visual signal and the absent proprioceptive signal, potentially correcting an altered cortical map. The original experimental work by V.S. Ramachandran and D. Rogers-Ramachandran, published in 1996, used a self-built “virtual reality box” with 10 patients.
One of those patients, D.S., had not felt movement in a phantom arm for 10 years. Within minutes of using the mirror box, he reported feeling movement again. He later used the box at home for 15 minutes a day; after three weeks, the phantom arm had telescoped completely into the residual limb. Ramachandran called it the first known “amputation” of a phantom limb. It is a memorable observation, but it is a single patient’s experience rather than a universal outcome.
Controlled research also suggests potential benefit. In a randomized trial of 22 people with lower-limb amputation, every participant in the mirror-therapy group reported less pain after four weeks. The median change was minus 24 mm on the visual analog scale, while the covered-mirror and mental-visualization control groups did not improve. A person whose phantom limb has telescoped may benefit less, because the reflected image of a full limb may not fit that person’s altered body representation.
How long mirror therapy can take to work
There is no reliable personal timetable. In the stroke studies summarized by Cochrane, treatment programs lasted from two to eight weeks, with a median of four weeks. The common schedule was five 30-minute sessions per week for four weeks. That is a useful structure for a clinician-led plan, not a deadline for recovery.
Phantom-pain reports show why it is unwise to promise a timeline. D.S. described movement within minutes and a fully telescoped phantom after three weeks of daily practice. In a case report, a lower-limb amputee who used mirror therapy at home for 25 minutes a day reported marked pain reduction within six weeks and no phantom pain after three months. The trial of 22 lower-limb amputees measured improvement after four weeks. Each is an example of what occurred in a particular study or case, not proof that anyone else will respond in the same way.
Consistency, a convincing visual setup, and an appropriate target movement all matter. So does the underlying condition. If symptoms are new, changing, or worsening, do not extend a home experiment in the hope that it will settle. Contact a physician or rehabilitation clinician for assessment.
How to use a mirror box safely in a home program
Home mirror therapy can follow initial instruction from a physical therapist or occupational therapist. NICE guidance supports independent home use after that instruction, and a hospital may lend the mirror box. A practical home schedule is five days each week, 30 minutes a day, or two 15-minute sessions, for at least four weeks.
- Prepare the view. Place the mirror at a right angle to the body in a quiet space. Remove jewelry and cover tattoos on the unaffected arm, since recognizable details can break the illusion.
- Hide the affected limb. Put it behind the mirror or inside the mirror box. Keep the unaffected limb on the reflective side so its image appears where the affected limb would be.
- Watch the reflection. Keep attention on the mirror image rather than looking behind the mirror. Looking behind it interrupts the visual illusion that the exercise requires.
- Begin with simple movements. Open and close a fist, touch the thumb to each fingertip, lift and lower the wrist, turn the palm, or “play piano” with the fingers.
- Advance only as instructed. Later exercises can include touching textures such as wallpaper, sandpaper, a brush, sponge, or cotton, and moving objects in the hand.
A standardized upper-limb protocol can combine up to 125 distal and proximal arm-movement combinations. The basic movement set includes shoulder flexion and extension, shoulder abduction, elbow bending and straightening, forearm pronation and supination, wrist bending and straightening, and opening and closing the fist. That variety is useful only when it fits the person’s plan; more movements do not automatically mean better therapy.
Downsides, limits, and myths worth clearing up
Mirror therapy is generally simple, but simple is not the same as risk-free. The home guidance lists clear reasons to pause: new pain, changed sensation, feelings of grief or confusion, dizziness, or sweating. Tell the therapist when any of these occur. People with severe cognitive impairment, visual-spatial neglect, or apathy may not be able to carry out mirror therapy correctly on their own and need continuing supervision.
Emotional discomfort can be real. In the lower-limb amputation trial, 2 of 22 participants experienced brief grief lasting less than two minutes when seeing the healthy leg in the mirror. With CRPS, mirror therapy can in individual cases trigger or worsen symptoms on the unaffected side. That possibility is one reason CRPS should not be managed with mirror therapy as a solo treatment.
- Myth: It is hypnosis. Mirror therapy uses visual feedback during ordinary movement practice; it does not require hypnosis.
- Myth: Mirror neurons prove exactly why it works. They are one proposed piece of the explanation, and the human evidence remains indirect.
- Myth: A mirror box replaces rehabilitation. It is an adjunct to rehabilitation, not a substitute for individualized medical, physical, occupational, or pain care.
- Myth: It works for everyone. Research reports average effects and individual responses vary. Small trials and methodological limits mean the true size of benefit remains uncertain.
See a doctor or rehabilitation clinician before starting if the diagnosis or symptom pattern is unclear. Seek their advice promptly for new or worsening pain, sensory changes, dizziness, confusion, sweating, or distress during practice.
Frequently Asked Questions
What is mirror therapy?
Mirror therapy places a mirror between the limbs so the reflection of the unaffected moving limb looks like the affected or missing limb. The visual feedback is used in rehabilitation for movement or pain-related goals.
Does mirror therapy actually work?
For stroke, a Cochrane review found moderate evidence of average improvements in motor function, motor impairment, and daily activities. It does not guarantee a result for an individual, and many studies were small or incompletely reported.
What is mirror therapy used to treat?
Its main uses are motor rehabilitation after stroke, phantom limb pain after amputation, and CRPS, especially post-stroke shoulder-hand syndrome. The treatment goal and safety plan should match the condition.
How does mirror therapy work for phantom limb pain?
The reflection shows an intact limb moving where the missing limb would be. This may help resolve a conflict between visual input and body signals and may alter an unhelpful cortical representation. The exact mechanism is not fully settled.
How long does it take for mirror therapy to work?
Stroke programs in the evidence commonly ran for four weeks, though studied programs ranged from two to eight weeks. Phantom-pain reports range from changes within minutes to improvement over weeks or months, so no individual timeline can be promised.
What are the downsides of mirror therapy?
Some people experience new pain, altered sensation, dizziness, sweating, confusion, or emotional distress. CRPS symptoms can occasionally worsen or spread to the unaffected side. Pause the exercise and contact a rehabilitation clinician or doctor if symptoms appear or worsen.
