Virtual embodiment, and how it differs from mirror therapy.

What is virtual embodiment?

Virtual embodiment is the experience of a virtual body as your own. It has two parts that researchers measure separately. Body ownership is the feeling that the body you see belongs to you. Agency is the feeling that you are the one moving it. In a well-built VR system both arrive quickly: look down, see a body where yours should be, move your arm and watch the virtual arm follow, and within a few minutes the brain stops treating the image as a picture and starts treating it as you.

That shift is the starting point for virtual embodiment as a treatment. Once the brain has adopted a body, what it sees that body do becomes evidence about its own body. A virtual arm that reaches overhead without incident is information. A virtual back that bends further than the real one has bent in months is information. The therapy is built on feeding the pain system that kind of information deliberately, in doses, over weeks.

Where the idea came from

The lineage runs through two experiments from the 1990s. In 1996, Ramachandran and Rogers-Ramachandran reported that amputees watching the mirror image of their intact hand felt the phantom hand move, and that for some the painful clenched posture of the phantom eased. In 1998, Botvinick and Cohen showed that stroking a visible rubber hand in time with a person's hidden real hand made them feel the touch on the rubber hand and sense that it was theirs. Both findings point the same way: the brain's sense of what its body is doing can be edited through the eyes. Virtual reality generalized this. Slater and colleagues showed that people in immersive environments respond to virtual events as if they were real, and later work extended that to whole virtual bodies. The clinical question since then has been whether the same effect can be pointed at pain.

How is virtual embodiment different from mirror therapy?

Mirror therapy and virtual embodiment are the same idea delivered two ways. A mirror reflects what a real limb is doing. A headset renders what a virtual body is doing, and the rendering does not have to match reality. That one difference changes what can be treated, what can be controlled, and how much evidence stands behind each.

The row that matters most for patients is the setting. A mirror needs a limb that comes in a pair, so it has nothing to offer the lower back, the neck, or anyone whose pain is on both sides. Our mirror therapy guide explains that limit in detail. A rendered body has no such constraint, which is why most of the chronic back and neck pain work in this area uses VR rather than mirrors.

How can seeing a virtual body change movement or pain?

The short answer is that vision carries unusual weight when the brain decides whether a body part is safe, and virtual embodiment lets that input be controlled. The longer answer involves three findings, and it is worth being clear that the mechanisms linking them are proposed rather than established.

The rubber hand illusion

Botvinick and Cohen's 1998 experiment is the foundation. When a visible rubber hand was stroked in synchrony with the participant's hidden hand, people felt the touch on the rubber hand and reported it as part of their body. When the stroking was out of sync, the illusion mostly failed. The lesson is that ownership is built from agreement between senses. Give the brain matching vision and touch, or matching vision and movement, and it will fold a foreign object into its body map.

Body ownership in VR

A head-mounted display does the same job with a whole body. Because the virtual body sits where the real one is, moves when the person moves, and is seen from the person's own eyes, the agreement across senses is strong and the ownership illusion is correspondingly strong. Laboratory work has since shown that changing the virtual body's appearance, for instance making a limb look larger, smaller, or more transparent, can shift pain threshold or perceived pain in healthy volunteers and in some people with chronic pain. The 2019 review by Matamala-Gomez and colleagues collects this work and is candid that most of it is small and short-term.

Visual feedback and the pain decision

Pain is produced when the brain judges that a body part needs protecting, and that judgment draws on tissue signals, memory, expectation, and what the eyes report. In long-standing pain the judgment can become stuck, so that ordinary movements are flagged as threats long after tissues have settled. The proposed role of visual feedback is to supply evidence that conflicts with the stuck prediction. A painful back seen bending comfortably, over and over, is hard for a prediction of danger to absorb.

Karuna's own small mechanistic study, described on the mirror therapy page, found that showing people their unaffected arm's movement as if it were the painful arm reduced pain-free range of motion in the healthy arm. Belief about which limb was moving changed what the body would do. That is the effect virtual embodiment tries to run in reverse. Imaging work on mirrored feedback after stroke, analyzed in our research commentary, suggests the visual signal reaches motor areas through circuits that process observed action, which is consistent with the account above without confirming it.

Which conditions is virtual embodiment used for?

Virtual embodiment has been tried wherever a distorted body map is suspected of keeping pain or disability going. The conditions below are listed by how long the visual-feedback idea has been applied to them, not by strength of evidence, which is thin everywhere.

• [Phantom limb pain](/conditions/phantom-limb-pain). The original mirror therapy condition. VR versions render a virtual limb in place of the missing one and let the person move it, which a mirror can only approximate.
• [Complex regional pain syndrome](/conditions/complex-regional-pain-syndrome). CRPS often comes with a distorted sense of the limb's size and position. Virtual embodiment can adjust the visible limb, and it fits as a more flexible final stage of graded motor imagery.
• Chronic low back and neck pain. The largest group by patient numbers, and the one a mirror cannot serve. Most VR programs for back pain pair embodied movement with graded exposure and pain education.
• Fibromyalgia and widespread pain. Studied because the pain is both-sided and diffuse, which again rules out a mirror. The evidence here is early and small.
• Stroke rehabilitation. Here the goal is movement rather than pain. VR mirrored feedback has been tested as a way to engage the damaged hemisphere, and the distinction between embodied VR and screen-based games matters a great deal when reading those trials, as our commentary on VR versus videogames explains.

Across these, the common factor is a nervous system that has become protective of a body region beyond what the tissue currently warrants. That is the profile of kinesiophobia and of a sensitized pain system generally. Virtual embodiment is not a treatment for a fresh injury, a structural problem awaiting repair, or pain that has not yet been medically assessed.

What are the limits of the evidence?

The honest summary is that the illusion is well established and the treatment is not. Those are separate questions and they are often run together.

• Most studies are small and short. Much of the embodiment literature is single-session laboratory work in healthy volunteers, measuring pain threshold rather than clinical pain. A shifted threshold in the lab is not the same as a better month at home.
• Few trials isolate embodiment. Clinical VR programs bundle embodiment with graded exposure, breathing practice, and pain education. When a bundled program helps, it is rarely possible to say how much the embodied body contributed.
• Blinding is difficult. A person knows whether they are wearing a headset. Comparison conditions vary from no treatment to non-embodied VR, and the choice changes what a trial can conclude.
• Mechanism is inferred. Body-map remapping and prediction updating are reasonable explanations drawn from the rubber hand and mirror literatures. They have not been demonstrated as the cause of clinical improvement.
• Not everyone embodies easily. A minority of people feel little ownership of a virtual body, and a few find the experience unsettling or feel motion sick. Programs need a way to find those people and adjust.

This is why reviews in the field tend to use words like promising and preliminary, and why we do the same. Virtual embodiment is a reasonable thing to try for appropriately selected patients inside a program that is also doing the other things known to help. It is not a standalone treatment, and anyone describing it as proven is ahead of the data.

Pain that comes with new weakness or numbness, loss of bowel or bladder control, unexplained weight loss, fever, pain after a major injury, chest pain, or thoughts of self-harm needs prompt medical attention, and no visual-feedback approach is a substitute for that.

How does Karuna use virtual embodiment training?

Karuna Virtual Embodiment Training (KVET)™ is a 12-week program done at home, built around the mechanism described above. Patients use a VR headset that Karuna provides, pre-configured, for short embodiment exercises most days. In the headset they see and control a virtual body and work through movements of the painful region that are graded to start within tolerance and expand over the weeks.

The headset is one part of a larger structure. Each patient has a dedicated pain coach they meet one-on-one every week, who teaches the pain neuroscience behind the exercises, helps interpret flares, and adjusts the plan. A clinician reviews the case before anything starts and provides oversight throughout, so the program complements existing medical care rather than replacing it. The approach combines pain neuroscience education, graded exposure, and pain-reprocessing principles; embodiment is the medium those are delivered through.

Karuna has run institutional review board-approved studies on the technology itself, including validation that range-of-motion measurement in VR matches real-world measurement and the mirror visual feedback study mentioned earlier. Program outcome data, with the caveats that belong next to it, is on the how it works page. The program is drug-free, non-invasive, and HSA/FSA eligible, starts with a free consultation, and is backed by a money-back guarantee. For how it fits alongside other home-based options, see our guide to at-home chronic pain rehabilitation.

What this means for you

If you are living with chronic pain

Virtual embodiment is worth considering if your pain has been medically evaluated, has lasted longer than three months, and comes with movements you avoid because of what you expect them to do. It is a better fit for a sensitized system than for a problem still being diagnosed. If you have tried mirror therapy for a limb and found it helpful, embodiment is the same principle with more room to work. If your pain is in the back, neck, or on both sides, embodiment is the version of visual feedback that can actually reach it.

Expect gradual change. Early shifts usually show up as easier movement or less dread before pain scores move, and a flare during graded work is common and is not a sign of damage. Keep your prescriber in charge of any medication decisions.

If you are a clinician

Treat virtual embodiment as a delivery mechanism for graded exposure and body-map retraining, with mirror therapy and GMI as the evidentiary anchors. The relevant screening questions are the ones you already ask for exposure-based work: fear-avoidance beliefs, catastrophizing, and whether the clinical picture is consistent with central sensitization. When reading VR trials, check whether the intervention was immersive and embodied or a screen-based game, since pooled reviews often mix the two. Expect bundled interventions and uncontrolled designs, and weigh outcomes accordingly. Karuna accepts referrals and the care team confirms coverage case by case, including workers' compensation.