Can you unlearn chronic pain?

Can chronic pain be unlearned?

For a large group of people, the research says yes. The word 'learned' is being used precisely. It describes pain that is kept going by a nervous system that has become sensitized and has come to expect threat from the body, rather than by damage the body is still reporting. Clinicians call this neuroplastic pain or nociplastic pain, and some use the older term psychophysiologic disorder. Whatever the label, the claim is the same: the mechanism is learning, and learning can be revised.

The clearest single test is a randomized trial of pain reprocessing therapy in 151 adults with chronic back pain lasting about a decade on average. After four weeks, 66% of the treated group reported being pain-free or nearly pain-free, against 20% after an open-label placebo injection and 10% with usual care, and a five-year follow-up found that much of the gain held. That is one trial, at one site, in one condition, with motivated volunteers. It is also the kind of result that does not happen if pain is a fixed readout of tissue.

The qualifications matter as much as the headline. Tissue, nerves and disease are real inputs, and some chronic pain is driven mainly by them: active inflammation, a progressive nerve compression, an undiagnosed illness. Those need a doctor before they need retraining, and retraining is not a substitute. Even in pain that fits the learned profile, some people improve a great deal, some a little, and some not at all. Nobody can tell you in advance which group you are in.

What does “learned pain” mean?

In the framework this series is built on, predictive processing, the brain does not wait for the body to report pain. It predicts what the body is about to feel, compares that prediction with the signals that actually arrive, and perceives something close to its own best guess. Pain is one of those guesses: a protective inference about whether the body is under threat, not a measurement of damage. This is a theory with growing but incomplete evidence, and this page uses it as a working model rather than a settled fact.

Three ideas from the model do most of the explanatory work.

• Priors are the brain's standing expectations. After months of a back that hurts when you bend, 'bending hurts' becomes a prior, and the brain starts producing pain at the first sign of a bend, sometimes at the thought of one.
• Precision is how much weight the brain gives to a prediction versus the incoming signal. In chronic pain the danger prediction tends to be held with high precision, so an ordinary stretch sensation gets read as the thing the brain was already expecting. Attention turns precision up and down, which is why where you put your attention is a treatment tool and not a distraction trick.
• Learning is the brain revising its priors when prediction and evidence disagree. The problem in chronic pain is that avoidance and guarding mean the evidence rarely arrives, so the prior is never tested. Active inference describes how the brain acts to confirm its own predictions, and in chronic pain that is exactly what protective behavior does.

It helps to think of this as a skill the brain has become very good at. A pianist who has practiced a passage with the wrong fingering for a year plays the mistake fluently, without deciding to. A brain that has produced pain in response to bending for a year does the same. Understanding that the fingering is wrong does not fix it; what fixes it is practicing the correct pattern, clumsily at first, until it is the one that runs on its own. The old pattern is never deleted. It is out-practiced, which is the plain meaning of neuroplasticity.

Two other threads feed into this. Expectation shapes pain directly, which is the mechanism behind the nocebo effect, and years of protective firing leave the spinal cord and brain more excitable, a process called central sensitization in which the nervous system amplifies its own signaling. The prediction and the physiology reinforce each other, and both are open to change.

How does retraining work, step by step?

No two programs run these steps in the same order or give them the same weight, and the names differ. Pain reprocessing therapy, pain neuroscience education paired with graded exposure, the approach described in Howard Schubiner's book Unlearn Your Pain, and Karuna's own VR embodiment program all belong to the same tradition. Underneath, the steps are recognizable.

1. Medical evaluation. A doctor rules out causes that need their own treatment: fracture, infection, inflammatory disease, tumor, progressive nerve compression. Retraining begins after this, never instead of it, and it continues alongside whatever medical care remains appropriate.
2. Education that changes what the sensation means. Pain neuroscience education explains how a protective system learns and why hurt does not reliably signal harm. On its own it moves fear more than pain, which is the point: the meaning of the sensation is what every later step works on. Our review of the education trials covers what it does and does not do alone.
3. Attention and precision training. Somatic tracking is the core exercise: attending to the sensation itself with curiosity and a sense of safety, describing it in neutral terms, and watching it shift. The aim is to lower the weight the brain gives to the danger prediction while that prediction is being tested against what is actually there.
4. Graded exposure to feared movement. The movements the person dreads are ranked and approached from the least threatening, in doses that feel manageable. Each repetition in which the feared outcome does not occur is a prediction error the brain can learn from. This is the direct treatment for kinesiophobia, and it is organized around fear rather than tissue load.
5. Working with emotion and trauma. High-precision threat priors often have a history. Stress, grief, unsafe relationships and past trauma shape how readily a nervous system reads the body as dangerous, so retraining that ignores them tends to stall. The work is trauma-informed and paced; trauma and chronic pain explains why.
6. Changing daily activity. Retraining has to leave the session. Returning to the chair, the commute, the garden and the sport, with planned pacing rather than pain-contingent pacing, is where the new pattern gets the repetitions it needs.
7. Consolidating a new self-model. The final stage is less about the back and more about the person: building a flare plan, practicing the new pattern under stress, and letting go of the identity that formed around protection. The next section explains why that last part is not optional.

Step four deserves a word about direction. People in chronic pain are usually told either to push through or to rest, and both fail for the same reason: neither one tests the prediction. Pushing through produces a flare that confirms the danger story, while resting never produces the evidence that would disconfirm it. Graded exposure sits in the narrow band between them, and finding that band for a particular person is most of what a good coach or clinician does.

Why is the sense of self part of unlearning pain?

Because the same predictive machinery that produces pain also produces the feeling of being a self with a body. In the predictive view, the brain builds its sense of 'me' from the stream of signals arriving from inside the body, from the ownership of the limbs it sees moving, and from the story it tells about who this person is and what they can do. None of that is handed to it. It is inferred, and it can be inferred differently. The series pillar on therapeutic neuroscience sets this out in full.

In chronic pain the self that gets built is organized around protection. The body is modeled as fragile, movement as risk, the day as something to get through. Over years this stops feeling like a belief and starts feeling like a fact about who you are: 'I am someone with a bad back.' That identity is adaptive while the threat is real. Once the pain is being maintained by prediction rather than damage, the identity becomes one of the things maintaining it, because a brain that believes it inhabits a broken body will predict pain from that body and perceive what it predicts.

This is why retraining that treats only the symptom tends to be fragile. The pain can drop while the self-model stays intact, and then a bad week, a new ache or a stressful month hands the old model exactly the evidence it was waiting for. What Karuna calls selfing flexibility is the alternative: loosening the fixed story of a damaged body so that a sensation can be noticed, described and let go of without it having to mean something about who you are. Like everything else here, it is practiced rather than decided.

In practice it is unglamorous: describing yourself to a new acquaintance without leading with the pain, or noticing the small surprise after a feared movement goes fine, or catching an 'I can't' and testing it once. Insight sometimes does part of this work in a single moment, when a person realizes with their whole body that the sensation was not what they thought it was. Mostly it is repetition.

How long does it take, and what does progress look like?

Weeks to months is the honest range, and anyone who gives you a precise number is guessing. The pain reprocessing trial delivered eight sessions over four weeks, and many of its responders shifted inside that window, but its participants had a single pain site, moderate pain and no complicating conditions. Long-standing pain, several pain sites, high fear or a trauma history usually mean a longer course. Karuna's program runs 12 weeks because that is roughly what consistent daily retraining seems to need; how it works describes the arc and reports the outcomes Karuna has measured, with their limits stated alongside.

Progress rarely arrives as pain going down first. The usual order is different.

1. Fear and confidence move first. A movement that was unthinkable becomes merely unpleasant, then ordinary. People often notice this before they would call anything 'better'.
2. Function follows. Standing long enough to cook, sitting through a meeting, a walk that used to be rationed.
3. Pain intensity moves last and least predictably. For some people it falls a long way. For others it falls modestly while life expands around it, and for some it does not change much at all.

Flares are part of the process rather than evidence against it. In the skill picture, a flare is the old pattern running again, usually under stress, fatigue or a sensation the brain had not yet filed as safe. It is uncomfortable and it is information: it shows which prediction is still held with high precision. The response that keeps progress going is a return to the practice at a tolerable dose, rather than a return to protection.

No promises are offered here, and you should be wary of anyone who offers them. The evidence supports a real possibility of substantial improvement for people whose pain fits this picture. It does not support a guarantee of it for any individual.

How does Karuna's program apply this?

Karuna Virtual Embodiment Training (KVET)™ is a 12-week, drug-free program done at home, and it maps onto the steps above with one addition. A clinician reviews each case before the program starts, to confirm the pain has been medically evaluated and that retraining is appropriate, and provides oversight throughout. A dedicated pain coach meets each person one-on-one every week to teach the neuroscience behind the exercises, help interpret flares, and carry what happens in the headset into sleep, work and daily movement.

The addition is the medium. Short daily VR sessions on a provided headset give the person a virtual body, seen from their own point of view, that moves when they move. The exercises are designed to create safe prediction errors around movement: the brain watches the painful region move further, more smoothly or more easily than it predicted, in a setting it has learned to treat as safe, and over weeks that first-person evidence is what updates the body map. Virtual embodiment therapy explains the principle, and our commentary on the embodiment research sets out how early the clinical evidence still is.

The later weeks are built for consolidation: expanding what feels safe from the headset into real-world activity, building a flare plan, and making sure progress continues after the program ends. Karuna takes workers' compensation referrals and the care team confirms coverage case by case; the program is also HSA/FSA eligible and backed by a money-back guarantee. The full arc, and the outcome data Karuna has measured along with its limits, are on how it works. A free consultation is the first step, and it includes being told if the program is not the right fit.

What does this mean for you?

If your pain has lasted longer than three months, has been medically evaluated, and the evaluation came back reassuring or found only changes that many pain-free people also have, then the possibility on this page applies to you. The pain is real, and it is probably being maintained by a nervous system that learned to protect you and never got the evidence it needed to stop. The steps above are learnable, with help, and the first of them, understanding what the sensation means, has already begun.

For clinicians, the implication is about sequencing and framing. Education that changes meaning has to come before graded exposure, or the exposure becomes a flare generator. Fear, function and confidence are worth measuring alongside pain, because they usually move first and they tell you whether the retraining is working. And the identity a patient has built around protection is a treatment target rather than background.

When not to start here

If the picture fits, take the neuroplastic pain quiz to see how many learned-pain features your pain shows, read pain reprocessing therapy for the clinic-based version of these steps, or talk with Karuna's care team about the VR version.