Can the Brain Unlearn Chronic Pain?
Evidence review & editorial synthesis
Chronic pain can change the brain. The unsettling question has always been what that means.
Does years of pain progressively damage the brain? Does the altered brain keep pain alive even after an injury has healed? Or can those changes disappear if the pain itself goes away?
A new study involving 35,052 people offers an unusually large clue. Persistent chronic pain was associated with widespread differences in brain structure. But among people whose chronic pain had resolved, most of those differences were no longer evident.
That fits an increasingly important idea in pain science: at least some brain changes accompanying chronic pain may represent plastic adaptation rather than irreversible damage. And if the nervous system can change in one direction, an obvious question follows.
Can it change back?
The longer pain persisted, the broader the brain differences
Researchers used UK Biobank data from people who had reported chronic pain at two assessments roughly nine years apart. Participants were classified as pain-free, newly experiencing chronic pain, persistently experiencing chronic pain or having recovered from previous chronic pain.
MRI measurements from the later assessment covered 313 measures of cortical thickness, cortical surface area and subcortical volume.
New chronic pain was associated with reduced surface area across sensorimotor, temporoparietal and prefrontal regions. Persistent pain was associated with a broader pattern that also included changes in the cerebellum, thalamus and basal ganglia.
Yet the resolved-pain group looked strikingly different. Most of the structural differences associated with current chronic pain were no longer present.
This does not literally prove that an individual's brain shrank during pain and regrew after recovery because MRI was available only at the second assessment. But across more than 35,000 people, the pattern is much easier to reconcile with reversible neuroplasticity than with simple permanent brain damage.
One exception remained. Reduced surface area in the primary somatosensory cortex persisted in the resolved-pain group. That region helps represent and interpret sensations from the body. Whether this residual difference is harmless, reflects a lasting history of pain or somehow contributes to future recurrence remains unknown.
Calling it a physical 'pain memory' would go beyond the evidence. But it raises exactly that question.
Pain can persist even when danger changes
Pain is not a direct readout of tissue damage. It is an experience created by the nervous system after integrating signals from the body with context, expectation, attention, previous experience and perceived threat.
Usually that system is enormously useful. A badly sprained ankle should hurt because protecting it promotes recovery.
But pain processing itself is plastic. Persistent nociceptive input can alter excitability, synaptic signaling, sensory processing, attention and brain networks involved in threat, emotion and body perception. In some chronic pain states, the relationship between current tissue damage and the intensity of pain can therefore become increasingly loose.
That does not mean the pain is imagined. The pain is real. The question is whether the nervous system has become unusually good at producing it.
Researchers sometimes describe parts of this process using terms such as maladaptive neuroplasticity or central sensitization. The exact mechanisms in humans remain debated, and no single process explains all chronic pain.
But the broader principle is increasingly difficult to ignore: the nervous system changes with prolonged pain.
One treatment asks the brain to reinterpret pain
Pain Reprocessing Therapy, or PRT, takes this idea unusually seriously.
For carefully selected people with primary chronic back pain, treatment aims to reduce the perceived danger attached to pain sensations. Patients learn to consider whether sensations that feel threatening may actually reflect non-dangerous nervous-system processing rather than continuing tissue injury. They gradually encounter sensations and movements with less fear while changing expectations about what the signals mean.
The results of one randomized trial were striking.
After four weeks, 66% of the 50 people assigned to PRT were pain-free or nearly pain-free, compared with 20% receiving an open-label placebo and 10% receiving usual care.
The difference did not immediately disappear. At five-year follow-up, among participants who provided follow-up data, 55% of the PRT group were nearly or completely pain-free, compared with 26% after placebo and 36% after usual care.
Those numbers come from a relatively small and selected chronic-back-pain trial, so they should not be generalized to every pain condition. But they are difficult to dismiss as a short-lived placebo effect.
The original trial also found changes in brain activity and connectivity. And in 2026, researchers examined structural MRI from the same research program. PRT increased gray matter in the anterior cingulate and orbitofrontal cortex compared with control groups. Clinical improvement was also associated with changes in prefrontal cortex, thalamus and amygdala.
Intriguingly, the specific treatment-induced gray-matter changes did not statistically explain why pain improved. The brain clearly changed alongside recovery, but researchers still cannot point to one structural alteration and say: this is how the pain was unlearned.
So can chronic pain really be unlearned?
For every form of chronic pain, no.
Pain from active inflammatory disease, cancer, progressive joint destruction or damaged peripheral nerves cannot simply be reclassified as a false alarm. Treating the biological source remains essential.
But chronic pain is not one disease. In some forms of primary or nociplastic pain, the nervous system's interpretation and amplification of signals may become an important part of the condition itself.
For those patients, 'unlearning' is a useful metaphor for a genuine biological possibility: repeated predictions of danger can potentially be updated, fearful responses can diminish and neural processing can reorganize again.
The new 35,000-person study does not prove that therapy can erase a pain memory. It shows something more fundamental.
The brain differences associated with chronic pain do not necessarily appear permanent.
Combined with randomized evidence that changing the interpretation of pain can produce large and durable improvement in some patients, that makes a once-radical possibility increasingly plausible: sometimes successful pain treatment may involve teaching the nervous system that it no longer needs to hurt.
Scientific References & Further Reading
- Anchor Study: “Brain gray matter changes across chronic pain trajectories: MRI evidence from 35,000 individuals.” (Zorina-Lichtenwalter K, Bango CI, Petre B, et al. Brain gray matter changes across chronic pain trajectories: MRI evidence from 35,000 individuals. Pain. 2026. doi:10.1097/j.pain.0000000000004125.)
- Pain Reprocessing Therapy Trial: “Effect of Pain Reprocessing Therapy vs Placebo and Usual Care for Patients With Chronic Back Pain: A Randomized Clinical Trial.” (Ashar YK, Gordon A, Schubiner H, et al. Effect of Pain Reprocessing Therapy vs Placebo and Usual Care for Patients With Chronic Back Pain: A Randomized Clinical Trial. JAMA Psychiatry. 2022;79(1):13-23. doi:10.1001/jamapsychiatry.2021.2669.)
- Five-Year Follow-Up: “Pain Reprocessing Therapy vs Placebo and Usual Care for Patients With Chronic Back Pain: 5-Year Follow-Up of a Randomized Clinical Trial.” (Ashar YK, Low EL, Knight K, et al. Pain Reprocessing Therapy vs Placebo and Usual Care for Patients With Chronic Back Pain: 5-Year Follow-Up of a Randomized Clinical Trial. JAMA Psychiatry. 2025;82(10):1049-1051. doi:10.1001/jamapsychiatry.2025.1844.)
- PRT Brain Structure Study: “Are gray matter changes following pain reprocessing therapy associated with clinical improvement? A voxel-based morphometry mediation analysis.” (Murillo C, Wager TD, Ashar YK. Are gray matter changes following pain reprocessing therapy associated with clinical improvement? A voxel-based morphometry mediation analysis. Journal of Pain. 2026;40:105644. doi:10.1016/j.jpain.2025.105644.)
- Mechanistic Context: “Is chronic pain caused by central sensitization? A review and critical point of view.” (Velasco E, Flores-Cortés M, Guerra-Armas J, et al. Is chronic pain caused by central sensitization? A review and critical point of view. Neuroscience & Biobehavioral Reviews. 2024. doi:10.1016/j.neubiorev.2024.105886.)
Disclaimer: This article is for informational purposes only and does not constitute medical advice. Chronic pain has many different causes, and treatments aimed at pain processing are not appropriate substitutes for evaluating or treating ongoing injury, inflammation, neurological disease or other medical causes.

