Phantom pain is caused by changes in the brain and nervous system that continue generating pain signals after a limb is lost or a nerve pathway is severed. Roughly 64% of people who undergo amputation experience phantom limb pain, and some estimates place that number closer to 72%. It is not imagined or psychological. It arises from measurable, physical changes at multiple levels of the nervous system, from the cut nerve endings at the amputation site all the way up to the brain’s sensory maps.
How the Brain’s Body Map Gets Rewritten
Your brain maintains an internal map of your entire body, spread across a strip of tissue called the somatosensory cortex. Each body part has its own dedicated zone. When a limb is amputated, that zone doesn’t go silent. Instead, neighboring zones begin creeping into the vacated territory, a process called cortical reorganization.
Research on upper-limb amputees illustrates this clearly. In people with phantom pain, the brain region that processes face sensations physically shifts toward the area that once processed hand sensations. The degree of that shift directly predicts pain intensity, with a strong correlation (r = 0.83) between how far the face zone migrates and how severe the phantom pain becomes. Similarly, the brain’s motor map for the muscles near the amputation site becomes enlarged and more excitable on the affected side compared to the intact side.
This remapping creates a mismatch. The brain still expects input from the missing limb, and when signals from adjacent body regions spill into the old territory, they can be misinterpreted as pain, tingling, or cramping in a limb that no longer exists. Amputees with no phantom pain show significantly less of this cortical reorganization, reinforcing the link between map distortion and pain.
What Happens at the Nerve Endings
The brain isn’t the only source. At the amputation site itself, severed nerves attempt to regrow but have nowhere to go. Without a target to reconnect to, these nerve fibers form tangled clusters called neuromas. Every severed nerve without a distal end will form one. A neuroma is essentially a ball of disorganized nerve tissue that can fire spontaneously, sending bursts of electrical activity up the spinal cord and into the brain.
Not every neuroma causes pain, and researchers still don’t fully understand what determines whether a particular neuroma becomes symptomatic. But when they do fire, they can produce shooting, burning, or electric-shock sensations that the brain interprets as coming from the missing limb. Some surgical techniques now aim to give transected nerves a physiological target to reinnervate, reducing the chance of a painful neuroma forming in the first place.
The Neuromatrix: Pain Without Input
One of the most important shifts in understanding phantom pain came from a theory developed by neuroscientist Ronald Melzack. The neuromatrix theory proposes that pain is not simply a response to signals arriving from the body. Instead, the brain contains a widely distributed neural network, the “body-self neuromatrix,” that generates its own characteristic patterns of nerve impulses. These patterns can be triggered by sensory input, but they can also arise entirely on their own.
This explains why phantom pain can persist even when peripheral nerves are blocked or neuromas are surgically removed. The brain has its own pain-generating circuitry, shaped partly by genetics and partly by a lifetime of sensory experience. Somatic sensory input from the body is only one of several influences converging on this network. Stress, emotional state, and attention all feed into the same system, which is why phantom pain often flares during periods of anxiety or fatigue.
Risk Factors That Increase the Odds
Not everyone who loses a limb develops phantom pain, and several factors shift the likelihood. The strongest predictor is persistent pain before the amputation. If the limb was painful for weeks or months before surgery, the brain’s pain circuits are already sensitized, and they tend to keep firing after the limb is gone. This suggests the nervous system essentially “learns” the pain pattern and replays it.
Other identified risk factors include the location of the amputation (more proximal amputations, meaning those closer to the trunk, carry higher risk), lower-limb amputation compared to upper-limb, ongoing pain in the residual stump, and the presence of non-painful phantom sensations like tingling or a sense that the missing limb is still there. Having phantom sensations doesn’t guarantee phantom pain, but it signals that the brain is still actively representing the absent limb, which increases vulnerability.
Phantom Pain Without Amputation
Phantom pain isn’t exclusive to amputees. People with spinal cord injuries can experience phantom sensations and pain in limbs that are still physically present but paralyzed and numb. In these cases, the spinal cord injury interrupts the flow of sensory information to the brain, creating a similar “deafferentation” effect. The brain loses its expected input and begins generating pain on its own, just as it does after amputation.
Similar phenomena occur after strokes, in certain types of epilepsy, and following nerve avulsion injuries where nerves are torn from the spinal cord. In one documented case, a patient with a spinal cord injury experienced not only phantom pain in his paralyzed legs but the sensation of a supernumerary (extra) phantom limb. These cases reinforce that phantom pain is fundamentally a disorder of how the brain processes body representation, not simply a wound-healing problem at an amputation site.
How Phantom Pain Is Managed
Because phantom pain has multiple causes operating at different levels of the nervous system, no single treatment works for everyone. The most well-known approach is mirror therapy, in which a mirror is positioned so the reflection of the intact limb appears where the missing limb would be. The idea is that by restoring a visual representation of the missing limb, the brain’s distorted sensory and motor maps may begin to normalize. The combination of visual feedback, sensation from the intact limb, and motor signals from both sides is thought to be the active ingredient.
The evidence for mirror therapy, however, is weaker than its popularity suggests. A 2023 systematic review of randomized, placebo-controlled trials found that only one out of five included studies showed a significant benefit of mirror therapy over a control condition, and that was at a single time point (four weeks). The review concluded that the current evidence does not allow firm claims that mirror therapy reduces phantom pain intensity, frequency, or duration. It may help with certain daily living activities over longer follow-up periods, but the overall picture is inconclusive.
Other approaches target different parts of the problem. Medications that calm overactive nerve signaling can address peripheral contributions. Techniques like transcutaneous electrical nerve stimulation (TENS) aim to override abnormal signals from neuromas. Graded motor imagery programs work on the brain’s body maps through a sequence of mental exercises. And because stress and emotional state feed into the neuromatrix, psychological approaches that reduce overall nervous system arousal can sometimes lower pain levels. Most people with phantom pain end up using a combination of strategies, adjusted over time based on what their nervous system responds to.