Sciatica, Posture, and the Prediction Your Body Won’t Let Go Of

The pain runs down the back of your leg. You have been told it is your disc. It might not be.

Sciatica is one of the most over-attributed diagnoses in musculoskeletal medicine. The nerve is real. The compression is sometimes real. But in many cases, the pain pattern is a nervous system prediction that outlived the original structural event.

This article makes an honest distinction. Structural sciatica is real. Disc herniation compresses nerve roots. Stenosis narrows the canal. These conditions require medical evaluation and sometimes medical intervention. Nothing in this article replaces that evaluation.

But there is a second category. Predictive sciatica. The nervous system generating pain along the sciatic distribution without current compression. Real pain. Real suffering. Different source. Different treatment.

The treatment depends on which one you have. Most people are treated for the first when they have the second.

Your brain maintains an internal model of your body called the body schema [5]. That model includes predictions about pain. When the model predicts threat in a region, it generates a protective response [4]. Sometimes that response is bracing. Sometimes it is pain. The pain is real. The question is whether the source is current structural compression or a prediction that has not updated.

What Structural Sciatica Actually Is

The sciatic nerve is the largest nerve in the body. It exits the lower spine at the L4-S3 nerve roots, runs through the buttock (through or adjacent to the piriformis muscle), and travels down the back of the leg to the foot.

Structural sciatica occurs when something physically compresses this nerve or its roots. A herniated disc pressing on the L5 or S1 nerve root. Spinal stenosis narrowing the canal. A bone spur. Spondylolisthesis.

This is real. This is measurable. This requires imaging and medical evaluation.

Red flags that require immediate medical attention: Progressive weakness in the leg or foot. Bowel or bladder changes. Numbness in the saddle area (inner thighs, groin). These are signs of cauda equina syndrome and require emergency evaluation. Do not wait. See your provider.

For non-emergency structural sciatica, the findings are more complex than most people are told. A systematic review published in the American Journal of Neuroradiology examined imaging in people with zero symptoms [2]. Thirty-seven percent of 20-year-olds showed disc degeneration on MRI. By age 60, the number exceeded 90 percent. Disc bulges. Herniations. Degeneration. All present. No pain.

This does not mean disc herniations are harmless. It means imaging findings do not automatically equal pain generators. You can have a disc herniation and no sciatica. You can have sciatica and a disc herniation that is not the source.

A study in the New England Journal of Medicine followed patients with sciatica over time [3]. Many disc herniations resorbed spontaneously. Some patients whose herniations resolved still had pain. Some patients whose herniations persisted had no pain. The correlation between what the MRI shows and what the person feels is weaker than most people expect.

Posture and sciatica are connected through the nervous system’s compensatory patterns. When the brain’s internal model generates a bracing pattern (Friston 2010), the posterior chain activates as a unit. This includes the piriformis and deep external rotators through which the sciatic nerve passes. Chronic bracing of these muscles can compress the sciatic nerve mechanically. However, the bracing is generated by a nervous system prediction, not by a muscular weakness (Paillard 1999). Research by Hodges and Moseley (2003) demonstrated that the nervous system reorganizes its motor strategy under pain, and this reorganization persists even after the original cause resolves. Posture does not “cause” sciatica in the simple mechanical sense. The nervous system’s postural prediction can create the muscular bracing pattern that compresses the nerve, and that pattern can persist independently of the original structural event.

What Predictive Sciatica Looks Like

David has pain that shoots down his left leg. An MRI two years ago showed a small disc bulge at L4-L5. His orthopedist said the bulge was mild and not surgical. His physical therapist gave him McKenzie extensions and piriformis stretches. The pain improved for two weeks after each visit. Then it returned.

Three practitioners. Three explanations. None of them mentioned the possibility that his nervous system was generating the pain pattern independently of the disc.

Pain neuroscience research has established something that changes the conversation [1]. Pain is not a direct report from the tissue. Pain is generated by the brain as a protective output. The brain evaluates threat. If it predicts threat in a body region, it generates pain in that region. The tissue does not have to be damaged right now for the brain to generate pain right now.

David’s disc bulge was real two years ago. It may have resorbed. It may still be present. Either way, the pain pattern running down his leg follows the same distribution it followed during the acute episode. The brain learned that pattern. It predicted it. The prediction persists [4].

This is not “the pain is in your head.” The pain is in David’s leg. He can feel it. It is real. The source of the pain generation is the nervous system’s prediction, not current structural compression. Different source. Same experience. Different treatment.

Hodges and Moseley documented this mechanism [6]. The nervous system reorganizes its motor and pain strategy in response to an acute episode. The reorganization persists beyond the episode. The brain maintains the altered strategy because it was never given evidence that the threat resolved. The prediction runs. The pain keeps coming back.

How to Tell the Difference

This is where honesty matters more than theory.

You cannot diagnose yourself. The distinction between structural and predictive sciatica requires clinical evaluation. Imaging when appropriate. Neurological testing. A qualified provider.

But you can notice patterns that suggest which category you might be in.

Signs that suggest structural compression: Consistent pain that worsens with specific spinal positions (flexion or extension). Progressive weakness in the foot or ankle. Loss of reflexes at the knee or ankle. Pain that follows a specific, dermatomal distribution (L4, L5, or S1 territory) with precision. These warrant imaging and specialist evaluation.

Signs that suggest a predictive pattern: Pain that fluctuates with stress, sleep quality, or emotional state. Pain that responds temporarily to manual therapy, then returns. Pain distribution that is broader or less precise than a single nerve root. Imaging that does not correlate with the severity of the symptoms. Multiple treatments that produce temporary relief but no lasting change. An original acute episode that resolved months or years ago, but the pain pattern persists.

Both can coexist. A person can have a real disc bulge and a predictive pain pattern layered on top of it. The structural component may account for some of the pain. The predictive component may account for the rest. The treatment that addresses only the structural component misses the prediction. The treatment that addresses only the prediction misses the structure.

This is why sciatica is confusing. It is often both. The honest answer is to screen for the structural component first, address it if present, and then address the predictive component that remains.

Sciatica can be structural (disc herniation, spinal stenosis compressing the nerve root) or predictive (the nervous system generating pain along the sciatic distribution without current compression). A systematic review by Brinjikji et al. (2015) found that 37% of 20-year-olds show disc degeneration on MRI with no symptoms, demonstrating that structural findings do not reliably predict pain. El Barzouhi et al. (2013) showed in the New England Journal of Medicine that many disc herniations resorb spontaneously, and that pain resolution does not always correlate with imaging changes. Pain neuroscience research (Moseley 2003) establishes that the brain generates pain as a protective output based on its prediction of threat. When the original structural event resolves but the nervous system’s threat prediction does not update, pain can persist along the same distribution. This is predictive sciatica: real pain, generated by a real nervous system mechanism, without current structural compression.

—

This is the distinction we work with inside the Posture Dojo. Not replacing medical evaluation. Working with what remains after the structural component has been addressed. The prediction that keeps running. Learn how we approach it at posturedojo.com.

—

Why Stretching Does Not Resolve Predictive Sciatica

David stretches his piriformis three times a week. The stretch feels productive. The muscle releases temporarily. The pain returns within hours.

The stretch is a motor command. David’s brain predicts the result. The predicted sensation matches. No new information reaches the body schema [4]. The model does not update. The pain prediction persists because the prediction was never violated.

The piriformis and deep rotators brace as part of a nervous system pattern. The bracing is not a local muscular event. It is a descending signal from the nervous system organizing the entire posterior chain under threat. Stretching one muscle in that chain without addressing the signal that braces it produces temporary compliance. The signal reinstates the brace.

The same reason stretching does not fix posture is the reason stretching does not resolve predictive sciatica. Both are predictions. Both persist until the prediction receives evidence to update.

What evidence would update the prediction? Sensory input the brain did not expect [7]. Attention to the painful region without a motor command. Noticing the quality of the sensation without trying to change it. The brain receives information about the region that does not match its threat prediction. A small prediction error. Over time, the prediction begins to shift.

Pandiculation addresses the muscles the brain has lost control over [4]. Voluntary contraction of the braced pattern, followed by extremely slow conscious release. The contraction re-establishes cortical access to the held muscle. The slow release generates unpredicted sensory input. The brain learns something new about the region it was protecting.

This is not a replacement for medical care. If you have structural compression, you need structural treatment. But if the structural component has been addressed and the pain persists, the prediction is the remaining variable. And the prediction does not respond to stretching, strengthening, or injections. It responds to evidence.

Persistent sciatica often reflects a nervous system prediction that has outlived the original structural event. Moseley (2003) demonstrated that pain is generated by the brain as a protective output, and Hodges and Moseley (2003) showed that the motor reorganization associated with pain persists beyond pain resolution. The brain learned to brace the posterior chain and generate pain along the sciatic distribution during an acute episode. The acute episode resolved. The prediction did not update because the nervous system was never given evidence that the threat had passed (Friston 2010, Clark 2015). Stretching the piriformis or hamstrings generates a predicted sensation: the brain expects the stretch, and the resulting feeling matches that expectation. No new information reaches the model. The prediction persists. Effective intervention requires sensory evidence the brain did not predict, which is why attention-based and sensory-first approaches often succeed where repeated stretching and strengthening do not.

The Prediction Can Update

Sciatica is not a life sentence. Neither the structural kind nor the predictive kind.

Disc herniations frequently resorb on their own [3]. The body has mechanisms for resolving structural compression. Medical guidance helps navigate this process. The structural picture is often better than people expect.

Predictive sciatica responds to the same principle that governs all postural change. The prediction updates when the nervous system receives evidence it did not expect. Not more of the same treatment. Not harder stretching. Not stronger muscles around the nerve. Evidence that the threat has passed.

David’s path forward is not more piriformis stretches. It is screening to confirm the structural picture (his provider can update imaging if warranted), then addressing the predictive pattern that remains. Attention-based work. Sensory input the brain did not generate. The slow process of giving the nervous system a reason to stop generating a pain pattern it has been running for years.

The pain is not separate from the posture. Both are outputs of the same prediction. Change the prediction and both outputs shift. Not instantly. Not from one session. But the model is not permanent. It is a prediction. And predictions update from evidence.

—

Sources

[1] Moseley, G.L. (2003). A pain neuromatrix approach to patients with chronic pain. Manual Therapy, 8(3), 130-140.

[2] Brinjikji, W., et al. (2015). Systematic literature review of imaging features of spinal degeneration in asymptomatic populations. American Journal of Neuroradiology, 36(4), 811-816.

[3] el Barzouhi, A., et al. (2013). Magnetic resonance imaging in follow-up assessment of sciatica. New England Journal of Medicine, 368(11), 999-1007.

[4] Friston, K. (2010). The free-energy principle: a unified brain theory? Nature Reviews Neuroscience, 11(2), 127-138.

[5] Paillard, J. (1999). Body schema and body image: A double dissociation in deafferented patients. In G.N. Gantchev et al. (Eds.), Motor Control, Today and Tomorrow.

[6] Hodges, P.W., & Moseley, G.L. (2003). Pain and motor control of the lumbopelvic region: effect and possible mechanisms. Journal of Electromyography and Kinesiology, 13(4), 361-370.

[7] Clark, A. (2015). Surfing Uncertainty: Prediction, Action, and the Embodied Mind. Oxford University Press.

—

About the author: Sam Miller is the creator of Syntropic Core and founder of Posture Dojo. Diagnosed with an 85-degree scoliosis at 18, he spent two decades mapping the nervous system mechanisms that conventional treatment misses. He works with people whose bodies did not respond to the standard playbook. His approach is built on the predictive neuroscience of posture, not the mechanical model that failed him.

---

---

Sources

1. Moseley, G.L. (2003). A pain neuromatrix approach to patients with chronic pain. Manual Therapy, 8(3), 130-140. [T1]
   
   Pain neuroscience. Pain as a brain output, not tissue report.
2. Brinjikji, W., et al. (2015). Systematic literature review of imaging features of spinal degeneration in asymptomatic populations. American Journal of Neuroradiology, 36(4), 811-816. [T1]
   
   Disc findings in asymptomatic populations. Imaging does not reliably predict pain.
3. el Barzouhi, A., et al. (2013). Magnetic resonance imaging in follow-up assessment of sciatica. New England Journal of Medicine, 368(11), 999-1007. [T1]
   
   Disc herniations frequently resorb. Pain resolution does not always correlate with imaging.
4. Friston, K. (2010). The free-energy principle: a unified brain theory? Nature Reviews Neuroscience, 11(2), 127-138. [T1]
   
   Predictive coding. Pain as a prediction that persists beyond the structural event.
5. Paillard, J. (1999). Body schema and body image: A double dissociation in deafferented patients. In G.N. Gantchev et al. (Eds.), Motor Control, Today and Tomorrow. [T1]
   
   Body schema. Internal model maintains pain predictions for specific body regions.
6. Hodges, P.W., & Moseley, G.L. (2003). Pain and motor control of the lumbopelvic region: effect and possible mechanisms. Journal of Electromyography and Kinesiology, 13(4), 361-370. [T1]
   
   Motor reorganization under pain persists beyond pain resolution.
7. Clark, A. (2015). Surfing Uncertainty: Prediction, Action, and the Embodied Mind. Oxford University Press. [T1]
   
   Predictive processing. Pain along sciatic distribution can be a model output.