Not Everyone Has The Gene. Here’s What’s Actually Generating the Curve.

Part 2 of a 2-part introduction. Part 1: Is Scoliosis Genetic? The Genes Point Beyond the Spine (the article referenced in the opening). For the full synthesis, see The Scoliosis Research That Points to the Nervous System.

In the last article, I traced a chain from a single gene to a curved spine. POC5 builds cilia in the inner ear. Mutate it, and the vestibular hardware that detects gravity is malformed. The sensing tilts. The spine follows. LBX1 helps build sensory interneurons in the spinal cord, some of which carry position signals from the trunk. Impair that wiring, and the body may not detect the asymmetry as it develops.

That article mattered because of where it pointed. One of the strongest gene signals in scoliosis sits at LBX1, a gene that builds sensing cells, not bone. Other replicated genes shape the spine’s own cartilage and growth plates, so the spine is part of the story too. But sensing is in it from the start.

But genetics is only one door into the building.

If you read that article and thought, “I don’t have a genetic mutation, so this doesn’t apply to me,” stay here. This is the article where the model opens up. Because the mechanism that produces a scoliotic curve is broader than any single cause. It is a loop. And the loop has more than one door.

What causes scoliosis if it’s not genetic?

The short answer, as the hypothesis proposes it: during growth the spine drifts a tiny amount each month. With clear body sense, the nervous system catches the drift and corrects it. With blurry body sense, the drift is too small to register, so nothing corrects it, while the brain’s map keeps quietly updating to match the body. Month by month, the curve becomes the map’s new “straight.”

This is the neural generation hypothesis.

The current version proposes that the part of the curve a person can actively correct “is maintained by the body schema (the brain’s non-conscious and precision-weighted predictive model of the body)” and that along one entry route “degraded sensing can precede and help initiate the curve.”

The Neural Generation Hypothesis is a preprint, open on Zenodo (doi:10.5281/zenodo.19342098). Version 3.0 (October 2026) sets out the mechanism in full, with written tests that could show it wrong.

The genetic evidence from the previous article is one traceable route into this model. It points to one chain: gene > sensing organ > degraded signal > asymmetric output > curve. But the model does not require a gene to activate. It requires degraded sensory input during growth. And plenty of things can degrade sensory input.

Genetics is one. Trauma is another. Developmental issues are another. Chronic postural habits are another. Vestibular damage, proprioceptive deficits, visual dysfunction. All of them converge on the same mechanism. Different doors, same room.

The rate mismatch: why adolescence is the danger zone

One concept ties everything together.

During adolescent growth spurts, the vertebral column grows fast. In some kids, staggeringly fast. The spine can add centimeters in months. But the nervous system that maps and manages that spine does not grow at the same rate.

The spinal cord, the vestibulospinal pathways, the proprioceptive networks that track trunk position, all of these develop on their own timeline. This is what researchers call “uncoupled neuro-osseous growth.” The bones and the nerves are not synchronized. They are running on different clocks.

In most adolescents, this mismatch is small enough that the nervous system compensates. It updates its body schema frequently enough, accurately enough, to keep pace with the skeletal changes. The map stays close enough to the territory.

But if the sensory systems feeding that map are already compromised, already degraded by genetics, by trauma, by vestibular dysfunction, by anything, then small drifts go uncorrected. The map absorbs them as normal. The body schema moves with the curve.

And an inaccurate body schema does not just fail to correct asymmetry. It actively generates it.

The nervous system organizes muscle activation based on its model of where the body is. If the model says the body is straight when it is not, the motor output will maintain the curve. If the model says vertical is tilted five degrees to the left, every righting reflex, every postural adjustment, every unconscious shift will organize around that tilted reference.

The curve is not a failure of structure. It is the faithful output of a map that no longer matches the territory.

The curve attractor loop

The rate mismatch explains how the curve starts. The curve attractor loop explains why it accelerates.

Once a small asymmetry begins, a cascade locks it in place, and each step of the cascade feeds the next.

Asymmetric loading changes bone growth. This is the Hueter-Volkmann law, one of the most established principles in orthopedic biomechanics. When one side of a vertebral growth plate is compressed more than the other, the compressed side grows slower. The tension side grows faster. Asymmetric muscle activation produces asymmetric spinal loading. Asymmetric loading, sustained through a growth spurt, produces vertebral wedging. The bone itself begins to encode the curve.

The curve degrades proprioception further. As the spine curves, the proprioceptive map of the trunk loses resolution. This is cortical smudging. The brain’s representation of the curved region becomes blurry, less differentiated. The person literally cannot feel the asymmetry as clearly as someone with a straight spine can feel their symmetry. The sensor that should detect the problem is degraded by the problem itself.

The body schema accepts the new shape as baseline. The gamma efferent system, the neural mechanism that sets muscle tension baselines, recalibrates to the curved position. The asymmetric contraction pattern that maintains the curve becomes the new default. The muscles are not “tight” or “weak” in the conventional sense. They are following an updated set point. The nervous system has decided this is normal.

The pattern becomes invisible to the person living in it. This is why people with scoliosis often say they feel straight. They are not in denial. Their body schema genuinely reports straightness because the schema itself has been updated to match the curve. The measuring instrument has been recalibrated to the wrong standard.

The loop feeds itself. Degraded sensing produces asymmetric output. Asymmetric output produces asymmetric bone growth. Asymmetric bone growth further degrades sensing. The sensing degradation means the drift never gets detected, so it never gets addressed. The curve deepens. The attractor strengthens.

Attractor is a technical term here, not a metaphor: in the language of nonlinear dynamics, the curve is an attractor state, and the basin of attraction gets deeper with every cycle of the loop. The longer the loop runs, the harder it is to escape.

Why this explains what nothing else does

The curve attractor model answers questions that the standard structural model cannot.

Why are 80% of cases “idiopathic”? Because the cause is the rate mismatch, and the rate mismatch is invisible on X-ray. You cannot see a degraded body schema on a Cobb angle measurement, and you cannot see a vestibular asymmetry on a spinal film. The diagnostic model tracks the shape, not the generator. The cause looks unknown because the tools are pointed at the wrong layer.

Why do curves progress during growth and stabilize after skeletal maturity? Because the rate mismatch resolves when growth stops. Once the vertebral column is no longer changing faster than the nervous system can map it, the primary driver of curve progression disappears. The curve stabilizes not because the problem was fixed, but because the window of vulnerability closed. The attractor is still there. It just stops deepening.

Why are girls affected 8:1 at severe curve magnitudes? Girls enter puberty earlier and have a longer, more gradual growth period. The window during which the rate mismatch can operate is wider. There is also emerging evidence that hormonal factors affect proprioceptive sensitivity, potentially degrading the sensing layer during the exact period when it needs to be sharpest. The gender disparity comes down to timing and sensing, not bone weakness.

Why does early intervention work better? Because the attractor is shallower early on. The loop has not had time to deepen, the bone has not yet permanently wedged, the cortical smudging is not yet severe, and the gamma efferent recalibration is not yet fully consolidated. Interrupt the loop early, and there is less to undo. Wait, and the attractor basin deepens with every growth cycle.

Why does “idiopathic” persist as a label? Because the standard diagnostic model was built to describe shapes, not to identify generators. “Idiopathic” names a limitation of the framework, not a mystery of the body.

The many doors into the same room

The genetic entry point was covered in detail in the previous article. But the neural generation hypothesis identifies at least six entry points into the curve attractor loop. All of them degrade the sensory input that feeds the body schema. All of them can initiate the rate mismatch.

Genetic. POC5 and LBX1 degrade vestibular and proprioceptive hardware from development. The sensing organs are structurally impaired before any curve begins. This was the focus of the previous article because it provides the most measurable, traceable chain. You can sequence the gene. You can image the malformed vestibular canal. You can trace the full path from mutation to sensing deficit to asymmetric output to curve.

Vestibular. Any asymmetry in the otolith organs tilts the gravitational reference frame. This does not require a genetic mutation. Developmental variation in semicircular canal geometry, birth trauma affecting the temporal bone, chronic inner ear infections during critical developmental periods, all of these can create a vestibular bias.

The animal evidence points the same way. Lambert and Straka (2009) removed vestibular organs from one side of frog larvae. The frogs developed scoliosis. Full lateral curvature, sagittal deformation, vertebral rotation. The pattern matched human scoliosis.

Proprioceptive. Adolescents with idiopathic scoliosis have measurably impaired trunk position sense compared to controls, a finding that repeats across the research. They cannot feel where their trunk is in space as accurately. Some studies find the deficit doesn’t track curve size, which hints that it isn’t only a result of the curve. But every one of them measured it after the curve appeared, so whether it comes first is still an open question.

In VEMP testing, 38% of AIS patients have abnormal vestibular evoked responses. Subjective visual vertical testing shows they perceive gravitational vertical inaccurately. These are sensing findings, not structural ones. Whether they come before the curve in people is the question nobody has answered yet.

Trauma. This is the entry point that the standard model ignores entirely. Thomas Hanna documented the Trauma Reflex: a protective lateral flexion pattern that is triggered by injury, surgery, sustained emotional threat, or any significant physical insult. The reflex pulls one hip up, shifts the trunk laterally, and rotates the ribcage. It is mediated by the tectospinal tract and maintained by the gamma efferent system.

If the reflex is never resolved, it consolidates. The nervous system treats the contraction as normal. The asymmetric muscle activation becomes the baseline. During an adolescent growth spurt, this chronic asymmetric activation feeds directly into the Hueter-Volkmann mechanism. The bones grow into the shape the trauma reflex keeps producing.

The threat cascade that trauma initiates does not just affect emotions. It reorganizes the physical structure. This is one of the most common and least recognized entry points into scoliosis.

Developmental. Retained primitive reflexes, specifically the Asymmetric Tonic Neck Reflex (ATNR) and the Tonic Labyrinthine Reflex (TLR), create asymmetric postural patterns that should have integrated in infancy. When they persist, they impose a bias on the motor output that the developing nervous system organizes around. During rapid growth, that bias becomes the seed of a curve.

Nutritional. Vitamin D deficiency is correlated with scoliosis, and the mechanism may be dual. Vitamin D affects bone density, obviously, making the skeleton more susceptible to deformation under asymmetric loading. But it also affects brain morphology and neural development. The sensing and the structure are both compromised simultaneously. This is an area where the research is still emerging, but the convergence is suggestive.

All roads lead to the same place: degraded sensory input. Body schema mismatch. Asymmetric motor output. Asymmetric loading. Curve.

The entry point varies. The mechanism is the same.

Why the genetic evidence matters even if you don’t have the gene

In the previous article, I focused on POC5 and LBX1 because they give the most traceable chain. You can sequence a genome. You can image a vestibular canal. You can measure the structural asymmetry in the sensing organ. You can trace the chain from mutation to malformed cilia to degraded gravitational signal to asymmetric vestibulospinal output to curve.

Each link in that chain can be studied. That is what makes it a useful place to start.

The other entry points are harder to measure. You cannot easily quantify the degree to which a trauma reflex has degraded trunk proprioception. You cannot put a number on how much a retained ATNR is biasing motor output during a growth spurt. The logic is the same, but the measurement tools are less precise.

The genetic evidence makes the sensing route plausible. It doesn’t prove it, and the hypothesis is written to be tested. If the model holds, the other entry points use the same mechanism. They just enter through different doors.

This is why “is scoliosis genetic?” is both the right question and the wrong question. Yes, genetics is one entry point. No, it is not the only one. The deeper question is: what is the mechanism that all the entry points share? And the answer is: sensory degradation creating a rate mismatch during growth, locking the spine into a curve attractor loop.

What my diagnosis taught me about this model

I had a scan and a diagnosis at thirteen. Nobody tested my vestibular function. Nobody assessed my proprioceptive accuracy. Nobody asked about trauma history. Nobody checked for retained reflexes.

They measured the curve. At eighteen, a surgeon declined to operate because it was too risky, and I walked out believing I would just have to live with it.

As a teenager, we tried alternative things without really knowing what we were doing. After that, I lived with it. Nothing addressed what was generating the curve.

The loop kept running. The attractor kept pulling.

When I finally found the upstream layer, the sensing layer, things shifted that had not shifted in two decades. I gained two inches of standing height, measured on a doorframe at home. No imaging was taken, so nobody knows what changed in my spine, or whether it did.

The vestibular recalibration. The proprioceptive restoration. The resolution of bracing patterns that had been running since adolescence. The body schema updated because it finally received accurate information. My story is one case. It sent me to the question; it is not proof of the answer.

I do not tell this story to suggest that everyone’s curve will respond the same way. I tell it because it demonstrates the model. The curve had a generator. The generator had an address. When I found the address, the system could reorganize.

The curve has a generator. The generator has an address.

If you have scoliosis and you have been told the cause is unknown, you were given an incomplete model, not a wrong answer.

The neural generation hypothesis does not reduce scoliosis to one cause. It identifies one mechanism that multiple causes feed into. Genetics, vestibular dysfunction, proprioceptive deficit, trauma, developmental retention, nutritional factors. These are different entry points into the same loop. The loop is the curve attractor. And the curve attractor is what keeps the spine where it is.

The standard model sees the curve and asks how big it is. The neural generation model sees the curve and asks what is generating it. Those are fundamentally different questions, and they lead to fundamentally different answers about what is possible.

The curve has a generator. The generator has an address.

If you are ready to look upstream from the shape to the source, our scoliosis-specific assessment identifies your entry point into the attractor loop and builds a protocol around the generator, not the curve.

---

Sam Miller is the founder of Posture Dojo and the author of UPRIGHT. He has spent the last 9 years developing a nervous system approach to postural change. He found it.

---

Sources

– Lambert FM, Malinvaud D, Glaunes J, Bergot C, Straka H, Vidal PP. “Vestibular asymmetry as the cause of idiopathic scoliosis: a possible answer from Xenopus.” J Neuroscience. 2009;29(40):12477-83. – Lambert FM, Straka H. “Restricted neural plasticity in vestibulospinal pathways after unilateral labyrinthectomy as the origin for scoliotic deformations.” J Neuroscience. 2013;33(16):6845. – Machida M et al. “Characterization of scoliosis after pinealectomy in the chicken.” Spine. 1997. – Machida M et al. “Production of scoliosis after pinealectomy in chickens, rats, and hamsters.” Spine. 1998. – Kucerova K et al. “Vestibular functions of adolescents with idiopathic scoliosis: a comprehensive assessment.” J Manipulative Physiol Ther. 2025. – Cakrt O et al. “Subjective visual vertical in patients with idiopathic scoliosis.” J Vestib Res. 2011;21(3):181-6. – Simoneau M et al. “Sensory reweighting is altered in adolescent patients with scoliosis: evidence from a neuromechanical model.” Gait Posture. 2015;43:164-169. – Wang D et al. “Sensory integration dysfunction in adolescent idiopathic scoliosis.” J Clin Neurosci. 2025;134:111267. – Shi L et al. “Lateral semicircular canal asymmetry in idiopathic scoliosis.” PLoS One. 2015;10(7):e0131120. – Hanna T. Somatics: Reawakening the Mind’s Control of Movement, Flexibility, and Health. Da Capo Press, 1988. – Defined G et al. “Vestibular morphological alterations in adolescent idiopathic scoliosis: a systematic review.” Symmetry. 2023;15(1):175. – Cyr JP et al. “Enhanced vestibular-evoked balance responses in adolescents with idiopathic scoliosis.” J Neurophysiol. 2025.

---

---

Related research

• Is Scoliosis Genetic? The Genes Point Beyond the Spine
• Why 80% of Scoliosis Has No Explanation (Until Now)
• Schroth vs Functional Patterns vs Somatic Exercises: Which Works for Scoliosis?
• Functional Patterns for Scoliosis: What It Gets Right and What It Misses

Keep reading

• Is Scoliosis Genetic? The Genes Point Beyond the Spine
• Why 80% of Scoliosis Has No Explanation (Until Now)
• Schroth vs Functional Patterns vs Somatic Exercises: Which Works for Scoliosis?