The research behind Posture Dojo
Posture Dojo's method grows out of research Sam Miller does in the open: on his own body, with the people he teaches, and across the scientific literature. He isn't a doctor or a physical therapist. He's an independent researcher and investigator, and he publishes his work so anyone can check it.
Who does the research
Sam Miller founded Posture Dojo in 2019. He had a scan and a diagnosis of scoliosis at 13. At 18 he went looking for surgery himself, and the surgeon told him it was too risky and to just live with it. He did, for about 20 years. For the last 9 of them he has been developing a nervous system approach to postural change, which he teaches as neurosomatic practice.
His authority is empirical, not clinical. It comes from years of looking closely at one problem from every side he could reach, and from writing down what he found in a form other people can test. You can find his published work on ORCID (0009-0000-4479-973X). More about Sam.
How the investigation works
His own body, as a long case study. At 33 Sam's body collapsed, and he came back to the problem with a different question. Not “how do I fix this shape?” but “why does my body keep generating it?” Since then he has gained about two inches of standing height, measured on a doorframe at home. No imaging was taken, so nobody knows what changed in his spine. One case can't prove a theory, and we never use his as proof. What it did was hand him the question his papers now ask.
The people he teaches. Live cohort sessions are recorded and studied: what members notice, where they get stuck, what shifts for them week by week. That shapes how the practice is taught. It stays out of the papers on purpose, because our members' reports can't be the test of our own idea.
The literature. Sam reads across fields that rarely talk to each other: the genetics of scoliosis, animal studies of position sense, balance science, pain science and predictive models of the brain. Then he puts them into one account and labels every claim by the strength of its evidence.
Auditing his own work. In October 2026 the team ran a link-by-link evidence audit of the Neural Generation Hypothesis, scoring each step of its chain the way a skeptical reviewer would. Version 3 of the paper was rebuilt from that audit, with sharper claims and tests that come with numbers. You can read the results, claim by claim, on our NGH and GPF fact check.
The two papers
| Neural Generation Hypothesis (NGH) | Proposes that the reducible part of a scoliotic curve, what active self-correction removes beyond passive flexibility, is kept going by the body schema, the brain's non-conscious predictive model of the body, and that degraded sensing can help start a curve along one route. Growth and loading set how large a curve gets, which NGH does not claim to explain. First posted 31 March 2026; version 3.0 posted 27 October 2026. doi.org/10.5281/zenodo.19342098 · SSRN doi.org/10.2139/ssrn.6493379 · Plain-language guide |
|---|---|
| Generative Posture Framework (GPF) | “Posture is generated, not held.” Proposes that the body schema generates posture as a running prediction, sets out the conditions under which that prediction updates, and offers testable predictions, each with a result that would count against it. First posted 14 April 2026; version 2.5 posted 27 October 2026. doi.org/10.5281/zenodo.19580170 · SSRN doi.org/10.2139/ssrn.6577078 · Plain-language guide |
Both are open-access preprints on Zenodo with permanent DOIs. Neither has been through journal peer review. They were written to be argued with, and both list the studies that could prove them wrong.
How we label evidence
Posture Dojo Research marks every claim T1, T2 or T3. On this site we use the same scale in plain words:
- Established science (T1): peer reviewed and independently replicated. We state it directly.
- Supported but still being studied (T2): real findings that are newer, from one group, or associations where cause and effect aren't sorted out yet.
- Our hypothesis, not yet tested (T3): our own synthesis. Reasoned from the evidence, open to testing, never stated as fact.
A claim with several parts gets a label for each part. That rule is where most of the honesty lives: a strong first step doesn't lend its strength to the steps that follow it.
How to check us
- A public corrections log. When a citation doesn't hold up, we fix the article and log it, dated, on the corrections page.
- Who pays for it. Posture Dojo Research is funded by Posture Dojo memberships. Not certification fees, brace revenue, referral fees or advertising.
- Our interest, stated plainly. We teach the method our research argues for, so we have a reason to want it to be right. That's why the papers are public and every claim carries its label.
- A directory that doesn't favor us. The Conservative Spine Care Navigator lists Schroth, SEAS and BSPTS therapists, bracing clinics and other conservative spine care. Nobody pays to be listed, and we don't profit from anyone in it.
What the research is for
Plenty of people with scoliosis or kyphosis have been told, one way or another, to just live with it. The research asks a more useful question: what keeps a pattern going, and which parts of that can a person work on themselves? Every answer so far is labelled for what it is, and the open ones come with the studies that would settle them. That's what makes them worth following.
Where to go next: the claims, rated one by one, the distinctions that change how you practice, and what practicing posture looks like day to day.
Questions about the research
Is Sam Miller a doctor or a scientist?
Neither title fits, and we don't use them. Sam is not a doctor, a physical therapist or a licensed clinician. He is an independent researcher and investigator: he studies his own body over the long run, studies what people report as they practice, reads across the scientific literature and publishes his reasoning with DOIs. His record is on ORCID. What he teaches is neurosomatic practice, education that works alongside medical care.
Are Sam Miller's papers peer reviewed?
No. Both are open-access preprints on Zenodo with permanent DOIs, also posted on SSRN. Anyone can read, cite and challenge them, which is why they are public. Each claim in them is labelled by how strong its evidence is, and we rate them claim by claim on our fact check.
Where can I read the Neural Generation Hypothesis and the Generative Posture Framework?
The Neural Generation Hypothesis is at doi.org/10.5281/zenodo.19342098 and the Generative Posture Framework is at doi.org/10.5281/zenodo.19580170. Those links always open the latest version. Plain-language guides to both are on Posture Dojo Research.
Does Posture Dojo use member results as scientific evidence?
No. Neither paper uses data from Posture Dojo's own practice. Member stories help us teach better and they show what people experience, but self-reported results from our own members can't test our own theory. That job belongs to independent, pre-registered studies, and the papers describe the ones that would do it.
How do I report a mistake in the research?
Email team@posturedojo.com. When a citation doesn't hold up, the article is fixed and the change goes in the public, dated corrections log on Posture Dojo Research.