Why Chiropractic Adjustments Don’t Hold

The adjustment felt amazing. For about fourteen hours.

You woke up the next morning and your neck was right back where it started. The thoracic spine. The shoulder. The low back. Whatever the chiropractor mobilized yesterday afternoon had quietly returned to its previous position by the time you finished your coffee.

So you went back. And the adjustment felt good again. And it wore off again. And you went back again.

This is not a story about a bad chiropractor. This might be the most important thing I say in this article: the adjustment itself is doing exactly what it claims to do. The problem is not what happens on the table. The problem is what happens after you leave the office.

What the Adjustment Actually Does

A chiropractic adjustment produces real neurological input. This is not disputed.

When the chiropractor applies a high-velocity thrust to a spinal segment, three things happen. The joint capsule is mobilized. Mechanoreceptors in the paraspinal tissues fire a burst of novel sensory information into the nervous system. And the pain gate temporarily closes, reducing the pain signal from that area [1].

The pop you hear during the adjustment is cavitation. Gas released from the joint capsule under rapid pressure change. It sounds dramatic. The neurological input it produces is real. Pickar’s research documented this clearly: spinal manipulation stimulates mechanoreceptors and produces reflexive changes in muscle tone and pain perception [1].

So the adjustment changes something. Joint position shifts. Muscle tone around the segment drops. Pain decreases. You stand up and you feel different. Taller. Looser. More open.

For about fourteen hours.

Why It Reverts

Your brain maintains an internal model of where every joint in your body should be positioned. This model is called the [body schema](/body-schema-posture-how-brain-controls) [3]. It runs continuously. It generates your posture as its output. Not your muscles. Not your bones. Not your willpower. The brain’s prediction of where each segment belongs is what produces the position you see in the mirror.

The adjustment temporarily moves a segment to a new position. This creates what neuroscience calls a prediction error. The joint is now somewhere the brain did not predict it would be [2].

Here is the problem. Your brain has been running its current prediction for years. Sometimes decades. It has very high confidence in that prediction. The adjustment produces a single event of contradictory evidence. One moment where the joint is in a different position than the model expected.

The brain compares the evidence from the adjustment against its own confidence in the existing model. The model has been confirmed thousands of times. The adjustment contradicts it once. The brain resolves the mismatch by regenerating the original position [2][4].

This is not stubbornness. This is how predictive processing works. The brain updates its model only when the incoming evidence outweighs the confidence of the existing prediction. A single adjustment event, no matter how skilled, does not outweigh a prior built from years of consistent positioning.

The adjustment moved the hardware. The software reverted.

Chiropractic adjustments produce real, measurable neurological effects. Research by Pickar (2002) demonstrated that spinal manipulation stimulates mechanoreceptors in paraspinal tissues, producing reflexive changes in muscle tone and pain perception. The adjustment temporarily changes joint position and creates a burst of novel sensory input. However, systematic reviews including a Cochrane update by Rubinstein et al. (2012) consistently show that the benefits of spinal manipulation diminish over follow-up periods. The neuroscience explains why. The body schema, the brain’s internal model of where each joint should be positioned (Paillard 1999), generates joint position as a continuous prediction. The adjustment temporarily changes the position, creating a prediction error. But the brain’s confidence in its existing model, built over months or years of habitual positioning, outweighs the single adjustment event. In the framework of predictive coding (Friston 2010), the prior has more precision than the evidence. The brain resolves the mismatch by regenerating the original position rather than updating the model. The adjustment moved the hardware. The software reverted.

The Muscles That Pull It Back

There is another layer. Thomas Hanna identified a pattern called Sensory Motor Amnesia: when the brain loses voluntary control over muscles it has been holding for months or years [5]. The muscles become involuntary. Not paralyzed. Involuntary. The brain is still activating them. You just cannot feel it happening and you cannot stop it.

The muscles surrounding the adjusted segment are operating under this involuntary control. The adjustment mobilizes the joint. The moment you stand up, the SMA-held muscles begin pulling the segment back to the position the brain’s model specifies. You did not tell them to do that. You cannot tell them to stop. They are executing a prediction you do not have conscious access to.

This is [why posture keeps going back](/why-posture-keeps-going-back) regardless of the intervention. The instruction set has not changed. The muscles are following orders from a model that was never updated.

Chiropractic adjustments can temporarily improve joint position and reduce pain, but they do not change the nervous system prediction that generates postural patterns. Hodges and Moseley (2003) showed that chronic pain and compensation reorganize the brain’s motor strategy centrally, not at the joint level. The brain decides which muscles to recruit, at what tension, and in what pattern. This central strategy generates the joint positions that adjustments temporarily change. Thomas Hanna identified a key mechanism: Sensory Motor Amnesia, where chronically held muscles lose voluntary cortical control. The muscles that pull the joint back to its pre-adjustment position operate under involuntary nervous system control. The adjustment mobilizes the joint. The involuntarily held muscles return it to the position the brain’s model specifies. This is not a failure of the adjustment. It is the brain’s prediction reasserting itself. For the adjustment to hold, the brain’s model of that joint’s resting position must update through a different kind of intervention.

The Cycle

Marcus did chiropractic twice a week for a year. “It felt good in the office. By the next morning, everything was back.” He was not imagining it. The Cochrane data confirms his experience: spinal manipulation produces short-term improvements that diminish over follow-up [7].

Sofia liked how it felt. She hated that it wore off. Her chiropractor was good. The results just did not stick.

Both of them were experiencing the same mechanism. The adjustment produced a genuine change at the joint. The brain’s prediction regenerated the original position because nothing in the adjustment updated the model running the prediction.

This is [why nothing works](/why-nothing-works-for-posture) when the intervention targets the output instead of the model. [Stretching](/why-stretching-doesnt-fix-posture) targets the tissue. The adjustment targets the joint. Both change the downstream expression. Neither changes the upstream instruction.

The same principle applies to every therapy that changes the output without updating the model. The distinction is not which therapy. The distinction is which system the therapy is talking to.

What Would Make It Hold

I am not anti-chiropractic. Adjustments produce real neurological input. The joint mobilization is real. The mechanoreceptor stimulation is real. The temporary pain relief is real.

The question is what happens next.

The adjustment creates a brief window. The joint is in a new position. The brain has registered a prediction error. For a few minutes, there is an opening. If the person learns to attend to the new position during that window, the body schema has an opportunity to register the change. Sensation first. Then integration [3][6].

But that is not what typically happens. The person stands up. Walks to the front desk. Pays. Drives home. The window closes. The schema never registered the new position as anything other than a temporary interruption. The model reverts because it was never given a reason not to.

What would make the adjustment hold is not a better adjustment. It is a complementary intervention that addresses the prediction itself. Something that gives the brain evidence its model needs to update. Not a single event of contradictory joint position. Sustained, repeated, self-generated sensory input that the brain cannot ignore [2][4].

The difference between chiropractic that holds and chiropractic that does not is not the quality of the adjustment. It is whether the nervous system’s model of where that joint belongs gets updated in the hours and days that follow.

The frequency of chiropractic visits reflects how briefly the adjustment overrides the body schema. The brain’s postural prediction is maintained by a precision-weighted model (Friston 2010). This model has been running for years, sometimes decades, with very high confidence in its output. Each adjustment creates a temporary prediction error: the joint is in a position the brain did not predict. But the brain compares that momentary evidence against its established prior. The prior wins. The joint returns to its predicted position. The next adjustment creates the same temporary change, and the prior wins again. This is how precision weighting works in predictive coding. The evidence from a single adjustment does not outweigh a prior built from years of consistent positioning. What would change this is not more frequent adjustments but an intervention that addresses the model itself. The brain needs evidence that updates its prediction of where that joint should be, not just momentary changes to where the joint is.

The Model Is the Villain

Your chiropractor is not the problem. The adjustment is not the problem. The model of treatment that addresses the output without updating the prediction is the problem.

The adjustment moves the joint. The brain moves it back. Not because the adjustment failed. Because the adjustment succeeded at the wrong level. It changed the position. It did not change the prediction generating the position.

[Pandiculation](/what-is-pandiculation) addresses the prediction. Self-generated sensory input that the brain did not expect. Voluntary contraction followed by slow, conscious release. The nervous system registers the new information because you generated it. Not because someone else applied it.

The distinction is not manual therapy versus movement therapy. The distinction is whether the intervention talks to the output or talks to the model. The chiropractor can do brilliant work at the joint. The model does not live at the joint. The model lives in the brain’s prediction of where that joint belongs [3][6].

Your [neck pain](/chronic-neck-pain-keeps-coming-back) comes back because the prediction comes back. Your mid-back tightness returns because the model that generates the tightness was never revised. The adjustment is real. The reversion is also real. Both are doing exactly what the nervous system designed them to do.

The question is not whether to get adjusted. The question is whether the adjustment is the entire strategy or just one input into a larger process of updating the model.

If it is the entire strategy, you already know the answer. You have been living it every fourteen hours.

—

We work on the prediction, not the position. If the cycle described in this article is your cycle, [join the free community at posturedojo.com](https://www.posturedojo.com) where we teach how to update the model that keeps regenerating the pattern.

—

Sources

[1] Pickar, J.G. (2002). Neurophysiological effects of spinal manipulation. The Spine Journal, 2(5), 357-371.

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

[3] 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.

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

[5] Hanna, T. (1988). Somatics: Reawakening the Mind’s Control of Movement, Flexibility, and Health. Da Capo Press.

[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] Rubinstein, S.M., et al. (2012). Spinal manipulative therapy for chronic low-back pain: an update of a Cochrane review. Spine, 37(19), E1167-E1175.

—

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. Pickar, J.G. (2002). Neurophysiological effects of spinal manipulation. The Spine Journal, 2(5), 357-371. [T1]
   
   Spinal manipulation stimulates mechanoreceptors and produces reflexive changes in muscle tone.
2. Friston, K. (2010). The free-energy principle: a unified brain theory? Nature Reviews Neuroscience, 11(2), 127-138. [T1]
   
   Predictive coding: the brain resolves prediction errors by regenerating the prior.
3. 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 as the brain’s spatial model generating joint position.
4. Clark, A. (2015). Surfing Uncertainty: Prediction, Action, and the Embodied Mind. Oxford University Press. [T1]
   
   Precision weighting in predictive processing.
5. Hanna, T. (1988). Somatics: Reawakening the Mind’s Control of Movement, Flexibility, and Health. Da Capo Press. [T1]
   
   Sensory Motor Amnesia: involuntary holding patterns pull joints back to predicted position.
6. Hodges, P.W., & Moseley, G.L. (2003). Pain and motor control of the lumbopelvic region. Journal of Electromyography and Kinesiology, 13(4), 361-370. [T1]
   
   Central motor strategy reorganization persists beyond tissue healing.
7. Rubinstein, S.M., et al. (2012). Spinal manipulative therapy for chronic low-back pain: an update of a Cochrane review. Spine, 37(19), E1167-E1175. [T1]
   
   Spinal manipulation benefits diminish over follow-up.