Why Massage Doesn’t Fix Chronic Tension (And What’s Actually Holding)

The relief lasts two days. You have been measuring.

Not consciously. Not with a calendar. But somewhere in the back of your mind, you know the schedule. Tuesday massage. Wednesday feels good. Thursday the tension starts creeping back. By Friday morning, the shoulders are exactly where they were before the appointment.

Sofia’s schedule: $120 every three weeks for four years. Her massage therapist is excellent. Technically skilled. Deeply attuned. The tension is still there. Every three weeks, the same muscles. The same knots. The same conversation between her body and someone else’s hands. The same temporary resolution.

She is not doing anything wrong. Her massage therapist is not doing anything wrong. The mechanism is doing exactly what it was designed to do. That is the problem.

What Massage Actually Does

I am not against massage. Massage produces real physiological effects. Let me be specific about what those effects are, because the real mechanisms deserve respect before I name the ceiling.

First: thixotropy. Your fascial tissue behaves like ketchup. Under sustained manual pressure, the ground substance in the fascia temporarily shifts from a gel state to a more fluid state [3]. The tissue becomes more pliable. This is not imagination. It is materials science. Schleip documented the mechanism clearly: the gel-to-sol transition in fascial ground substance under sustained pressure produces measurable tissue softening.

Second: H-reflex inhibition. When your massage therapist holds pressure on the junction between muscle and tendon, the stretch reflex temporarily turns down [4]. The muscle’s reflexive spring-back response decreases. You feel the release as the muscle stops resisting the pressure. This is a genuine neurological effect. The reflex arc is being temporarily inhibited.

Third: parasympathetic activation. Massage activates the body’s rest-and-recover response. Heart rate drops. Breathing deepens. Cortisol decreases. You feel calm. You feel safe. This is real and it matters.

Three mechanisms. All real. All temporary.

Massage produces real physiological effects through two primary mechanisms. First, thixotropy: sustained manual pressure temporarily changes the viscosity of fascial ground substance, making tissue more pliable (Schleip 2003). Second, H-reflex inhibition: pressure on the myotendinous junction temporarily suppresses the stretch reflex, reducing the muscle’s reflexive resistance to lengthening (Behm & Wilke 2019). Both effects are genuine and both are temporary. The brain maintains an internal model of what each muscle’s resting tone should be (Clark 2015, Friston 2010). This model, called the body schema, generates muscle tone as a continuous prediction. Massage temporarily overrides that prediction through mechanical means. But the prediction itself was never updated. Within 24 to 48 hours, the brain’s model reasserts and the muscle returns to its previous tension. The massage changed the output. It did not change the instruction set generating the output. This is why chronic tension returns on a predictable schedule regardless of how good the massage is.

Why It Reverts

Your brain runs an internal model of what each muscle’s resting tone should be. This model is called the body schema [1][5]. It is not a suggestion. It is an instruction. The brain generates muscle tone continuously, the way a thermostat generates temperature. The setting determines the output.

Massage temporarily overrides the output. The tissue softens. The reflex turns down. The tone drops. But the thermostat setting did not change. The brain’s prediction of what that muscle’s resting state should be is exactly the same as it was before the appointment.

Within 24 to 48 hours, the prediction reasserts. The tone returns. The tissue re-gels. The knot comes back. Not because the massage wore off. Because the instruction set that generates the tension was never updated [2][5].

This is the same reason stretching does not fix posture. The same reason nothing seems to work when the intervention targets the output instead of the model. The treatment changed the tissue. It did not change the brain’s prediction of what the tissue should be doing.

The Holding Pattern

Thomas Hanna identified something critical. Chronic muscle tension is not a tissue problem. It is a brain problem. He called it Sensory Motor Amnesia: when the brain loses voluntary control over muscles it has been holding for months or years [1].

The muscles are not short. They are not damaged. They are not stuck. They are being held at elevated tone by a nervous system that has decided this is the necessary resting state. The decision is involuntary. You cannot feel it happening. You cannot override it with willpower. You cannot override it with more effort.

Your massage therapist can feel the knot. She can work it until the tissue softens. But the moment the mechanical pressure is removed, the brain begins regenerating the tone. Because from the brain’s perspective, that tone is correct. It is the predicted resting state. The massage was the interruption. The tension is the plan.

The brain is not confused. The brain is executing a strategy. The strategy may be outdated. The strategy may be based on a threat that resolved years ago. But the brain has high confidence in the strategy because it has been running it successfully for a long time [2]. A single massage session, no matter how skilled, does not outweigh that confidence.

The nervous system maintains chronic muscle tension through a centrally generated holding pattern, not through local tissue shortness. Thomas Hanna (1988) identified this as Sensory Motor Amnesia: the brain loses voluntary control over muscles it has been holding for months or years. The tension is involuntary. The brain has decided this is the necessary resting state for that muscle and holds it there without conscious participation. Massage temporarily reduces the tone through mechanical tissue effects and reflex inhibition. But because the holding pattern is centrally maintained, the brain regenerates the tension as soon as the temporary mechanical effects wear off. Research in predictive coding (Friston 2010) explains the timeline: the brain’s model has high confidence in its current prediction of what that muscle should be doing. The massage temporarily contradicts that prediction at the tissue level. The brain resolves the contradiction by regenerating the predicted state rather than updating the model. The evidence from the massage is not the right kind of evidence to change the prediction.

Why the Brain Does Not Count Massage as Evidence

This is the part that matters. The brain updates its model only when something unexpected happens. Novelty. Prediction error. A sensory signal the brain did not anticipate [2][5].

During a massage, the parasympathetic activation opens the nervous system’s receptive window. That is one of the three conditions required for the body schema to update. The gate is open. The system is listening.

But listening for what?

The sensation during a massage is expected. You booked the appointment. You got on the table. You knew what was coming. The brain predicted pressure, received pressure, and confirmed its prediction. No prediction error. No update signal. The experience matches the model of what a massage feels like [2].

A meta-analysis across 25 studies confirmed the pattern: sustained pressure interventions produce temporary improvements in flexibility and pain that do not persist [4]. The mechanism is not wrong. It is incomplete. It changes the tissue without changing the brain’s model of what the tissue should be doing.

The brain needs evidence it did not predict. Self-generated sensory input. A movement it did not expect. A sensation that contradicts its current model of what that muscle’s resting state should be. Massage provides none of these. Massage provides exactly what the brain expects massage to provide. The brain has learned the pattern. It no longer treats it as new information.

Massage activates the parasympathetic nervous system, producing genuine relaxation, reduced pain perception, and reduced anxiety. These are real neurological effects. The parasympathetic activation also temporarily opens the nervous system’s receptive window, one of three conditions required for the body schema to update. But the other two conditions are not met. For the body schema to change its prediction of a muscle’s resting tone, the sensory evidence must be novel enough to outweigh the brain’s confidence in its current model. Massage sensation is expected, generating minimal prediction error. The temporary tissue change matches what the brain already knows massage does: tone drops briefly, then returns. The brain has learned the massage pattern. It no longer treats it as new evidence. Wiewelhove et al. (2019) confirmed across 25 studies that sustained pressure interventions produce temporary improvements in flexibility and pain that do not persist. The mechanism is not wrong. It is incomplete.

What Would Change the Model

The question is not whether massage effects are real. They are. The question is whether those effects update the model or just temporarily override it.

Override is what you have been experiencing. Two days of relief. Then the model reasserts. Then you book the next appointment. The cycle is not the massage therapist’s fault. The cycle is the logical outcome of changing the output without changing the instruction set.

Updating the model requires a different kind of input. Not applied from the outside. Generated from the inside. Sensory information your brain did not predict, arriving through movement you initiated, in a pattern the nervous system has to account for because it contradicts the current model [2][5].

This is the difference between someone else pressing on the muscle and you slowly contracting and releasing that same muscle while paying attention to the sensation. The external pressure is expected. The self-generated contraction-release produces a prediction error. The brain must reconcile the new sensory input with its existing model. If the evidence is precise enough and the system is safe enough, the model updates [1][6].

Not for two days. The model updates because the instruction changed.

We have covered the specific case of massage for neck pain elsewhere. The mechanism is the same whether the tension lives in your neck, your shoulders, your low back, or your jaw. The location changes. The principle does not. The brain generates the tone. The brain maintains the tone. The brain will regenerate the tone until its prediction of what that tone should be gets revised.

Sofia’s Math

$120 every three weeks. Fifty-two weeks per year. That is roughly $2,080 per year. For four years, that is over $8,000 spent on temporary tone reduction.

Sofia’s massage therapist is excellent. The sessions feel genuinely good. The parasympathetic activation is real. The stress relief has value that extends beyond the muscular effects.

But the tension is still there. The same tension. In the same muscles. After four years and $8,000.

That is not a story about bad massage. That is a story about a model of treatment that addresses the output without updating the prediction. The massage changes the muscle. The brain changes it back. Not because the brain is broken. Because the brain is doing its job. The prediction is the job. The tension is the output of the prediction. Change the prediction, the output changes on its own.

Sofia does not need a better massage therapist. She does not need deeper pressure or more frequent sessions. She needs an intervention that talks to the model generating the tension. Not the tension itself.

The muscle is not the problem. The muscle is the printout.

—

If you have been managing tension instead of resolving it, the model is the missing piece. Join the free community at posturedojo.com to learn how the brain generates chronic holding patterns and what it takes to update the prediction.

—

Sources

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

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

[3] Schleip, R. (2003). Fascial plasticity: a new neurobiological explanation. Journal of Bodywork and Movement Therapies, 7(1), 11-19.

[4] Wiewelhove, T., et al. (2019). A meta-analysis of the effects of foam rolling on performance and recovery. Frontiers in Physiology, 10, 376.

[5] Clark, A. (2015). Surfing Uncertainty: Prediction, Action, and the Embodied Mind. Oxford University 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] Behm, D.G., & Wilke, J. (2019). Do self-myofascial release devices release myofascia? Rolling mechanisms and effects on range of motion and performance. Current Sports Medicine Reports, 18(6), 233-240.

—

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. Hanna, T. (1988). Somatics: Reawakening the Mind’s Control of Movement, Flexibility, and Health. Da Capo Press. [T1]
   
   Sensory Motor Amnesia: the brain holds muscles involuntarily.
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 massage effects by regenerating the prior prediction.
3. Schleip, R. (2003). Fascial plasticity: a new neurobiological explanation. Journal of Bodywork and Movement Therapies, 7(1), 11-19. [T1]
   
   Thixotropy: fascial ground substance shifts from gel to sol under sustained pressure.
4. Wiewelhove, T., et al. (2019). A meta-analysis of the effects of foam rolling on performance and recovery. Frontiers in Physiology, 10, 376. [T1]
   
   Sustained pressure interventions produce temporary flexibility improvements.
5. Clark, A. (2015). Surfing Uncertainty: Prediction, Action, and the Embodied Mind. Oxford University Press. [T1]
   
   Precision weighting: the brain’s model outweighs single-session evidence.
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.
7. Behm, D.G., & Wilke, J. (2019). Do self-myofascial release devices release myofascia? Current Sports Medicine Reports, 18(6), 233-240. [T1]
   
   H-reflex inhibition through myotendinous pressure.