What Biology Says

11. Fascia

The key word of the contemporary field, and the one with the greatest distance between claim and data.

What is documented

1. It is continuous. Fascia is not separate “wrappings” but a continuous network of connective tissue that pervades the whole body.

2. It is richly innervated — and this is the most important. Studies have shown a high density of free nerve endings in fascia, especially the thoracolumbar.

That is, fascia is, at least as much as a mechanical organ, also a sensory organ.

3. It contains contractile cells. There are myofibroblasts that can produce slow tension — but the forces are small and the timescale is tens of minutes to hours, not seconds. It does not explain what happens in a 60-second stretch.

4. It transfers force laterally. Part of a muscle’s force is transmitted through the fascia to adjacent structures, not only through the tendon.

5. It changes in chronic pain. Ultrasound studies have shown thickened and less mobile thoracolumbar fascia in people with chronic low back pain. (Cause or effect — open.)

What is not documented

1. “Release” by mechanical deformation. Chapter 9.

2. The “lines” as functional chains. The popular system of myofascial lines, circulating since 2001, is based on anatomical dissections and is partly anatomically supportable. But the functional claims — that a tension in the sole explains a pain in the neck through a specific line — have not been tested and do not constitute established anatomy. It is one author’s scheme.

3. “Fascia holds memory / trauma.” No documentation. It is a metaphor circulating as finding.

The honest summary

Fascia is genuinely important and under-studied for decades. And the therapeutic claims about it run far ahead of the data.

And the worst: it has become the new catch-all — what used to be explained with “energy” is now explained with “fascia”, with the same way of thinking and a slightly more scientific sound.

12. Pain is output

The most important chapter of the document.

If you keep one thing, keep this — because it rearranges the whole field, and is among the best-documented things in modern pain physiology.

The thesis

Pain is not a measurement of tissue damage. It is a protective output the brain produces based on the estimate of threat.

Two different things:

• Nociception — the nerve signal from stimulated receptors. Input.
• Pain — the experience. Output.

You can have one without the other, in both directions.

The data

1. Imaging does not predict pain. A large meta-analysis in asymptomatic people — with no pain at all — found:

• Disc degeneration in 37% of twenty-year-olds
• And in 96% of eighty-year-olds
• Disc bulges in a significant percentage of asymptomatic people

That is: what shows up on MRI is, in large part, normal ageing — and announcing it to someone in pain can make them worse.

2. Pain without damage. There is a famous case of a construction worker who arrived at emergency with a nail that had pierced his boot and in extreme pain. He needed sedation. When the boot was removed, the nail had passed between the toes. There was no damage.

The pain was real. The threat estimate was wrong.

3. Damage without pain. Serious injuries in battle or in sport, felt hours later.

4. Phantom limb pain. A limb that does not exist, pain that does.

And the therapeutic consequence — which is critical

If pain is produced by threat estimation, then:

Changing the estimation changes the pain.

And this is not theory: pain neuroscience education — explaining to the patient how pain works — has documentation from randomised trials as a supplementary intervention. Effect sizes are modest, but real and reproducible.

The explanation is the intervention.

And why it rearranges the field

Because it gives a plausible, non-metaphysical mechanism for nearly everything the field does:

If pain is a response to threat, then gentle touch, safety, attention, slow breathing, non-threatening new movement, and the sense of being understood are all mechanistically plausible interventions — without any need for energy, for fascial release, or for memory in the tissues.

The field already has the mechanism it is looking for. It is just not spectacular.

13. Interoception

Sherrington’s third term (chapter 2) — and the legitimate scientific home of a large part of this field.

What it is

The perception of the body’s internal state: heart rate, breathing, hunger, thirst, temperature, gut motility, muscle fatigue, nausea, sexual arousal.

The pathway is known: specialised fibres ascend from the body to the brainstem and thalamus, and end in the insular cortex — a region considered central to the synthesis of the sense of “how I am now”.

And the three dimensions that change everything

Modern research distinguishes three different things, and it is the most useful tool of the chapter:

And the finding is important: these dissociate.

People who believe they read their body excellently often do not.

And conversely.

This is directly continuous with chapters 1 and 3 — and is the experimental version of them: self-reported bodily awareness does not predict actual bodily awareness.

Clinical connections

• Anxiety — often excessive sensitivity to cardiac signals, with catastrophic interpretation
• Depression — often blunted perception
• Eating disorders — disturbed interpretation of satiety and hunger signals
• Alexithymia — difficulty naming internal states

The honest note

The classical measurement method — counting one’s own heartbeats without a pulse — has received serious methodological criticism: it is influenced by what you think your pulse is and by time-estimation ability.

The field is correcting this.

Why it matters

Because it gives the field a research programme that needs no mystification:

Somatic practices plausibly train the capacity to observe and correctly interpret internal signals. This is measurable, clinically relevant, and interesting.

And it is much more interesting than the explanations the field usually gives for itself.

14. Breath and the autonomic

The most boring and the most reliable part of the text.

What is solidly documented

1. Slow breathing, around six breaths per minute, maximises heart rate variability.

There is a “resonance frequency” where breath, heart rate and the baroreflex synchronise. The phenomenon is consistently reproducible.

2. Longer exhalation shifts toward parasympathetic dominance within the breathing cycle. Known mechanism.

3. Heart-rate-variability biofeedback has decent documentation for anxiety and stress, with modest effect sizes in meta-analyses.

4. The “physiological sigh” — double inhalation with long exhalation — has been studied and in one well-designed trial outperformed mindfulness meditation on mood, over one month. (One study, self-reported outcomes. Encouraging, not definitive.)

What is overstated

Heart rate variability as a general “meter” of nervous system state.

It is influenced by: the rate of breathing, its depth, posture, age, medication, time of day, hydration.

Reading a number off a watch and concluding “the state of my nervous system” is oversimplification — and will become a more serious issue in chapter 20.

The conclusion

Slow breathing works. The mechanism is known, mechanical and boring. And exactly for that reason it is reliable.

15. Prediction, delay, synergy

The chapter that was missing — and without it, chapters 1 and 3 remained with a weak explanation.

Chapter 3 said proprioception “calibrates to habit”, by analogy with sensory adaptation to smells and noises.

The analogy is correct but shallow. The real mechanism is more interesting, and documented.

Efference copy: perception of movement is partly prediction

In 1950, independently, Erich von Holst with Horst Mittelstaedt, and Roger Sperry, formulated the same idea:

When the brain sends a motor command, it keeps a copy of it — and uses it to predict what sensory input its own movement will produce.

The term: efference copy (or corollary discharge).

And the proof is everyday:

• Your own hand does not tickle you. The system predicted the stimulus and cancelled it
• The world does not “shake” when you move your eyes — although the image on the retina sweeps violently. But if you gently push your eyelid with your finger, the image moves: same displacement, no command, no cancellation

And so: forward models

Daniel Wolpert formalised mathematically the next step.

Forward model — the brain maintains an internal simulation of the body’s dynamics. Given a command, it predicts the outcome before sensory confirmation arrives.

Inverse model — the reverse: given a desired outcome, which commands produce it.

And the classical finding (Wolpert, Ghahramani & Jordan, Science, 1995):

The brain’s estimate of hand position behaves like a weighted combination of prediction and sensory measurement — with weights proportional to the reliability of each.

When sensation is noisy, the system trusts the prediction. When prediction fails, it turns to sensation.

And now the correct explanation of chapter 1

Alexander did not discover that sensation is “maladapted”.

He discovered, without having the language for it, that:

Perception of what you are doing with your body is not measurement. It is a weighted estimate — and the weight falls on prediction, which is produced by the very broken pattern itself.

The system tells you what it expected to feel. And if the model is built on twenty years of specific use, the prediction is internally consistent and externally wrong.

That is why the mirror worked: it is external measurement that does not pass through the model.

The loop delay

Norbert Wiener, in Cybernetics (1948), argued that organisms and machines solve the same formal control problem. From there the terms:

• Closed loop — measure → compare to goal → correct → repeat
• Open loop — execute a preset command, without checking the result
• Negative feedback — the correction that reduces error

And his critical observation:

Every feedback loop has delay. If the delay is large relative to the system’s dynamics, control becomes unstable or ineffective.

And so the correct explanation of inhibition

Conscious control is a loop with large delay.

For a fast movement — a step, a balance, a reaction — conscious correction always arrives late. And when it arrives late, it does not correct: it oscillates.

The forward model exists precisely to bypass this delay — the simulation runs faster than reality.

Alexander’s inhibition, in these terms, is not “not doing”. It is:

Do not send the command that will activate the old pattern, while there is still time — because once sent, the correction loop is too slow to stop it.

The pause is not moral. It is temporal.

Bernstein: the degrees-of-freedom problem

And now the third piece, from the most important and most under-known person of this chapter.

Nikolai Bernstein (1896–1966), Soviet physiologist, who worked with cinematographic recording of movement in workers, runners and pianists.

The problem: the human body has hundreds of joints and thousands of muscles. The number of possible combinations is astronomical.

If the brain had to compute every parameter separately, no movement would be feasible in real time.

So it does not.

The solution: synergies. Groups of muscles that function as a single functional unit, controlled with one command instead of many. The brain does not command muscles — it commands configurations.

“Repetition without repetition”

His famous phrase, and one of the most important in the whole document.

When you strike a hammer ten times on the same spot, the ten movements are muscularly different each time, although the result is the same.

The system does not store a recipe. It solves the same problem anew each time, with what is available at that moment.

And this explains why skill is flexible, not stereotyped — the expert adapts to each variation without “losing” the skill.

And the hierarchy of levels

Bernstein proposed that movement is organised in levels, from muscular tone to symbolic action.

And the critical thing for learning:

As you learn, control descends levels.

The beginner controls consciously. The expert has handed control to lower, faster levels — and that is why they are both faster and less able to explain what they do to you.

(His work was suppressed in the USSR around 1950 as anti-Pavlovian, and became widely known in the West only after the 1967 translation — after his death. He is, with Gindler and Todd, the third great example of chapter 5: visibility in this field does not follow value.)

And why this changes the reading of Feldenkrais

Feldenkrais was a physicist and engineer, and chapter 8 noted it without exploiting it.

With this chapter’s vocabulary, his principles acquire exact content:

• “Differentiation” is work on synergies: breaking a consolidated functional unit into parts that can be recombined differently
• “Novelty instead of repetition” is, essentially, repetition without repetition: if the system solves the problem anew each time, then variety of conditions is the lesson
• And reducing effort allows reorganisation, because high force locks the synergy in its existing form

That is: Feldenkrais had the right background and did not use it explicitly. And the field that inherited him used it even less.

16. And the opposing school

And here there is real scientific disagreement — not just different emphasis.

The previous chapter described the computational view: internal models, representations, prediction, error signals.

There is a school that holds none of this is needed.

Kelso: spontaneous transition

The HKB model (Haken–Kelso–Bunz, 1985) argues that coordinated movement emerges from the dynamics of coupled oscillators, without a central plan.

And the classical experiment you can do now:

Wave the index fingers of both hands alternately — one up, the other down. And gradually increase the frequency.

At a critical point, the fingers spontaneously transition to synchronised movement — both together.

You do not decide it. And you cannot easily prevent it. The system changes state on its own, like a physical system passing a critical point.

Entrainment

And from there the concept of entrainment: the spontaneous synchronisation of two oscillators through weak coupling.

It does not require computation. It requires only that there is coupling and that you do not prevent it.

And this describes, with unusual accuracy, a host of situations the field names otherwise: two people walking who synchronise; the breathing rhythm that follows music; the group that moves together; and learning to move something that oscillates — where trying to “find the rhythm” is, literally, interference with the coupling.

Gibson: perception is direct

James J. Gibson (1904–1979) and ecological psychology go a step further:

Perception is not mediated by representations. It is direct.

The organism perceives affordances — possibilities for action — directly from the environment.

You do not build a model of a staircase and then calculate whether it can be climbed. You perceive the “climbable” — and specifically in relation to your own body, not in absolute units.

(And there is documentation for this: judgement of whether a step is climbable or an aperture passable scales with the dimensions of the observer’s body, not with centimetres.)

The disagreement, unresolved

The two sides are still debating, and there are hybrid approaches.

I will not adjudicate — for the same reason I did not adjudicate any analogous disagreement in this series.

And why it matters for the field

Because the field borrows vocabulary from both, without knowing it is two.

When a teacher says “let the body find its own solution”, they are speaking dynamical systems. When they say “you are building a new pattern, a new map”, they are speaking internal models.

Both formulations are legitimate. But they are different claims, with different practical consequences — and switching between them within the same sentence is not synthesis; it is vagueness.

17. Motor learning

The chapter that quietly explains most of what the field does — and dissolves its vocabulary.

The stages

Fitts and Posner (1967) described three stages, and they remain useful:

And this is, in other words, Bernstein’s descent of levels (chapter 15).

What the rest of the literature says

1. Repetition entrenches; variety teaches.

Variable practice produces worse immediate performance and better long-term learning and transfer than fixed repetition. One of the most stable findings in the field.

And now we know why: if the system solves the problem anew each time (chapter 15), then learning is not inscription of a recipe but expansion of the solution space.

2. The uncomfortable finding: external focus wins.

Decades of experiments show consistently that when attention turns to the outcome outside the body (“push the floor away”), performance and learning are better than when it turns to the body itself (“squeeze your glute”).

And it is uncomfortable, because an entire field is based on “feel inside your body”.

3. And the phenomenon that explains it: destabilisation by conscious control.

It is documented — and connects with the so-called “choking under pressure”:

The expert who begins to consciously monitor what they are doing, gets worse.

And the mechanism, with chapter 15’s vocabulary, is clear:

Conscious attention to the detail of execution dissolves the synergy — forcing the system to raise control to a later level, with greater loop delay.

And so the reconciliation, now with mechanism

Chapter 15 gives the distinction that was missing:

Internal attention is a tool of reconstruction, not of execution.

• In the cognitive stage, when you are building or rebuilding a pattern, slowly, outside pressure — internal attention is the right tool, because that is exactly when you want the existing synergy to dissolve
• In the autonomous stage, in real time, under pressure — it is sabotage, for exactly the same reason

The same tool, opposite effect, depending on where you are.

And the field almost never makes this distinction — it teaches “feel your body” as a universal instruction, which is wrong in half the cases.

And the finding that dissolves the word “release”

The increase in range of motion after stretching is not primarily due to tissue elongation. It is due to increased stretch tolerance — that is, the nervous system allows more before signalling “stop”.

Nothing lengthened. The threshold moved.

And so

When someone says “it released” — and they truly experience it, because the change is real — what probably happened is one of the following, or a combination:

• The threat estimation changed, and the protective threshold shifted (chapter 12)
• A synergy was reorganised — and a movement that required three muscles is now done by five (chapter 15)
• The forward model was updated — the prediction stopped being wrong

The result is real. The explanation is wrong. And that phrase could be the subtitle of the whole document.