Part 4: There Are Sensors on the Inside — The Vagus Nerve and Interoception

Breathing, Revisited — Part 4 of 4

This is the fourth and last of four pieces in the series Breathing, Revisited. The series returns to breathing from the side of the research, eight years into offering Rolfing sessions. It stands in a different place from the four-part breathing series written back in 2015, during basic training. This final piece takes up the sensing of the body’s interior itself, breathing included.

What bodily awareness is

Since June 2015, I have offered Rolfing sessions in the Shibuya, Ebisu, and Daikanyama area of Tokyo.

Why hands-on Rolfing produces effects that last, unlike massage or seitai, I described in “<a href=”https://rolfing-zero.hidefumiotsuka.com/en/2023/05/09/87183/”>How Rolfing Differs from Seitai and Massage — “Symptom” or “Structure”?</a>” Here I want to take that further: what is the good of cultivating bodily awareness — interoception — through Rolfing, and how does it relate to intuition and to the vagus nerve?

By cultivating bodily awareness, Rolfing improves posture, and the effect lasts. “Bodily awareness” means noticing the state of the body’s interior.

Alongside the five senses — sight, hearing, taste, smell, touch — which take in information from the outside world, the body also has sensors that take in information from within and send it to the brain. These sensors are what we can call bodily awareness, or interoception.

The signals arise throughout the body: in the organs, within muscle, within bone. Strung together, they build up a single picture — the current state of oneself, how one feels in this very moment, and what to do in order to keep the body’s inner sensations in balance.

Much of what follows draws on an episode in which Andrew Huberman of Stanford University discusses the connection between brain and body: how the organs shape the workings of the brain, and how the brain governs the organs. Conscious awareness of that dialogue is what he calls interoception.

The vagus nerve and the organs — communication, and the making of emotion

One structure plays a key role in this sensing: the vagus nerve. The vagus is the one cranial nerve that reaches all the way from the brain to the abdomen, connecting the brainstem to the gastrointestinal tract, the bronchi, and the heart, and taking part in the work of the internal organs.

What is striking is how much it reports. Information from the lungs (breathing), the heart (heart rate, blood pressure), the stomach and intestines (hungry or full; which nutrients have arrived), the stomach (whether acid is being secreted), the gut (the balance of intestinal bacteria), the spleen (immune activity) — all of it travels continuously from the vagus to the brain. The vagus’s signals become bodily awareness as they reach the brain.

What travels divides roughly into two kinds: mechanical forces — pressure, pain, volume — and chemical factors, such as the acidity of the gut and the diversity of the microbiome. The two combine to shape mood, inflammation, immune activity, and even recovery from injury.

After a meal, for instance, how much protein, fat, and carbohydrate has entered the gut is relayed to the brain through the vagus within thirty minutes — well before the small intestine finishes absorbing the nutrients, which takes some four to six hours. On that basis, the body comes to feel sated, or to seek more.

The vagus is usually noticed for its parasympathetic role, calming the body. But it also works in the other direction. After sugar is eaten, signals travel from the vagus to the brain that can prompt the body to want still more of it — so the vagus can drive activation as well.

This is a point Huberman himself returns to. The idea that activating the vagus always leads to relaxation is not accurate. Electrical stimulation of the vagus can, by way of the nucleus of the solitary tract, raise alertness instead. It is not a nerve that only calms.

And when the body is under stress, vagal activity drops, and the communication between brain and organs falls quiet. It is less that stress directly worsens the gut than that, with signals from the gut no longer getting through, problems like indigestion set in.

Research on interoception supports this from its own side. Interoceptive accuracy correlates positively with emotion regulation and stress resilience, and reduced accuracy has been observed following acute stress. Under stress, the interior becomes harder to read.

Finally, the vagus also integrates. From the heart (heart rate, blood pressure), the lungs (breathing), the gut (digestion, absorption), hormones in the blood, and more, it helps shape which emotion or mood the body expresses, and relays that to the brain. In the end, emotion shows on the face.

The vagus does not only descend

Let me pause here to sort something out.

The vagus runs in two directions: down from the brain to the organs, and up from the organs to the brain. The first is efferent, the second afferent.

Everything this piece has covered so far belongs to the upward side. Information rises continuously from the organs, and that becomes bodily awareness.

The downward side I wrote about eleven years ago, in a separate piece. It was February 2015, during the Phase III training, and it set out the polyvagal theory I had learned in Phase II (see “The Face Makes Safety — Polyvagal Theory and the Two Vagus Nerves“).

What that piece addressed was Stephen Porges’s theory. The vagus has myelinated fibres (ventral) and unmyelinated ones (dorsal); the former is something mammals developed anew, connecting to the cranial nerves that govern facial expression and vocalisation. The new vagus, the sympathetic system, and the old vagus form a hierarchy: feel safe and the topmost works; let a crisis arrive and the system descends.

The same nerve, seen from opposite faces.

On the theory

That theory, however, is now the subject of considerable dispute.

Andrew Huberman notes that the word “polyvagal” reflects the multiple branches of the vagus, and grants the complexity that implies — but he points out that Porges’s model does not align entirely with modern anatomical understanding. He also objects to the way it is sometimes used to make diagnostic claims about psychological and physical conditions without substantial physiological evidence.

In 2026 came a larger criticism. Paul Grossman, formerly research director of psychosomatic medicine at the University Hospital Basel, published a paper with thirty-eight other researchers of the autonomic nervous system arguing that polyvagal theory is untenable. Thirty-nine experts were invited to evaluate it; the work is described as two decades in the making.

Three points carry the argument.

First, that using RSA — respiratory sinus arrhythmia — as a direct index of vagal tone is not supported. RSA reflects multiple influences and cannot uniquely mark ventral vagal activity. It is shaped by the rate and depth of breathing, by baroreflex function, by carbon dioxide levels, by sympathetic activity, by ageing, and by the intrinsic mechanisms of the heart.

Second, the anatomy: the functional distinction drawn between the nucleus ambiguus and the dorsal motor nucleus, and the emphasis on myelinated versus unmyelinated fibres, are said to be oversimplified and unsupported by current evidence.

Third, evolution: the claim that mammalian vagal pathways are unique is said to be overstated, since similar structures appear in reptiles and fish.

Porges has responded. The critique, he argues, conflates neuroanatomy with neurophysiology, reducing a systems-level theory to a dispute about measurement. And Porges himself grants that HRV is not a direct readout of vagal state but a proxy influenced by multiple interacting variables.

Nothing here is settled. Criticisms of this kind have appeared in the peer-reviewed literature since around 2007, with Porges responding to each; third-party assessments divide between those who find the critiques legitimate but see no theory in collapse, and those who find the theory no longer defensible.

One thing matters for this piece.

Every one of those criticisms points at the downward side: the ventral–dorsal distinction, the evolutionary sequence, and the reading of RSA as an index of vagal tone. That information rises from the organs to the brain is not itself in dispute. Research on interoception has accumulated along its own line.

What this piece addresses is that upward side.

That said, the second part of this series took up HRV (see “Part 2: The Diaphragm Reaches the Heart — Breath, Heart Rate, and Two Nuclei in the Brainstem“). HRV is RSA, and RSA is the first of the points above. High means parasympathetic predominance, low means sympathetic — it is not that simple.

Why interoception matters — the unconscious rhythm of heartbeat and breath

One remarkable thing about the human body is that, in the presence of someone we are close to, the rhythms of heartbeat and breath fall into sync without our noticing.

Why should that synchrony matter? People with higher interoceptive accuracy have been found to experience more intense emotional reactions and heightened sensitivity to the emotions of others. Reading one’s own interior and reading another person appear to come from the same place.

Turn that around: when the interior goes unread, the synchrony happens anyway, and one is carried along by the other person’s state without noticing. That is why, in adversity, on meeting someone one clashes with, or in the middle of a challenge, taking a little time to settle those signals makes it more likely one can get past the moment.

The relation between empathy and interoception I took up later, at length, in a piece on the neural circuits of empathy and self-sensing. Along the circuit that runs from the mirror neurons through the insula to the medial prefrontal cortex, it is the insula that takes another person’s movement or expression and registers it as a sensation on one’s own inside.

Interoception can be cultivated

Interoception, one might say, is an intuition that can be cultivated within oneself.

The word intuition carries a suggestion of guesswork with nothing behind it. The somatic marker hypothesis, which addresses the role of emotion in decision-making, takes a different view: emotional processes are integral to deciding, and they rest on bodily sensations bound up with emotion. A gut feeling, on this account, is the reading of a signal from the body.

And the encouraging part is exactly that it can be cultivated.

This has been tested. In groups that practised a body scan over eight weeks, interoceptive accuracy — measured by a heartbeat perception task — improved significantly. Interoception can be improved by long-term interventions that direct attention to bodily signals. The perception of the heartbeat itself is likewise known to be trainable.

Rolfing, which settles posture by working with the fascia, is one way to cultivate it.

Conclusion

In this piece I have looked at the relation between bodily awareness — interoception — and Rolfing, through the role of the vagus nerve.

Four parts of following the breath, and where the last of them arrives is this: breathing is one part of the sensing of the body’s interior.

Among the signals rising continuously from the organs, there is breath. There is heartbeat. Whether one can register them changes both one’s own state and one’s distance from another person.

Breathing is one of the few interoceptive signals that can be reached consciously. Heart rate and digestion cannot be moved by will; breath can be felt the moment attention turns to it, and can be altered. That is why the ten-session series begins with breathing.

Breathing, Revisited (Four Parts)

The breathing series written in 2015, during basic training, is here.

Breathing Series (Four Parts)

Note, September 2026

I wrote this piece in June 2023, addressing what the vagus nerve carries from the organs to the brain. The lungs, the heart, the gut, stomach acid, intestinal bacteria, the spleen. What rises from there becomes bodily awareness.

I had written about the same nerve eleven years earlier.

It was February 2015, during the Phase III training, and the piece set out the polyvagal theory I had learned in Phase II. What it addressed was the two vagus nerves, myelinated and unmyelinated; facial expression and vocalisation; and the hierarchy of new vagus, sympathetic system, and old vagus.

The two pieces look at opposite faces of the same nerve.

What 2015 was looking at was the side that descends from the brain to the organs. A facial expression becomes a signal of safety, and that signal reaches the other person’s nervous system. Work directed outward.

What 2023 was looking at was the side that rises from the organs to the brain. The state of the organs is reported continuously, and comes into focus as one’s own felt sense. Work directed inward.

While writing them, I thought only that I was writing about the vagus nerve. That the directions were opposite had not occurred to me.

And over those eleven years, the two faces came apart.

The descending side is now in considerable dispute. As the body of this piece describes, in 2026 Paul Grossman and thirty-eight researchers published a paper arguing that polyvagal theory is untenable. The ventral–dorsal distinction, the evolutionary sequence, the reading of RSA as an index of vagal tone. Porges has responded, and nothing is settled.

The ascending side — that information rises from the organs to the brain and becomes bodily awareness — is not in dispute. Research on interoception has accumulated along its own line.

The 2015 piece set out Porges’s account as he gave it. At the time, that was the current knowledge. Eleven years on, the foundation of that framework is what is being questioned.

There is something to hold onto here.

Seeing how this went, people sometimes take it to mean the theory was simply wrong. The case is the reverse.

Polyvagal theory made specific claims. The ventral vagus is myelinated; it is unique to mammals; RSA reflects its activity. Because it said so specifically, it became clear which parts could be checked and how. That is why thirty-nine researchers could spend twenty years examining it, why Porges could respond, and why the argument continues.

Being refutable is not a weakness of a science but its condition. An account that fits whatever happens cannot be checked. What cannot be checked will never be overturned — and will never move forward either.

In the world of bodywork, this theory has been widely taken up. Safety, danger, shutdown: three states that give names to what happens in a session. Being easy to grasp is itself a kind of power. But ease of grasp is a separate thing from being right, and what is in question now is the foundation underneath.

That it is being examined at all is, in itself, a mark that the theory stands on the side of science.

In the body of this piece I wrote that what it addresses lies outside the dispute. That was not because the piece was careful. It happened to be written about that side.

What looks well-founded now may not look so eleven years from now. There is no writing except without knowing that. But what one writes can be left in a form that allows it to be checked afterward. That is what this note is for.

Bio

Hidefumi Otsuka

Archives