← Did You Know?
🫀Mind8 min read

Polyvagal Theory

It has reshaped how a generation of therapists think about trauma, given millions of people a vocabulary for their own nervous systems, and sold enormous numbers of books. It is also, according to a determined group of physiologists, wrong in most of its specifics.

In plain English

The vagus nerve is the longest of the cranial nerves. It wanders from the brainstem down through the neck and chest into the abdomen, touching the heart, lungs and gut along the way. "Vagus" is Latin for wandering, the same root as vagrant and vague.

It is the main channel of the parasympathetic nervous system, the half of the autonomic system associated with rest and digestion rather than fight and flight.

For most of the twentieth century the standard model was a see-saw. Sympathetic activation on one side, parasympathetic on the other. Threat pushes you one way, safety lets you fall back the other.

In 1994, Stephen Porges proposed that this misses something important. His observation was that the see-saw model cannot easily explain shutdown states: fainting, dissociation, the frozen collapse that people describe after severe trauma. These look parasympathetic in some ways, since heart rate drops, but they are clearly not restful.

His answer was that the vagus is not one system but two, with different evolutionary ages and opposite functions.

The older branch, unmyelinated and originating in a region called the dorsal motor nucleus, produces immobilisation. Think of a reptile going still, or an animal playing dead. Applied to humans under extreme threat, this is collapse, dissociation, shutdown.

The newer branch, myelinated and originating in the nucleus ambiguus, does something quite different. Porges links it to a cluster of functions around social engagement: facial expression, vocal tone, listening, and a calm heart. This is the state where connection is possible.

Stack those with the sympathetic system in the middle and you get a hierarchy of three responses rather than a see-saw of two. Faced with threat, the theory says, you drop down the ladder in evolutionary order: first you try social engagement, then fight or flight, and only when both fail do you shut down.

Porges also coined a useful word for how this switching happens: neuroception. Your nervous system evaluates safety and danger continuously and below conscious awareness. You do not decide to feel unsafe. You notice that you already do.

Five things to file under "wait, what?"

  • The clinical world adopted it far faster than the scientific one. Polyvagal theory is close to standard vocabulary in trauma therapy. It runs through Bessel van der Kolk's The Body Keeps the Score, underpins Deb Dana's widely used clinical framework, and appears constantly in somatic and yoga-informed therapy. Meanwhile it has never achieved consensus acceptance in comparative physiology, and the gap between those two facts has been widening for twenty years.

  • The claim that myelinated vagal fibres are uniquely mammalian is not correct. This matters, because the evolutionary story is load-bearing for the theory. Myelinated vagal fibres controlling the heart have been documented in non-mammalian vertebrates, including some fish and reptiles. Critics argue the neat evolutionary sequence, ancient unmyelinated branch then modern mammalian myelinated branch, does not survive contact with comparative anatomy.

  • The main measurement it relies on is contested. Respiratory sinus arrhythmia, the slight speeding of the heart on inhalation and slowing on exhalation, is used throughout the field as an index of "ventral vagal tone". Paul Grossman and Edwin Taylor have argued at length that RSA is influenced by breathing rate and depth, posture and other factors, and cannot be cleanly read as a measure of a specific vagal branch, let alone of social-engagement capacity.

  • Its most prominent critic thinks the therapy might work anyway. Grossman's position is not that trauma-informed breathing and grounding practices are useless. It is that they do not need polyvagal theory to be true in order to help, and that dressing clinical intuition in unsupported neuroanatomy makes it harder, not easier, to find out what is actually going on. A treatment can work for reasons other than the ones given.

  • "Fight, flight or freeze" got its third term from somewhere. The freeze response entered popular vocabulary largely through this route. Whatever happens to the theory's specifics, it changed the default cultural script about trauma from "why didn't you fight back or run" to "shutdown is a thing nervous systems do", which for a lot of people was the first framing of their own experience that did not sound like an accusation.

The full story

Where it came from

Porges introduced the theory in his 1994 presidential address to the Society for Psychophysiological Research, published the following year. He was working on heart rate variability and had run into a puzzle that became known as the vagal paradox.

In premature infants, vagal activity appeared to be both protective and dangerous depending on the measure used. Bradycardia, a dangerously slow heart rate, seemed vagally mediated. So did respiratory sinus arrhythmia, which was associated with good outcomes. One nerve appeared to be doing two opposite things.

Two branches with different origins and different jobs resolves the paradox neatly. That was the theory's original motivation, and as a piece of problem-solving it is genuinely clever.

The clinical appeal

The theory arrived at a moment when trauma treatment was moving away from purely cognitive approaches and toward the body, and it offered three things that were in short supply.

It gave a physiological account of shutdown, which patients recognised immediately and which previous models handled badly. It reframed symptoms as automatic protective responses rather than failures of will, which is enormously useful in a clinical room. And it pointed at intervention: if the ventral vagal state is associated with breath, voice, facial expression and safe social contact, then slow exhalation, humming, and being with a calm person become plausible routes into regulation.

Clinicians report that this works. That report should be taken seriously, and it is a separate question from whether the mechanism is as described.

The scientific objection

Grossman has published sustained critiques over roughly two decades, most comprehensively in a 2023 paper in Biological Psychology going through the theory's core premises one at a time and arguing that each is either unsupported or contradicted by the comparative literature.

The objections cluster into three areas.

The evolutionary claim, that the myelinated vagal system is a mammalian innovation tied to social behaviour, conflicts with evidence of myelinated cardiac vagal fibres in other vertebrates.

The measurement claim, that RSA indexes a specific ventral vagal circuit, is challenged on the grounds that RSA varies with respiration in ways that have nothing to do with social engagement, and that the inference from a heart-rate pattern to a psychological state is far looser than the field's usage implies.

The functional claim, that the dorsal vagal complex produces the shutdown state in humans, is argued to rest more on analogy with reptilian immobilisation than on direct evidence in humans.

Porges has responded to these criticisms and disputes them. The exchange is ongoing and, at times, quite sharp.

How to hold it

This is a case where the honest answer is genuinely mixed, and flattening it in either direction loses the interesting part.

Treating polyvagal theory as established neuroscience is not supportable. It is presented that way constantly, in training courses, popular books and social media, usually with far more confidence than the evidence carries.

Dismissing it entirely also misses something. It identified a real gap in the two-branch model, it named states that patients recognise, and it generated practices that many people find useful. Those are real contributions even if the anatomy turns out to be wrong.

The most defensible position is probably this: the descriptive vocabulary is useful, the clinical practices may well help, and the mechanistic story should be held loosely and stated as a hypothesis rather than a finding. That is less satisfying than a clean verdict. It is also where the evidence currently sits.

Go deeper

For the curious:

On YouTube:

Keep going

More things worth knowing

All topics →