The Vagus Nerve: Anatomy, Vagal Tone, and What Stimulation Actually Shows

The vagus is the cranial nerve that leaves the head

Eleven of the twelve cranial nerve pairs exit through the base of the skull and stay in the head and neck. The tenth pair travels on, down the neck, into the chest, and through the diaphragm into the belly. Anatomists named it the vagus, from the Latin for wandering, the same root that gives us vagrant and vagabond.

Start with one plain definition. A nerve is a living wire: a bundle of long thin fibers, each one a projection of a single cell, carrying messages as tiny electrical pulses.

Most of the body's wiring leaves the brain, runs down inside the bony spine, and exits between the vertebrae to reach the arms, legs and trunk. The cranial nerves take the direct route out of the skull instead. They serve the eyes, the face, the tongue, the ears, and the muscles that move the head.

The vagus is the exception, and that single anatomical fact is the reason it became a household word. No other nerve touches so many organs at once. When people say the vagus connects the brain and the body, they are pointing at something true. What they usually get wrong is the direction of the traffic, and the picture of what the nerve is doing while it travels.

The vagus reaches the throat, lungs, heart and most of the gut

The vagus has the widest reach of any nerve in the body. It begins as a set of cell clusters in the brainstem, the stalk where the brain meets the spinal cord. This is the oldest and most automatic part of the nervous system, the part that runs breathing, heart rate, swallowing and blood pressure without asking permission.

Fibers gather and leave the skull through an opening called the jugular foramen. On each side of the neck the nerve runs in a sheath alongside the carotid artery and the jugular vein, which is why a surgeon working on the neck must know exactly where it is.

It sends branches to the throat and the voice box, so speech and swallowing depend on it. It sends branches to the windpipe and the lungs. It reaches the heart. It passes through the diaphragm and spreads across the stomach, the liver, the pancreas and the biliary system, and it continues along the intestine.

The neuroanatomists Hans-Rudolf Berthoud and Winfried Neuhuber, who study the wiring between the gut and the brain, reviewed the tracing studies that follow individual fibers to their endpoints. They wanted to know which tissues vagal sensory fibers actually reach and what kinds of ending they form there. Their 2000 survey of the afferent vagal system put the endings in the esophagus, the lower airways, the heart and the aorta.

It traced more endings through the entire gastrointestinal tract, the liver, the portal vein and the pancreas. The spleen was not on the list. They also described distinct specialized terminals shaped for different jobs, and recordings showing sensors for mechanical stretch, for chemistry, for temperature and for the concentration of dissolved particles.

Hold on to that last detail. The vagus is a cable of many separate lines, each tuned to a different internal event, and never a single wire carrying one kind of feeling.

One vagal cable carries traffic in both directions

Two words organize everything about the vagus. A fiber carrying a signal outward from the brain to an organ is called efferent, and it delivers instructions. A fiber carrying a signal inward from an organ to the brain is called afferent, and it delivers reports.

The vagus contains both, bundled together in the same sheath. Large nerves normally carry both directions. The vagus breaks the norm in its ratio, which the next section counts.

Inside the brainstem the two directions have separate addresses. Incoming reports arrive at a long thin column of cells called the nucleus of the solitary tract, where news from the lungs, the heart, the blood vessels and the gut lands first. Outgoing slowing signals leave from two other clusters, the dorsal motor nucleus and the nucleus ambiguus. One nerve, two separate addresses, running in opposite directions at the same time.

The neurologists Hong Yuan and Stephen Silberstein, who study headache and neuromodulation, wrote a 2016 review of the vagus and its clinical stimulation. They describe four vagal nuclei governing the cardiovascular, respiratory and digestive systems. They note that most of the fibers coming from the internal organs are thin and unmyelinated.

And they characterize the nerve as an integrative control center that takes in the body's internal state and answers with an adjustment. That word, integrative, is the important one. The vagus is a conversation, not a broadcast.

The vagus is mostly a listening nerve

Roughly 80 percent of vagal fibers are afferent. Four fifths of the traffic runs up to the brain. The figure is repeated everywhere, usually without a source, and it is worth knowing where it comes from and how far to trust it.

It comes from 1957. The respiratory physiologist Emilio Agostoni, working with J. E. Chinnock, M. de Burgh Daly and J. G. Murray, set out to answer a straightforward anatomical question. What is actually inside the vagus and its branches to the heart, lungs and abdominal organs? They combined function testing with counting fibers under a microscope in the cat. Their fiber counts and functional studies are the primary source that the widely quoted four fifths traces back to.

A figure repeated a million times is still only as good as its measurement, and this one has a known spread. The classic count was made in cats, and the proportion differs between branches and between species.

The neuroscientists James Prechtl and Terry Powley checked the accepted ratios by building a complete electron-microscope inventory of every axon in the rat's abdominal vagus and its five main branches. Their count of nearly 27,000 fibers led them to conclude that efferent fibers may represent over a quarter of the total, higher than all the widely accepted estimates in the literature.

So state it precisely. The vagus is predominantly a sensory nerve, by a large margin, in every species examined. The exact fraction depends on where you cut and what you are counting. Even at the most conservative count, most of the traffic is the body reporting to the brain, and the popular picture of a command cable is upside down.

The vagus is the body's reporting line first and its command line second. Read that way, it stops being a switch and starts being a sense organ.

The vagus carries many distinct internal senses

What the vagus reports is a catalog of separate channels rather than one feeling. The molecular neurobiologists Sara Prescott and Stephen Liberles asked how many genetically distinct kinds of internal sensor the vagus actually carries, using modern labeling techniques that tag one cell type at a time.

Their 2022 review of the internal senses of the vagus describes a remarkable diversity of sensory neuron types, each innervating particular organs, many still poorly understood. Their point is that interoception, the sense of the body's own interior, is not one sense. It is a whole sensory system with as many channels as vision has colors.

An earlier study from the same laboratory makes it concrete. Erika Williams and colleagues wanted to know whether the vagal neurons that sense stretch in the gut are the same neurons that sense food arriving. By labeling and activating each population separately in mice, they showed the two are different lines entirely, one tuned to mechanical fullness and one tuned to nutrients. The stomach tells the brain that it is stretched and that it has been fed as two separate messages.

How the vagus slows the heart, and what vagal tone means

The body steers its organs with two opposing wiring systems. The sympathetic system is the accelerator: it speeds the heart, tightens blood vessels, opens the airways and readies the body for effort. The parasympathetic system is the brake, and in the chest and abdomen the vagus is its main cable. It slows the heart, quiets the airways, and turns on digestion.

Neither is good and neither is bad. A car with only a brake is as useless as a car with only an accelerator. The pair is a steering system, and health lives in how well the two are matched to the moment.

Loewi's two frog hearts find the transmitter

Nerves speak to their organs with a released chemical, and the proof came from one of the most elegant experiments in the history of medicine. In 1921 the pharmacologist Otto Loewi wanted to settle an argument: did nerves speak to organs purely by electricity, or did they release a chemical? He took two frog hearts, kept them beating in fluid, and stimulated the vagus of the first until it slowed.

Then he took the fluid bathing that first heart and applied it to the second heart, which had not been touched. The second heart slowed too. Something released into the fluid had carried the message. Loewi called it Vagusstoff, meaning vagus substance, and it was later identified as acetylcholine. A historical account of the discovery is why we can now say exactly how the brake works. Vagal endings release acetylcholine onto the heart's natural pacemaker, and the pacemaker fires more slowly.

The physiologist Matthew Levy then asked what happens when both systems act at once, since in a living body they always do. His work on sympathetic and parasympathetic interaction in the heart showed that the two inputs do not simply add up. Vagal slowing becomes more powerful when sympathetic drive is high. The brake is not a fixed force subtracted from a fixed engine. Its effect depends on the state of the system it acts on, and that dependence returns in every stimulation trial below.

The intrinsic heart rate reveals the standing brake

Vagal tone sounds mystical and is perfectly concrete. It means how much brake is standing on, right now. Not a mood, not a personality trait, not a virtue. A physiological quantity.

Here is the cleanest demonstration that it exists. The heart has its own built-in pacemaker and will beat on its own with no nerve supply at all. So what rate does it choose when left alone? The cardiologists A. D. Jose and D.

Collison answered this in people by giving drugs that block both the accelerator and the brake at the same time, then measuring what the heart did unattended. Their measurement of the intrinsic heart rate in man established two things that still stand. The resting heart rate is lower than the intrinsic rate, and the intrinsic rate falls with age.

Read that again, because it is the whole concept in one line. A calm resting pulse in the sixties or seventies is not the heart running free. It is the heart running fast and being held back. Something is standing on the brake continuously, all day, without your involvement. That standing pressure is vagal tone.

Which means the interesting question is never simply whether the brake works. It is how well the braking is matched to what the moment requires, and how quickly it can come off and go back on.

Respiratory sinus arrhythmia couples the heartbeat to the breath

Put two fingers on your wrist and breathe slowly. If you are quiet enough, you will feel your pulse quicken slightly as you breathe in and settle as you breathe out. That rhythm is respiratory sinus arrhythmia, and it is the vagus made audible at the skin.

The mechanism is timing. Vagal traffic to the heart is not a steady stream. It is gated on and off within every breath. During inhalation the brake lifts and the heart speeds. During exhalation the brake reapplies and the heart slows. The cardiovascular physiologist Dwain Eckberg studied this directly by measuring the intervals between heartbeats in healthy volunteers while controlling how fast and how deeply they breathed.

He wanted to know whether the beat-to-beat pattern could be used to read vagal traffic itself. His measurements of human sinus arrhythmia showed a close relationship between the depth and interval of breathing and the quantity, periodicity and timing of vagal outflow to the heart. Slower breathing produced larger swings. Deeper breathing produced larger swings, though less than you might expect.

Does the rhythm do anything useful, or is it just a side effect of the wiring? The physicians Fumihiko Yasuma and Junichiro Hayano asked exactly that. Their review of why the heartbeat synchronizes with the breath gathered evidence that the rhythm improves the efficiency of gas exchange in the lung.

Matching heartbeats to the breath aligns blood flow with the moments when air is actually available. The same review notes that the rhythm and general vagal control of the heart can come apart, which matters for everything that follows.

Heart rate variability indexes the vagal brake

Your heart does not beat like a metronome. The gap between one beat and the next is always changing, by a few thousandths of a second at a time. Heart rate variability, usually shortened to HRV, measures those tiny differences, and its fast component is the standard index of vagal traffic to the heart.

Two of its numbers are worth understanding, because your device is probably showing you one of them. The first is RMSSD, which sounds technical and simply means the average size of the difference between one beat gap and the next, calculated in a way that keeps large differences from canceling small ones.

It captures the fast, beat-to-beat wobble. The second is high-frequency power, which is the part of the variability that cycles at the speed of breathing. Both are ways of asking the same question. How much is the beat interval moving at the pace the vagus works at?

Blockade experiments earned the vagal label

The reason those particular numbers are read as vagal is not tradition. It is a blockade experiment. Give someone a drug that blocks the vagus and watch what disappears. Solange Akselrod and colleagues, working in cardiovascular signal analysis, built the spectral method for this. They tested it with autonomic blocking drugs, asking which nervous inputs produced which frequencies in the heart rate signal.

Their power spectrum analysis of heart rate fluctuation showed that the sympathetic and parasympathetic systems make frequency-specific contributions to the variability. Bruce Pomeranz and colleagues, working in human autonomic physiology, then repeated the logic in people, using blocking agents and changes in posture. Their assessment of autonomic function by spectral analysis found that the high-frequency fluctuations at the breathing rate are mediated solely by the parasympathetic system.

That is the evidence. The fast, breathing-paced component of heart rate variability vanishes when the vagus is blocked, so it can be used to index vagal traffic to the heart. A task force of two cardiology societies later fixed the standards for measuring and interpreting it in 1996, which is why the same metrics appear in research everywhere.

The psychophysiologists Fred Shaffer and Jay Ginsberg wrote a plain overview of the metrics and their normal ranges for clinicians entering the field. It remains the clearest place to see what each number represents.

One line stays sharp throughout. That HRV is a validated index of autonomic state is established science. Reading HRV as a window onto tone, in the sense the Unified Model of Tone uses the word, is the model's interpretation. The measurement is medicine's. The reading is ours.

What a high or low HRV number means, and the second window

People arrive at heart rate variability wanting a score. Higher is better, lower is worse, and mine is 42. The reality is more interesting and much more personal, and it starts with age.

Age moves the numbers more than almost anything else

The cardiologist Ken Umetani and colleagues set out to define the normal range across the whole human lifespan, measuring 24-hour recordings in 260 healthy people from the second decade of life to the tenth.

Their survey of variability across nine decades found a steady decline with age, and the fast vagal measures fell fastest, reaching under half of their young-adult values by the sixth decade before leveling off. They also reported something that should make anyone cautious with cutoffs. In healthy people over 65, a meaningful fraction fell below the published thresholds that are supposed to mark increased risk.

Low variability predicts risk at population scale

Now the risk side, which is real. Hisako Tsuji and colleagues in the Framingham Heart Study wanted to know whether low variability predicted trouble in ordinary people rather than in cardiac patients. They reprocessed ambulatory recordings from 2,501 participants with no clinically apparent heart disease and followed them.

Their analysis of reduced variability and cardiac risk found that lower HRV predicted new cardiac events after adjusting for the usual risk factors. Each standard deviation of decline in one common measure carried about a 47 percent increase in hazard. Read that with the size in mind. The follow-up averaged three and a half years and produced 58 events in all.

Both facts are true at once, and holding them together is the whole skill. At the population level, lower variability tracks worse outcomes. At the individual level, a single number tells you very little, because the spread between healthy people of the same age is enormous. HRV is interpretable against your own baseline over weeks, not against your friend's ring.

The variability is a brain measurement too

The psychophysiologist Julian Thayer and colleagues pooled the neuroimaging studies that had measured brain activity and heart rate variability at the same time, asking which brain regions track the vagal signal. Their 2012 meta-analysis of variability and neuroimaging found consistent associations with the amygdala and the ventromedial prefrontal cortex, regions involved in threat appraisal and its regulation. The brake at your heart is connected to the part of you that decides whether the world is safe.

The baroreflex is the second window on the same brake

In the walls of the large arteries in your neck sit stretch sensors that feel how hard the blood is pressing with each beat. They report to the brainstem, and the vagus answers within a single heartbeat by slowing the heart. The physiologists Erica Wehrwein and Michael Joyner described how this arterial baroreflex regulates blood pressure as the fastest guardian of the circulation. How much the heart period changes for a given change in pressure is called baroreflex sensitivity, and it can be measured.

Maria Teresa La Rovere and colleagues, who developed much of the method, set out the ways it is measured and what it means clinically. In the ATRAMI study the same group followed 1,284 patients after a heart attack to test whether autonomic measures predicted death. Their finding that low baroreflex sensitivity and low variability both predicted cardiac mortality is one of the strongest pieces of evidence that the regulation itself carries the prognosis, not only the pump.

What a ring or watch can and cannot see

Millions of people now wake up to a vagal number computed while they slept. What produced it matters, because the instrument sets the limits of the reading.

A medical HRV measurement uses electrodes on the skin to catch the heart's electrical spike directly, and the timing is accurate to a few thousandths of a second. A ring or a watch does something different.

It shines light into the skin and watches the reflected signal change as blood pulses through, then estimates where the beat was. The estimate is good under ideal conditions, which is exactly why most consumer devices compute your headline number during sleep, when you are still and warm and the signal is clean.

How well do these devices see sleep itself? The sleep scientist Evan Chinoy and colleagues tested seven consumer sleep trackers against polysomnography, the laboratory standard that records brain waves, eye movement and muscle activity, over three nights including a deliberately disrupted one. Their head-to-head comparison with polysomnography found that the devices were very good at knowing you were asleep, with sensitivity above 0.93, and much weaker at knowing you were awake, with specificity between 0.18 and 0.54.

Sleep stage assessments were inconsistent, and performance got worse on the nights when sleep was poor. Simply put, these devices know sleep from wake reasonably well and know sleep stages badly, and they struggle most on exactly the bad nights you most want explained.

Comparing your HRV number to another person's is close to meaningless, because the between-person spread swamps almost any effect you are looking for. A single night means little, since a late meal, a drink or a hard workout can move it. What is worth reading is your own trend over weeks, and what it does when something in your life changes.

An index is not the thing it indexes

One more correction comes from inside the field. The psychophysiologist Paul Grossman has spent years arguing a careful point about respiratory sinus arrhythmia. That breathing-linked wobble, which most vagal indices depend on, is an approximate index of a vagal process rather than a direct readout of vagal tone.

His 2023 challenge to the premises of polyvagal theory calls the equation of that rhythm with general vagal tone a category mistake, confusing an approximate index with the thing itself. He is right. An index is enormously useful. It is still an index.

The vagus reports and restrains inflammation

The inflammatory reflex is the discovery that made the vagus famous outside cardiology, and it is the clearest proof that the nerve is a two-way instrument. A sensory limb detects inflammation, and a motor limb damps it.

Start with inflammation itself. When the body is injured or infected, immune cells release small signaling proteins called cytokines. They are chemical alarm bells. One of the most important is tumor necrosis factor, usually shortened to TNF. Cytokines are essential in the right amount and dangerous in excess, and runaway release is what turns an infection into septic shock.

Tracey's endotoxin experiments find the circuit

In the late 1990s the neurosurgeon Kevin Tracey was studying a drug intended to block inflammation in the brain. He injected it into the brains of rats and found, unexpectedly, that inflammation was suppressed in the body as well, far away from the injection. That is not how a drug in the brain should behave. Something was carrying the message out, and the obvious candidate was a nerve.

His group tested it directly. Lyudmila Borovikova and colleagues asked whether stimulating the vagus could change the immune response to endotoxin, the bacterial fragment that triggers septic shock. Their demonstration that vagal stimulation blunts the response to endotoxin showed two things. Acetylcholine, the vagus transmitter, suppressed the release of TNF and other alarm cytokines from human immune cells in a dish.

Electrically stimulating the vagus in living rats during lethal endotoxemia lowered TNF in the liver and blood and prevented shock. Tracey named the circuit the inflammatory reflex, and the name is exact. Like any reflex it has a sensory limb that detects and a motor limb that responds.

The nerve tells one cytokine from another

The sensory limb is the part that returns us to the 80 percent. Theodoros Zanos and colleagues asked whether the vagus can tell one cytokine from another, which would mean the nerve reports what is wrong and not merely that something is. They recorded raw electrical traffic from the vagus in mice exposed to TNF or to interleukin-1 beta, sorted the impulses by shape, and fed the firing patterns to a decoder.

Their decoding of cytokine-specific neural signals identified which cytokine the animal had received about 83 percent of the time, against a chance rate of 33 percent. Mice lacking the receptor for one cytokine no longer produced its signal, which is exactly the control the claim needs.

The human step in rheumatoid arthritis

Then came the human test. The rheumatologist Frieda Koopman and colleagues, working with Tracey's group, asked whether stimulating the vagus in people would inhibit TNF and change disease. They studied patients with implanted vagus stimulators, first in epilepsy patients and then in patients with rheumatoid arthritis, a disease driven by TNF.

Their finding that stimulation inhibited cytokine production and reduced arthritis severity established that the reflex exists in humans and can be engaged deliberately. Read it with its limits attached. It was a small open-label study with no sham arm, a real weakness in a disease where expectation influences reported symptoms. It is a proof of principle, and a striking one.

What is true about the gut and the brain, and what is oversold

The vagus is the physical basis of the phrase gut feeling, and this is where popular writing goes furthest past the evidence. So take the strongest real findings first.

For most of the twentieth century the gut was thought to speak to the brain only slowly, by releasing hormones into the blood. The neuroscientist Diego Bohorquez and his colleague Melanie Kaelberer asked whether there was a faster route. They looked at enteroendocrine cells, the hormone-releasing cells sitting in the gut lining, and noticed they had features of nerve cells.

Their discovery of a gut-brain neural circuit for nutrient sensing showed that these cells, which they renamed neuropod cells, form true synapses onto vagal neurons and signal with glutamate in milliseconds. The gut is wired to the brainstem, one synapse away, with the speed and precision of a nerve.

Next, the bacteria. Javier Bravo and colleagues in Cork fed mice a specific Lactobacillus strain and measured both behavior and brain receptor expression, asking whether a gut bacterium could change the brain of a healthy animal. Their finding that the effects vanished in vagotomized mice is the important part. The bacteria altered stress hormone responses, anxiety-related behavior and GABA receptor expression in the brain, and cutting the vagus abolished those effects. The vagus was the road the message traveled.

Mouse work is not human proof

Now the correction. The microbiologist Jens Walter and colleagues systematically reviewed studies that transplanted human gut microbes into rodents and then reported that the animals had acquired the human donor's disease phenotype. Their review found that 95 percent of such studies, 36 of 38, reported a successful transfer of pathology.

They argue that a success rate that high across species is implausible on its face, and that it overstates the role of the gut microbiome in human disease. When a whole literature reports that everything works, something in the reporting is wrong.

The summary of the gut line reads like this. The wiring is real and fast. The vagus genuinely carries information from the gut, including information about its bacterial residents, and that traffic can change behavior in animals. The leap from that to a supplement that will fix your mood through your vagus is marketing, not science.

What vagus nerve stimulation is, and what the trials show

When a neurologist says vagus nerve stimulation, they mean surgery. A thin electrode is coiled around the left vagus in the neck and tunneled under the skin to a generator implanted below the collarbone, much like a pacemaker. It fires in cycles, on for a few seconds and off for a few minutes, around the clock.

Epilepsy delivered the first strong evidence

The neurologist Adrian Handforth and colleagues ran a multicenter trial in patients whose seizures had not responded to drugs, comparing presumed therapeutic high stimulation to low stimulation as a control. Their randomized trial in partial-onset seizures found a 28 percent average reduction in seizure frequency with high stimulation against 15 percent with low, a real but moderate difference. Note the control they had to use.

They could not use a completely inactive device, because stimulation of the vagus in the neck changes the voice, and patients feel it. High stimulation caused more voice alteration and shortness of breath than low. This is the central methodological problem of the entire field. It is genuinely hard to blind a treatment that the patient can hear in their own speech.

Depression is a null at ten weeks and a signal at five years

Depression is the second approved indication and a more complicated story. The psychiatrist John Rush and colleagues ran the ten-week randomized sham-controlled trial in 235 patients with treatment-resistant depression. Their acute-phase trial found response rates of 15.2 percent with active stimulation against 10.0 percent with sham. The difference did not reach statistical significance, and the authors concluded that the study did not yield definitive evidence of short-term efficacy. That is a null result and it should be stated as one.

Longer observation looks different. The psychiatrist Scott Aaronson and colleagues followed 795 patients with treatment-resistant depression for five years in a registry comparing stimulation plus usual treatment to usual treatment alone. Their five-year comparison reported a cumulative response rate of 67.6 percent with stimulation against 40.9 percent without. The design limit is structural and large. The study was open-label and not randomized, so patients and doctors knew who had a device, and the groups were not assembled by chance. It is suggestive evidence, not proof.

The largest randomized test to date, the RECOVER trial, enrolled 493 patients with markedly treatment-resistant depression. A report from that trial states plainly that the primary outcome showed no statistically significant difference between the treatment conditions. The report then examines whether different outcome definitions and longer observation windows detect change better in this severely ill population.

Hold both halves. A large randomized trial missed its primary endpoint, and the effects that appear over a year in this group are small and slow enough that how you measure them decides whether you see them.

Heart failure produced the sharpest nulls

The heart failure trials are important precisely because the rationale was so good. Patients with failing hearts have too much accelerator and too little brake, so adding vagal stimulation should help. In NECTAR-HF, the cardiologist Faiez Zannad and colleagues randomized implanted patients to stimulation on or off for six months.

Their sham-controlled trial found no benefit on the primary measure of heart remodeling or on the other objective measures, while quality of life and symptom class did improve. In INOVATE-HF, the cardiologist Michael Gold and colleagues randomized 707 patients and followed them for a mean of sixteen months. Their trial found no reduction in death or heart failure events, while quality of life, symptom class and walking distance again improved.

The non-surgical cousins split their results

Then there are the non-surgical devices, which are what most consumers actually encounter. One is a clip or electrode worn in the outer ear, targeting a small vagal branch that supplies part of the ear canal and the concha.

The neurogastroenterology researchers Mohammed Butt and colleagues mapped the anatomical basis for stimulating the vagus through the ear. They found no consensus in the literature on which parts of the ear the branch actually reaches, with the cadaveric maps differing between individuals. The other is a handheld device pressed against the side of the neck.

In the PRESTO trial the neurologist Cristina Tassorelli and colleagues randomized 248 people with episodic migraine to that device or a sham version. Their trial found more pain freedom at 30 and 60 minutes than sham, and missed its primary endpoint at 120 minutes. In the ACT2 trial for cluster headache, the neurologist Peter Goadsby and colleagues found no difference from sham across the whole group.

Across everyone treated, 14 percent responded against 12 percent on sham. The episodic subgroup showed a large effect, 48 percent against 6 percent. In the chronic subgroup it was 5 percent against 13 percent, which is to say the device did nothing there and the sham did slightly more.

Read the field as a whole and a pattern appears that most summaries miss. The same intervention, delivered to different bodies, produces strong effects, no effects and opposite-looking effects. That pattern is not noise to be averaged away. It is the finding.

Ice baths, humming and gadgets, tested against sham

Search the phrase vagus nerve and you will be sold a dozen ways to hack it in ninety seconds. Some of it carries real evidence, most of it far less than the marketing implies, and the difference is measurable.

Ear clips fail their central claim

Start with the ear clips and neck devices sold for calm, because there the evidence is unusually clear. Vanessa Wolf and colleagues asked whether short sessions of transcutaneous ear stimulation actually change vagally mediated heart rate variability, which is the biomarker the whole field has leaned on.

They pooled sixteen single-blind studies comparing real stimulation to sham in healthy volunteers using Bayesian methods, which can distinguish an absent effect from an untested one. Their living meta-analysis reported strong evidence for the null, with an effect size near zero and odds of roughly 25 to 1 favoring no effect. That is not a study failing to find something. That is a body of evidence actively saying the effect is not there.

Cold engages the brake, and the famous protocol works through the accelerator

Cold is more interesting than the debunkers allow. Sylvain Laborde and colleagues pooled randomized trials of physical recovery methods used after hard exercise, asking which of them speed the return of parasympathetic activity. Their meta-analysis of post-exercise recovery techniques found a moderate to large effect for cold water immersion on the fast vagal measure, larger than any other technique tested. Cold on the face and body does engage the brake. That much is real.

What is not established is the leap from that to the claims stacked on top of it. The clearest test comes from an unexpected direction. The intensive care researcher Matthijs Kox and colleagues trained volunteers for ten days in the Wim Hof method, which combines meditation, forceful breathing and immersion in ice water.

All participants were then injected with bacterial endotoxin to provoke a controlled inflammatory response. Their trial found that the trained group had a genuinely blunted inflammatory response and fewer flu-like symptoms. The mechanism, though, was the opposite of the popular story.

The training produced a surge of adrenaline through voluntary activation of the sympathetic system, and the anti-inflammatory effect tracked that surge. The ice bath did something real to inflammation, and it did it through the accelerator. One warning before anyone copies the protocol.

Cold immersion is not for people with heart or blood pressure conditions, and the cold shock response can provoke dangerous rhythms in a vulnerable heart. Forceful breathing drills should never be done in water or before entering it, because they blunt the urge to breathe and can cause a blackout under the surface.

Humming rests on a pilot; slow breathing rests on 223 studies

Humming and chanting are the softest claim of all. The most cited study is a small pilot in twelve volunteers, in which Bangalore Kalyani and colleagues imaged the brain during audible chanting to see which regions changed. Their pilot imaging study reported deactivation in limbic regions and raised the possibility that vibration might stimulate vagal branches in the ear.

Notice what was measured. Brain blood flow, in twelve people, with no measurement of vagal activity at all. The hypothesis is reasonable and the evidence is a pilot. Anything stated more strongly than that is somebody selling something.

Breathing is where the evidence is strongest, which is a pleasant irony given that it is free. Laborde and colleagues, in a separate review, pooled 223 studies of voluntary slow breathing and its effect on vagally mediated variability. Their meta-analysis found increases during the breathing, immediately after a single session, and after multi-session training.

The psychologists Paul Lehrer and Richard Gevirtz then asked why the effect is so reliable. Their account of how heart rate variability biofeedback works points at resonance. Breathing at roughly six breaths a minute puts the breathing rhythm in step with the natural delay of the baroreflex, so the two reinforce each other and the oscillation grows.

Sit with that last mechanism, because it is the model's argument in miniature. Slow breathing works by putting two coupled rhythms into phase with each other. No button is pressed anywhere.

The vagus is one voice in the chord of tone

The physiology above is established, and it leaves a question hanging. Why does the same nerve, stimulated the same way, transform one person and do nothing for another? The Unified Model of Tone answers that question directly.

Tone, in this model, is the integrated organization of the body's state, held across every scale at once. It rides many mediums at the same time. It is carried electrically along nerves, mechanically in the tension of tissues, chemically in transmitters and cytokines, and hydraulically in the pressure of blood and fluid.

The same information, present in many forms, held in step. Health is the width of the range this organization can move through and return from. Illness is the collapse of that range to a narrow, stuck setting.

Read the vagus through that lens and its role changes shape. Popular writing crowns the vagus the seat of calm. The model reads it instead as one voice in a coupled chord, and largely the voice that reports rather than commands. The heart rate variability you measure at the wrist is the sound of that chord, not the note the vagus is playing.

The breath sets its rhythm. The baroreflex sets its timing. The vascular tone sets its amplitude. The immune state, the fluid state and the appraisal running in the prefrontal cortex all voice into it. A number that so many systems write into cannot belong to any one of them.

A measured variable is a chord sounded by many coupled voices. Silence one voice and you have not found the cause. You have removed one singer from a choir that will go on singing.

Input meeting tone explains the split trials

This is why the stimulation literature looks the way it does, and the model treats that literature as a result rather than a disappointment. When an electrode drives the vagus, it does not restore an organization. It injects a signal into one channel of a system that is already coupled everywhere.

Where the dysregulation genuinely lives in that channel, the input lands and the effect is large. Where the distortion is spread across the other voices, the system routes around the input and the trial reads null. The same current meets a different tone and becomes a different event.

Look back at the trials with that reading in hand. Cluster headache splits cleanly, with a 48 percent response in the episodic form and nothing in the chronic form, in the same trial with the same device. Heart failure improves in how patients feel and function while the pump measurements do not move.

Depression shows almost nothing at ten weeks and something at a year, in a condition where the organization has to reorganize slowly if it reorganizes at all. These results are the signature of an input meeting different tones, not contradictions to be averaged into a modest pooled estimate.

Two objections, answered in place

And the third objection, the one that matters most. If the model explains strong effects, absent effects and opposite effects, does it predict anything at all? It does, and here is the prediction with teeth.

Restoring tone versus masking output, and the test that decides

A drug that lowers heart rate lowers it in everyone who takes it. That is what a good drug does. It overrides an output reliably and in one direction, and for many people that override prevents a stroke or a death, which is not a small thing. An electrode driving a nerve does the same kind of work in a different medium. It overrides an output by supplying current.

The model calls that masking, and uses the word descriptively rather than as an insult. Masking manages the output. The organization underneath it is unchanged, which is why the management must continue.

Restoring tone is a different aim. It means widening the range the system can move through on its own, so that the output settles where the moment requires without being held there. A restored system does not merely have a better number. It has its range back.

Bidirectional restoration is the prediction with teeth

That distinction sounds philosophical until you make it measurable, so here is the model's sharpest prediction. A genuine tonal correction moves a dysregulated value toward the healthy middle from either side. If a group of people begins with values that are too high and another group begins with values that are too low, a real restoration moves both toward the center. A push-intervention cannot do that. It moves everyone the same direction, because that is what a push is.

Call it bidirectional restoration, and treat it as the test that tells restoring from masking. The experiment is not exotic. Take a measure with a healthy middle, such as resting heart rate, heart rate variability or baroreflex sensitivity. Split participants by whether they start above or below their expected range.

Apply the intervention and look for convergence on the middle rather than a uniform shift. If a claimed tonal intervention pushes everyone the same way, it is a push-intervention wearing different language. It masks the output instead of restoring the regulation, and it carries half the sample further from the middle while it helps the other half.

Notice how naturally the vagal literature already sits with this prediction. Slow breathing raises variability most in those who start low. Cold water immersion produces its clearest effect in the hours after hard exercise, when the brake has been driven off. The effects cluster where there is range to recover. That is suggestive, and it is not proof, and the difference between those two words is a discipline this library keeps.

What is worth measuring, and what still needs a doctor

The return matters more than the value. Track your own resting heart rate over weeks, your own variability trend against your own baseline, and how quickly both return to their usual place after a hard day, a bad night or an illness. A system with range moves and comes back. A system without range either sits stuck or fails to recover.

The second is what is worth doing, stated without overreach. Slow breathing at around six breaths a minute has the best evidence of any freely available practice for raising vagally mediated variability, and it works through a coupled mechanism rather than a switch. Regular aerobic movement, adequate sleep and treatment of sleep-disordered breathing all act on the same regulation. None of this is a treatment for a disease, and none of it should replace care that is working.

The third must never be dropped. Findable causes must still be found. A slow heart rate can come from a conduction problem in the heart. Fainting, swallowing difficulty, a hoarse voice that will not resolve, unexplained weight loss and a heart rate that will not rise with exertion are findings for a physician, not readings for an app. A model that explains regulation is not a reason to skip a diagnosis, and no page on the internet can examine you.

What the model offers is a different question to ask alongside the standard one. Medicine asks what is damaged. The Unified Model of Tone asks what has lost its range, and whether it can be given back.

The vagus is one of the clearest places in the body to watch that question play out, because it is the line the body uses to report itself to itself. It is a magnificent nerve. It is not a switch, and the people selling you the switch are selling you the smallest possible version of it.

How the vagus nerve relates to the rest of the library

The vagus is anatomy: one cable inside a regulation the whole library studies from different windows. Three distinctions keep the neighboring pages straight, and each neighbor leans on this nerve in a specific way.

Start with the two pages readers conflate with this one. The autonomic nervous system is the system; the vagus is one nerve inside it, the main parasympathetic cable of the chest and abdomen, and nothing here makes sense without that anatomy. Heart rate variability is the instrument; this page is the physiology that instrument samples, and the blockade experiments of 1981 and 1985 are why its fast component earned the vagal label.

• Dysautonomia is what it looks like when the regulation the vagus serves becomes the presenting problem, with the baroreflex and the vagal brake at the center of the testing.
• Inflammation runs on the cytokines the vagus both senses and restrains, and the inflammatory reflex is the circuit connecting the two pages.
• Autoimmune conditions are where that reflex met human disease, in the rheumatoid arthritis stimulation work.
• Gut health travels the vagal line in both directions, from neuropod cells signaling in milliseconds to bacteria whose behavioral effects vanish when the nerve is cut.
• Anxiety is the state read by the same amygdala and prefrontal circuitry that tracks heart rate variability, the appraisal side of the brake.
• And neurophysiology holds the ground rules, from Loewi's chemical transmission to the blockade logic, that every claim on this page stands on.

The chord itself is defined on the tone page. Tone within its healthy range is health, because the organization keeps the flexibility to adapt. Tone that drifts or distorts outside that range is what manifests as illness and disease, and the vagus is one of the clearest places to watch which of the two is happening.

Frequently asked