The vagus nerve is the tenth cranial nerve and the main component of the parasympathetic nervous system. Its name means “wandering”: it runs from the brainstem down through the neck and chest into the abdomen, reaching the heart, the lungs and much of the digestive tract. It carries information from those organs back to the brain as well as signals out to them — a point Breit et al. (2018) make about its afferent fibers and the gut in particular.
It is also the nerve most HRV research is asking about. HRV is used to index cardiac vagal tone, which Laborde, Mosley and Thayer (2017)define as “the contribution of the parasympathetic nervous system to cardiac regulation.” The popular version goes further and says every HRV-raising intervention works through the vagus. That is too strong: exercise training, for one, appears to slow the heart through more than vagal change alone, as covered below.
What the vagus actually does at the heart
The heart sets its own pace, and nerves adjust it. Gordan et al. (2015)put the sinoatrial node’s rate at about 100 beats per minute in the absence of nerve and hormone input. Sympathetic fibers push it up; vagal fibers pull it down. A resting heart rate below that is the net effect of vagal slowing outweighing sympathetic drive.
The fastest part of that control follows the breath: heart rate rises a little with each inhale and falls with each exhale. That rhythm, respiratory sinus arrhythmia, is the part of HRV most often used as an index of cardiac vagal activity.
It is an index, not a readout. Grossman and Taylor list the caveats: breathing rate and depth confound it; within one person it tracks vagal control well, but between people the association may be modest; physical activity and beta-adrenergic tone shift it; and RSA and vagal tone can dissociate. “High HRV means a strong vagus” is a reasonable first guess, not a fact about any one reading.
“Unawareness of these issues can lead to misinterpretation of cardiovascular autonomic mechanisms.”
What slow breathing does to it
Two coupled mechanisms.
First, respiratory sinus arrhythmia. Slow breathing makes the breath-linked swing in heart rate much larger. Hirsch and Bishop (1981) found the swing stays large and stable below about six breaths a minute, shrinks steadily above a corner frequency of roughly seven, and grows in proportion to the volume of each breath. Both halves of the breath matter: Strauss-Blasche et al. (2000)found that a short inhale followed by a long exhale enlarged the swing mainly by making heart rate rise more steeply during the inhale, as the vagal brake released — not only by deepening the slowdown on the exhale.
Second, baroreflex resonance. The pressure-feedback loop that steadies blood pressure has its own rhythm. Breathing at a person’s resonant frequency lines the two up, and Vaschillo, Vaschillo and Lehrer (2006)describe the result as large oscillations in heart rate, blood pressure, vascular tone and pulse amplitude. The resonant rate is individual — lower in men and in taller people in their data — which is why six breaths a minute is a starting point, not a law.
What happens over weeks is a narrower claim than the usual “this trains the vagus.” Laborde et al.’s 2022 meta-analysis of 223 studies found vagally mediated HRV higher after multi-session slow-breathing interventions, and Vaschillo’s group reported that resonance training increased baroreflex gain in healthy people. That is the evidence behind “vagal tone improvement.” It is a finding about an index, averaged across many studies, and it doesn’t promise a size for any one person.
Vagal tone vs HRV — the same thing?
Close, but not identical. Vagal tone is the underlying physiological property: how much vagal output your nervous system is producing at the heart. HRV is the observable consequence — the beat-to-beat variation that results. Grossman (2023)calls treating RSA as equivalent to cardiac vagal tone a “category mistake”: confusing an approximate index with the thing it indexes.
Higher vagal tone generally goes with higher HRV. But HRV also moves with breathing rate and depth, physical activity and beta-adrenergic tone, which is the Grossman and Taylor list again. RMSSD is the metric the studies on this page use as their marker of cardiac vagal activity.
For consumer purposes the distinction matters more than it is usually given credit for. A rise in your HRV over weeks of practice is consistent with more vagal activity, but HRV also moves with sleep, alcohol, illness and training, and a wearable can’t tell you which of them moved it.
Other ways to activate the vagus
Several other things move the same measures. None has been ranked against slow breathing in a head-to-head trial we know of.
- Cold exposure. Jungmann et al. (2018) applied a cold thermode to the neck, cheek and forearm of 61 adults. Cold on the side of the neck raised RMSSD and lowered heart rate compared with a no-cold control. We have not found controlled evidence that a few minutes of whole-body cold a few times a week raises HRV for days afterwards, so this page no longer says so.
- Humming and chanting. The usual explanation is that the vagus supplies the larynx, so vibration stimulates it. The studies point elsewhere. Woo and Kim (2025) found humming raised HRV about as much as paced breathing at six breaths a minute, and a follow-up, Kim, Lee and Woo (2026), concluded that humming’s effect appears tied to the slow breathing it induces rather than to the sound itself. Both are small, preliminary studies.
- Aerobic exercise (cumulative). Sandercock et al.’s (2005) meta-analysis found exercise training increased high-frequency HRV power and RR interval. The authors add that the weak link between the two suggests training slows the heart through more than vagal change alone.
- Sleep. Trinder et al. (2001) found that autonomic balance shifts toward parasympathetic dominance during non-REM sleep, mainly through increased parasympathetic activity, while REM sleep looks much like wakefulness.
- Probiotics, fasting, certain dietary changes. Breit et al. describe the evidence that gut bacteria affect mood partly through the vagus as preliminary. We have not found evidence strong enough to say these move HRV, so this page doesn’t.
Polyvagal theory — what to believe
Polyvagal theory, introduced by Stephen Porges in 1995, distinguishes two vagal source nuclei in the brainstem and argues that only fibers from one of them, the nucleus ambiguus, produce respiratory sinus arrhythmia — an arrangement it presents as a mammalian development. Porges (1995) framed it in evolutionary terms from the start. Later work linked the two branches to defensive immobilization and to social engagement, and the theory has become enormously influential in therapy circles, particularly trauma therapy.
The honest version: its core premises have been challenged by the autonomic researchers whose field it borrows from. Grossman and Taylor (2007) argued that it misdescribes how vagal control of heart rate evolved. Grossman (2023) takes its five basic premises one at a time, including the claim that RSA doesn’t occur in reptiles.
“I will here briefly document how each of these basic premises have been shown to be either untenable or highly implausible based on the available scientific literature.”
Grossman is a long-standing critic and the argument is not over. But the burden sits with the theory, and this page doesn’t lean on it.
Practical translation: the techniques that travel under the polyvagal label — slow breathing, humming, cold — have their own evidence, covered above, and none of it depends on the theory being right. The sequential “ladder” of autonomic states is a metaphor, not a finding. Use the techniques; don’t confuse the metaphor with the physiology.
Vagal nerve stimulation devices
Implanted vagus nerve stimulators are used for drug-resistant epilepsy, where a 2025 meta-analysis, Muniyandi et al., found that 46 percent of patients had their seizures cut by at least half. For treatment-resistant depression, Breit et al. describe the evidence as preliminary. Either way it is surgery, not a consumer product.
Consumer ear-clip devices — transcutaneous auricular VNS — are sold partly on HRV. On that specific claim the evidence is fairly clear. A living Bayesian meta-analysis of 16 sham-controlled studies in healthy people found strong evidence that acute stimulation does not change vagally mediated HRV compared with sham.
“To conclude, there is no support for the hypothesis that vmHRV is a robust biomarker for acute taVNS.”
That doesn’t settle whether the devices do anything else. It settles the HRV pitch. If HRV is the reason you’re considering one, slow breathing has the evidence the device doesn’t, and it costs nothing.