Deep breathing for the vagus nerve.

Start with what “activating the vagus nerve” can mean in a study. Breathing trials don’t record traffic on the nerve. They record heart rate variability — mostly the fast, breath-linked part of it — which researchers use as an index of cardiac vagal activity. The index is useful and imperfect, and every claim on this page is a claim about it.

On that measure, slow breathing has a large and consistent evidence base. A 2022 meta-analysis pooled 223 studies and found that vagally mediated HRV rose during slow breathing, immediately after a single session, and after multi-session interventions.

Cold, humming and the rest have evidence too, covered below. We have not found a trial that ranks them against slow breathing head-to-head, so this page doesn’t rank them.

Two breathing variables matter, and rate matters most. Hirsch and Bishop (1981) found that the breath-by-breath heart-rate swing stays large below about six breaths a minute and shrinks steadily above a corner frequency of roughly seven. Depth matters too: in their data the swing scaled linearly with the volume of each breath. The exhale adds something on top, at slow rates:

In that study, a long exhale relative to the inhale produced more reported relaxation, and more high-frequency HRV power at six breaths a minute but not at twelve. In 64 athletes breathing at six a minute, Laborde et al. (2021) found RMSSD higher when the exhale was longer than the inhale, even with an exhale only about a quarter longer. Rate first, exhale second.

Three protocols, and what’s known about each

Three common variants. The research compares them less directly than their marketing does.

Daily resonance breathing (6 breaths per minute). Inhale 5 seconds, exhale 5 seconds. Three to ten minutes. This is the protocol the Lehrer/Vaschillo biofeedback literature is built around. Their explanation is resonance: breathing at a person’s own resonant frequency produces large oscillations in heart rate and blood pressure, and Vaschillo et al. (2006) report that training at that frequency increased baroreflex gain in healthy people. The resonant rate varies by person; six is an average, not a law.

Extended exhale at 1:2 ratio.Inhale 4 seconds, exhale 8. No holds. 3–10 minutes, which works out at five breaths a minute. The closest tested pattern is Van Diest’s long-exhale condition, an inhale-to-exhale ratio of about 1:2.4 at six breaths a minute, which produced more high-frequency HRV power than the reverse ratio. That study had no equal-ratio condition, so it doesn’t show that 1:2 beats resonance breathing.

4-7-8 breathing. Inhale 4, hold 7, exhale 8. Four cycles. Each cycle takes 19 seconds, a little over three breaths a minute. It is popular for winding down. We have not found a trial comparing it with plain slow breathing on vagal measures, so this page makes no claim that the hold adds anything.

The mechanism, in one paragraph

Vagal fibers slow the heart, and the amount of vagal braking is modulated by respiration. The result is respiratory sinus arrhythmia— your heart speeds up on the inhale and slows on the exhale. Slow breathing makes the swing much larger: Hirsch and Bishop measured it at around 20 beats per minute, give or take 5, in seated adults breathing slower than six times a minute. RSA is the main thing HRV-based “vagal” measures pick up, which is why the caveats matter. Breathing rate and depth change RSA directly, differences between people track vagal tone only modestly, and the two can come apart.

Whether weeks of practice raise resting vagal HRV is supported at the level of the Laborde meta-analysis, which found increases after multi-session interventions. It doesn’t tell you how much to expect, and this page no longer quotes a number.

How to verify it’s working

This is the part most “vagus nerve breathing” advice skips. You can’t feel vagal activity directly. Feeling calmer is real and worth having, but it isn’t a measurement of anything vagal.

The best available proxy is HRV, with the caveats above. Laborde, Mosley and Thayer (2017) wrote their methodological recommendations because HRV findings are easy to misconstrue. One easy way to misconstrue them here is to read the effect of breathing slowly as a change in you.

Two specific things to watch:

• Within-session change. HRV rises while you breathe slowly. That is the respiratory effect described above: it confirms you did the exercise, not that anything has changed at rest. A reading taken during or straight after a session is not a baseline.
• Trend over weeks.Track your seven-day rolling average against your own history, not against anyone else’s — Grossman and Taylor’s point that between-person comparisons are weak applies to you too. Sleep, alcohol, illness and training all move it, and “no clear change” is a common, real answer.

HRV Breathe shows both, without verdicts. On Apple Watch, Live Pulse shows the heart-rate swing during a session as it is recorded. The Insights tab plots a rolling average against your own spread, marks the days you practised, and says no clear change when the change is inside your normal scatter.

Other ways to activate the vagus

Several other things move the same measures. None has been compared head-to-head with slow breathing in a trial we know of, so treat this as a list, not a ranking.

• 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 for the popular claim that a few minutes of whole-body cold a few times a week raises HRV for days, so we have dropped it.
• Humming and chanting. They raise HRV, but the evidence points at the breathing, not the vibration. Bernardi et al. (2001) found that reciting the rosary or a yoga mantra at six breaths a minute produced strong increases in cardiovascular rhythms and baroreflex sensitivity. Kim, Lee and Woo (2026), a preliminary study of 24 adults, found that humming slowed breathing toward six a minute and concluded its effects appear tied to that rate rather than to the sound itself.
• Aerobic fitness (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.

None of these compete with each other, and no trial we know of has tested them as a stack. Pick the ones you will actually keep doing.

What “deep breathing for vagus nerve” advice usually gets wrong

• Emphasizing “deep” over “slow.”Depth does matter — the heart-rate swing scales with breath volume — but rate matters more, and one deep breath is not a slow-breathing session. A longer exhale than inhale adds a modest amount at slow rates. The shape matters; the pace matters more.
• Suggesting Wim Hof breathing for the vagus. The best-known controlled trial of it is titled “Voluntary activation of the sympathetic nervous system”: Kox et al. (2014)found profoundly increased adrenaline in volunteers trained in the method. That’s a real effect, in the opposite direction from the one vagus-nerve advice is usually after.
• Treating every breath practice as interchangeable. They aren’t, but the differences are less tidy than the marketing. Box breathing is often dismissed as not vagal, yet in Röttger et al.’s (2021) comparison it produced lower physiological arousal than prolonged exhalation, while prolonged exhalation produced better task performance. In Balban et al.’s (2023) month-long trial, exhale-focused cyclic sighing produced more improvement in mood than mindfulness meditation. Pick the tool for the goal, and know how thin the comparisons are.

Practical recommendation

For most adults who want to practise this:

1. 5 minutes of resonance breathing every morning at 6 BPM
2. Watch the seven-day rolling average rather than any single reading
3. Give it several weeks before reading anything into the average
4. Keep up regular aerobic exercise
5. Protect your sleep

It costs about five minutes a day. The HRV trend will tell you whether your number moved. It can’t tell you why, and it can’t tell you that your vagus nerve did it.