Slow breathing and the vagus nerve: how pranayama calms you

Six breaths a minute is not folklore. It lines your breath up with the body's blood-pressure reflex. Here is the mechanism, the trials, and where the claims run ahead of the data.

Pages of a 19th-century Devanagari manuscript copy of the Hatha Yoga Pradipika
Photo: Ms Sarah Welch, CC BY-SA 4.0, via Wikimedia Commons

Slow breathing calms you mainly through the vagus nerve's control of the heart: each long exhale lets vagal output slow the pulse, and breathing near six breaths a minute (about 0.1 Hz) lines the breath up with the baroreflex, the blood-pressure feedback loop, so heart-rate swings grow several-fold. That raises vagally mediated heart-rate variability within minutes. The mood and anxiety benefits in trials are real but modest, and the strongest evidence is for short-term effects, not cures.

Key takeaways
  • ~6 breaths a minute (0.1 Hz) is the resonance rate where breath and baroreflex push heart rate in step.
  • The long exhale is the calming phase: vagal output to the heart rises as you breathe out.
  • Reciting the Ave Maria or a mantra slowed 23 Italian adults to about 5.6 breaths a minute (Bernardi, BMJ 2001).
  • 5 minutes a day of cyclic sighing lifted mood more than mindfulness in one small Stanford trial (2023).
  • In patients, pranayama's pooled effect is small: SMD −0.27 against passive controls, none against standard care.

"Breathe slowly and you will calm down" is advice every Indian child hears from a parent or a yoga teacher. The physiology behind it is now well mapped, which is useful, because it also tells you what slow breathing cannot do.

This post walks through the mechanism (respiratory sinus arrhythmia, the baroreflex, the "vagal brake"), the numbers from the best trials, and what a 2025 meta-analysis found when it pooled pranayama trials in patients.

~6/minresonance breathing rate, about 0.1 Hz (Lehrer & Gevirtz 2014)
5 min/daydose in the Stanford breathwork trial (Balban et al. 2023)
SMD −0.27pranayama vs passive controls in patients (Frontiers in Psychiatry 2025)

What does the vagus nerve have to do with breathing?

The vagus nerve (cranial nerve X) carries the parasympathetic "brake" signal to the heart's pacemaker, and that signal rises and falls with every breath. When you breathe in, vagal output to the sinoatrial node is briefly damped and the heart speeds up; when you breathe out, the brake comes back on and the heart slows.

This beat-to-beat rhythm is called respiratory sinus arrhythmia (RSA), and it is the reason a healthy resting pulse is never perfectly regular.

Why the exhale does the calming

Two points follow:

  • You steer the nerve through the breath. You cannot control the vagus nerve directly, but you can control the breath, and the breath gates vagal output.
  • A long exhale buys more brake time. It spends more of each cycle in the "brake on" phase. That is why most calming techniques, from bhramari (humming bee breath) to the 2:8 inhale-exhale pattern used in a 2026 Neuron study, stretch the out-breath.
One slow breath: 4 s in, 6 s out (6 per minute) Inhale 4 s Exhale 6 s 0 s4 s10 s Heart rate brake eases,pulse rises brake back on,pulse falls Longer exhale = more of each cycle with the vagal brake applied The size of this swing is what HRV measures (RSA)
Respiratory sinus arrhythmia in one breath. The curve is schematic, but the direction of each phase follows the physiology described by Lehrer and Gevirtz (2014).

How researchers read "vagal tone"

Heart-rate variability (HRV) is the standard way to read this. Measures such as RMSSD and high-frequency (HF) power track the beat-to-beat changes that the vagus produces, so "higher vagal tone" in a paper usually means "higher vagally mediated HRV at rest". It is an index, not a direct reading of the nerve.

Slow breath~5 s in, ~5 s out Heart rateup on inhale, down on exhale Blood pressureoscillates in turn Baroreceptorscarotid sinus, aortic arch Brainstemsets vagal output(the "vagal brake") RSA At ~0.1 Hz the breath and the baroreflex push in step, so heart-rate swings grow
The breath-heart-baroreflex loop. Breathing near six a minute times the breath to the reflex's own delay, which is why heart-rate swings are largest there (after Lehrer and Gevirtz 2014).

Why does six breaths a minute matter?

Six breaths a minute (0.1 Hz) is close to the resonance frequency of the cardiovascular system, the rate at which breathing and the baroreflex reinforce each other. Lehrer and Gevirtz's 2014 review in Frontiers in Psychology sets out the mechanism.

At about 0.1 Hz, heart rate and blood pressure oscillate 180 degrees out of phase, and breathing at that rate is in phase with heart rate. So every breath pushes the reflex in the same direction it was already going.

The story

In Pavia and Florence, Bernardi's team was studying how reading and mental arithmetic shift the balance of the nervous system. They wanted a calm, low-arousal control task, so they had volunteers recite the rosary. The surprise was the timing: one cycle of the Ave Maria in Latin, priest's line and congregation's reply, took almost exactly 10 seconds, the same period as the body's slow blood-pressure waves.

So they tested it properly on 23 healthy adults with no yoga experience, adding the mantra om-mani-padme-om. Breathing fell from 14.1 a minute at rest to 5.6 during the prayer and 5.7 during the mantra, as regular as breathing to a metronome. Baroreflex sensitivity rose from 9.5 to 11.5 ms/mm Hg compared with free talking. The BMJ printed it on 22 December 2001.

Bernardi L et al.: Effect of rosary prayer and yoga mantras on autonomic cardiovascular rhythms, BMJ 323:1446 (2001)

How the baroreflex loop works

Pressure sensors in the carotid sinus and aortic arch detect a rise in blood pressure, and the brainstem answers by increasing vagal output to slow the heart; a fall in pressure does the opposite. The reflex responds with a lag.

When the breathing cycle is about 10 seconds long, that lag and the breath line up, and the swings add together instead of cancelling.

How big and how fast the effect is

The effect is large and fast. Lehrer and Gevirtz report that HRV amplitude can rise to many times its resting level within a fraction of a minute, even in people new to the technique, and baroreflex gain rises during sessions.

After about three months of twice-daily home practice, resting baroreflex gain was also higher, which they read as a trained change in the reflex. They give a refined average resonance of about 0.09 Hz (5.5 breaths a minute), with real variation between people: taller people and men tend to sit lower.

A prayer line that lasts ten seconds and a blood-pressure wave that lasts ten seconds: Bernardi's team suggested the match may be no coincidence.Summarising Bernardi et al., BMJ 2001

What breathing rate does your pattern give?

What does slow breathing do to the brain and body?

Heart, brainwaves and mood

A 2018 systematic review by Zaccaro and colleagues found that breathing at 10 breaths a minute or slower reliably raised HRV and RSA and shifted the EEG towards more alpha and less theta power. They screened 2,461 abstracts and kept 15 studies that actively slowed the breath (pranayama, paced breathing), excluding passive breath-awareness and mixed yoga packages.

The psychological findings in that review were softer: reports of relaxation, comfort, more vigour and lower arousal, with some studies showing less anxiety, anger or confusion, and many null results.

The studies were small (roughly 9 to 58 people), none was a randomised trial, and the authors could not pool them in a meta-analysis. They also flag a technical trap: at slow rates, the breathing rhythm itself falls into the low-frequency (LF) HRV band, so a rise in "LF power" during slow breathing is partly the breath showing up in the signal.

Slow breathing and risk-taking

Newer work links slow breathing to decision-making. In a 2026 Neuron study from the German Institute of Human Nutrition and Charité, 41 healthy adults made gambling choices in an fMRI scanner while breathing normally or with a 2-second inhale and 8-second exhale.

Slow breathing lowered heart rate, raised HRV, increased activity in the ventromedial prefrontal cortex and precuneus, and made people more willing to take risks, apparently by raising sensitivity to rewards rather than lowering fear of losses. Calmer is not the same as more cautious.

How Breath-Control Can Change Your Life: A Systematic Review on Psycho-Physiological Correlates of Slow BreathingZaccaro et al. · Frontiers in Human Neuroscience · 2018

A PRISMA review of MEDLINE and Scopus that kept 15 of 2,461 screened studies on techniques that slow breathing to 10 breaths a minute or fewer. Slow breathing generally increased HRV and RSA, and several studies found more EEG alpha and less theta. Emotional outcomes leaned positive (relaxation, vigour, lower arousal) but many were non-significant. Limit: small within-subject and pre-post studies, no randomised trials, so the review describes acute physiology better than lasting benefit.

Read the paper (free)

Which breathing technique worked best in the Stanford trial?

In a 28-day remote trial at Stanford, five minutes a day of cyclic sighing (two inhales, then a long full exhale) raised daily positive mood more than mindfulness meditation, and lowered resting breathing rate. Balban and colleagues published it in Cell Reports Medicine in January 2023.

Of 108 people enrolled, 30 did cyclic sighing, 21 box breathing (equal inhale, hold, exhale, hold), 33 cyclic hyperventilation with breath retention, and 24 a mindfulness practice of watching the breath.

Daily rise in positive affect after 5 minutes of practice (28-day average)
See the data as a table
Positive affect, pre-to-post change (points)
Mindfulness meditation1.22
Cyclic sighing1.89
Box breathing1.84
Cyclic hyperventilation1.97

Source: Balban et al., Cell Reports Medicine 2023 · Only cyclic sighing differed significantly from mindfulness in the mixed model; the mindfulness change itself was borderline (p = 0.06).

Every group got calmer. State anxiety and negative affect fell in all four arms (p < 0.0001 within groups) with no significant difference between breathwork and mindfulness. The table puts the numbers side by side.

Protocoln enrolledPositive affect changeState anxiety changeNegative affect change
Mindfulness meditation24+1.22−3.95−1.62
Cyclic sighing30+1.89−3.85−1.48
Box breathing21+1.84−3.75−0.83
Cyclic hyperventilation with retention33+1.97−1.81−0.62

The fine print on the Stanford trial

Read the fine print before you treat this as settled:

  • Registration: the trial was registered after it ran.
  • Adherence: practice was self-reported, and subgroups were small.
  • Sample: most participants came from a Stanford undergraduate class.
  • Devices: heart data came from a WHOOP wearable while one co-author was a WHOOP advisor.
  • Null results: resting heart rate and HRV did not change in any group, and neither did trait anxiety or sleep.

The honest summary is that a few minutes of slow, exhale-heavy breathing lifts mood on the day, and that cyclic sighing was the standout in one small trial.

Brief structured respiration practices enhance mood and reduce physiological arousalBalban et al. · Cell Reports Medicine · 2023

A remote randomised trial: 108 adults did 5 minutes a day for 28 days of cyclic sighing, box breathing, cyclic hyperventilation, or mindfulness. Breathwork raised positive affect more than mindfulness (+1.91 vs +1.22 points), driven by cyclic sighing, which also cut respiratory rate the most. Anxiety and negative affect fell in all groups equally. Limit: retrospective registration, self-reported adherence, small arms, a student sample and wearable-based physiology, with no change in resting HRV.

Read the paper (free)
First page of a Sanskrit commentary on the Hatha Yoga Pradipika in Devanagari script
The opening page of a commentary on the Hatha Yoga Pradipika, from the Raghunath temple library. Hatha yoga texts prescribed breath ratios centuries before anyone measured a baroreflex.Photo: Ms Sarah Welch, CC BY-SA 4.0, via Wikimedia Commons

Is pranayama scientifically proven to reduce anxiety?

Partly. Slow pranayama reliably lowers arousal in the short term, and small randomised trials show less state anxiety, but the best pooled evidence in patients shows only a small effect, and it does not beat standard treatment. The cleanest test is a 2025 systematic review and meta-analysis in Frontiers in Psychiatry by Mütze, Mitzinger and Haller at University Hospital Essen.

What the 2025 meta-analysis found

They found six randomised trials (seven publications, 517 adult patients) of pranayama for PTSD, depression and mixed non-psychotic disorders. Against passive controls such as wait lists, pranayama reduced symptom severity by a standardised mean difference of −0.27 (95% CI −0.52 to −0.03) in intention-to-treat analysis.

Against standard care (psychotherapy, antidepressants, ECT) there was no significant difference. When the high-risk-of-bias studies were removed, the intention-to-treat effect stopped being significant. Most trials used Sudarshan Kriya Yoga, which mixes slow and fast breathing, and the samples were 94% male.

The safety signal

Their safety finding matters for anyone choosing a technique. Adverse events were reported more often with fast techniques, especially kapalabhati, and included anxiety, breathlessness, dizziness and headache in one German PTSD trial, where they led to dropouts. Safety reporting across trials was too thin to pool.

The authors' conclusion is plain: pranayama may add short-term benefit to care, but should not replace standard therapy.

Effectiveness of pranayama for mental disorders: a systematic review and meta-analysis of randomized controlled trialsMütze, Mitzinger & Haller · Frontiers in Psychiatry · 2025

Pooled six RCTs (517 patients with PTSD, depression or mixed disorders). Pranayama beat passive controls on symptom severity (SMD −0.27, 95% CI −0.52 to −0.03) but not standard care; quality of life improved only in per-protocol analysis (SMD 0.59). Fast techniques, especially kapalabhati, had more adverse events. Limit: few trials, high risk of bias in two-thirds of domains, unblinded therapists, mostly male samples and doses of 1 to 6 weeks.

Read the paper (free)

A smaller randomised trial in healthy people points the same way. Novaes and colleagues (Frontiers in Psychiatry, 2020) gave 30 young adults four weeks of pranayama or a control activity (crosswords, card games). The pranayama group reported lower state anxiety and negative affect and showed changed activity in the amygdala, anterior cingulate, anterior insula and prefrontal cortex during an emotion task. The authors call it preliminary, and with 30 people it is.

Slow or fast pranayama: which calms and which arouses?

Slow, exhale-led techniques push the system towards vagal dominance; fast techniques do the opposite in the moment and carry more side effects. That difference is why "pranayama" is too broad a word for a research question.

PatternExamplesRateMain acute effectEvidence and caution
Slow, paced, long exhaleResonance breathing, nadi shodhana done slowly, bhramari~4.5 to 7 per minHigher RSA and HRV, lower arousalConsistent acute physiology (Zaccaro 2018; Lehrer & Gevirtz 2014)
Double-inhale sighCyclic sighingSlow, exhale-heavyLower breathing rate, better moodOne small trial (Balban 2023)
Equal-ratio with holdsBox breathingSlowCalming; mood gain similar to sighingOne small trial (Balban 2023)
Fast, forcefulKapalabhati, bhastrika, cyclic hyperventilationFastArousing in the momentMore adverse events reported (Frontiers in Psychiatry 2025)
Mixed cyclesSudarshan Kriya YogaSlow and fastVaries within a sessionMost of the patient RCTs; cannot isolate which part works
A scanned page from a 1919 textbook on the anatomy and physiology of the autonomic nervous system
A page from a 1919 textbook of Vegetative Neurology. The autonomic system's two branches were described a century ago; measuring the vagal brake beat by beat came much later.Photo: Internet Archive Book Images, No restrictions, via Wikimedia Commons

Does slow breathing "reset" the nervous system?

No. Slow breathing shifts autonomic balance for as long as you do it and somewhat afterwards, but nothing is reset, and a healthy system is meant to swing between arousal and recovery.

An October 2026 piece in The Irish Times, syndicated from The Guardian, made this point against the wave of "nervous system reset" content: the target is flexibility, the ability to come back down after stress, not permanent calm. It cites the same practical dose researchers use, about six breaths a minute (roughly 5 seconds in and 5 out) for 5 to 10 minutes.

Three limits are worth keeping in mind:

  • HRV is an index. A higher RMSSD during a session shows the vagal brake working harder; it is not proof of a lasting change in "vagal tone".
  • Acute is not chronic. The Stanford trial found no change in resting HRV after 28 days. Lehrer and Gevirtz report resting baroreflex changes only after about three months of twice-daily practice.
  • Expectation is hard to blind. Breathing trials cannot hide the intervention from participants, and anxiety scores are self-reported. The replication crisis in psychology is a reminder of how far small, unblinded samples can drift.

How would I use this, practically?

Here is how I would apply the evidence, keeping to what the studies actually tested:

  1. Pick slow over fast. For calming, use an exhale-heavy pattern near six a minute (for example 4 in, 6 out). Leave kapalabhati for when you want arousal, and learn it with a teacher.
  2. Dose it like the trials. Five minutes a day was enough to move mood in the Stanford trial; 5 to 10 minutes is the common resonance-breathing prescription.
  3. Expect day-level effects. The clearest benefit is lower arousal and better mood on the day. Lasting change in baroreflex function took months of twice-daily practice in HRV-biofeedback work.
  4. Do not swap it for treatment. For a diagnosed anxiety disorder, PTSD or depression, the 2025 meta-analysis supports pranayama as an add-on only.
  5. Notice what it does to decisions. If the 2026 Neuron result holds, a calm exhale may make you bolder, which is good before a hard conversation and less good before a bet.

If you want to compare techniques head to head, the evidence-ranked list is in pranayama techniques ranked by research, and the history of laboratory yoga research in India is in Kuvalayananda and the science of yoga. For the attention side of meditation, see the default mode network and the sceptical counterweight in mindfulness hype vs evidence. The wider set of essays on the mind lives on Manthan.

Questions people ask

What does "increased vagal tone" actually mean?

It means the vagus nerve is exerting more of its slowing influence on the heart, which researchers infer from higher vagally mediated HRV (RMSSD, HF power) at rest. It is a statistical index of parasympathetic control, not a direct measurement of the nerve. Slow breathing raises it during practice; whether it stays raised depends on months of regular practice.

Is pranayama proven to reduce anxiety?

For short-term state anxiety in healthy people, small trials say yes (Novaes 2020; Balban 2023, where mindfulness worked about as well). For diagnosed disorders, the 2025 meta-analysis of six RCTs found a small benefit over passive controls (SMD −0.27) and none over standard care, so it is an add-on, not a replacement.

Should I breathe into the chest or the belly?

The physiology above depends on rate and exhale length, not on a particular style. Slow breathing at around six a minute is easier with a full, relaxed diaphragmatic breath, because shallow chest breathing at that rate tends to leave you short of air. Zaccaro's review notes that many studies did not even report nasal versus mouth breathing, so claims about one "correct" style run ahead of the data.

Is kapalabhati dangerous?

It is not dangerous for most healthy people under instruction, but it is the opposite of calming in the moment, and the 2025 meta-analysis found adverse events (dizziness, breathlessness, headache, anxiety) reported more often with fast techniques, especially kapalabhati. People with heart disease, uncontrolled blood pressure, epilepsy or pregnancy should get medical advice first.

How quickly does slow breathing work?

The heart-rate swings at resonance appear within a minute (Lehrer & Gevirtz 2014). Mood effects showed up after single 5-minute sessions in the Stanford trial. Changes in resting baroreflex function took about three months of twice-daily practice.

Sources

  1. Lehrer & Gevirtz: Heart rate variability biofeedback: how and why does it work? Frontiers in Psychology (2014)
  2. Zaccaro et al.: How Breath-Control Can Change Your Life: A Systematic Review on Psycho-Physiological Correlates of Slow Breathing, Frontiers in Human Neuroscience (2018)
  3. Balban et al.: Brief structured respiration practices enhance mood and reduce physiological arousal, Cell Reports Medicine (2023)
  4. Mütze, Mitzinger & Haller: Effectiveness of pranayama for mental disorders: a systematic review and meta-analysis of RCTs, Frontiers in Psychiatry (2025)
  5. PsyPost: Four weeks of pranayama reduces anxiety and negative affect (on Novaes et al., Frontiers in Psychiatry 2020) (2020)
  6. ScienceDaily: Slow breathing impacts inter-organ dynamics modulating brain function and risk behavior (Huang, Park et al., Neuron) (2026)
  7. The Irish Times: There is no such thing as resetting the nervous system (2026)
  8. Bernardi L et al.: Effect of rosary prayer and yoga mantras on autonomic cardiovascular rhythms: comparative study, BMJ 323(7327):1446-1449 (2001)
  9. Ministry of Health and Family Welfare: Tele-MANAS
  10. Psychology Stack Exchange: What is meant by increased vagal tone?
slow breathingvagus nervepranayamaheart rate variabilitybaroreflexresonance breathingcyclic sighing
Sapta

Sapta (Saptarshi Sarkar) is a mechanical engineer turned marketer in Bengaluru and a solo touring rider. He writes the seven Saptaverse blogs. Work with him or read more of Mindblog.

Keep reading in Mindblog

From the other blogs