Slow breathing at about six breaths a minute has by far the best evidence. It raises heart rate variability (HRV) strongly while you do it and for a while after. HRV biofeedback HRV biofeedback A breathing practice in which you watch your heart rate live and breathe slowly, often near 6 breaths a minute, to make the heart's rise and fall with each breath as large as possible. More about HRV biofeedback builds on the same breathing. Cold water on the face gives a short rise. Humming seems to work because it slows your breathing. Gargling has no HRV study at all, and pressing points on the ear rests on one research group.
What “vagus nerve exercise” can mean
The vagus nerve is the main nerve of the calming branch of the nervous system, and its fast signals to the heart drive most beat-to-beat HRV. Nobody can measure its activity at the heart directly in everyday life, so researchers judge it from HRV, mainly RMSSD RMSSD Root mean square of successive differences: an HRV metric built from how much each heartbeat interval differs from the next, in milliseconds. It mainly reflects vagal activity. More about RMSSD and high-frequency power HF power The variance of the heartbeat intervals that falls in the 0.15 to 0.40 Hz band, in ms²: the range of normal breathing. It mainly reflects vagal activity when breathing is 9 to 24 breaths a minute. More about HF power (Laborde and colleagues, 2017). On this page an exercise “works” if studies found higher HRV, the window most apps and studies use. Claims about digestion, inflammation or mood are a separate question that this page does not grade.
Two time scales matter. An exercise can raise HRV during and just after a session, or it can raise your resting HRV, the number you track each morning or night. Most studies measure the first. Fewer have tested the second, and the changes they found are smaller.
The evidence at a glance
We grade evidence as strong when reviews of many studies agree, some when a few small studies agree or a review covers a wider group of methods, little when it rests on one small study or on studies from a single research group, and none when we found no study or pooled trials found no effect.
| Exercise | What studies found | During or just after | Resting HRV over weeks |
|---|---|---|---|
| Slow breathing, about 6 a minute | Higher vagal HRV during practice (172 studies), after one session (16) and after repeated practice (49) | Strong | Some |
| HRV biofeedback | Slow breathing with a live heart rate display; HRV rose against control conditions across 10 remote studies (g = 0.44) | Strong | Some |
| Cold water on the face | Higher vagal HRV in 8 volunteers at rest and a faster return of vagal HRV in 13 men after exercise | Some | None |
| Humming | In two small pilot studies from one research group, raised HRV as much as paced slow breathing in 16 adults; in 24 adults the rise went with slower breathing, not with hearing the sound | Little | None |
| Singing | In healthy 18-year-olds, heart rate rose and fell with the breathing a slow song set | Little | None |
| Cold on the side of the neck | Higher RMSSD during 16-second cold pulses in one lab study of 61 adults | Little | None |
| Gargling | No HRV study found | None | None |
| Pressing points on the ear | Higher SDNN SDNN The standard deviation of all NN intervals in a recording, in milliseconds. It measures overall variability, and its value depends strongly on how long the recording lasts. More about SDNN and RMSSD than a sham for at least 5 minutes in 114 adults, in a trial from one research group | Little | None |
| Ear-stimulation devices | No change in vagal HRV against a sham across 16 studies in healthy people | None | None |
Slow breathing: the best evidence
Your heart already speeds up a little as you breathe in and slows as you breathe out. At about six breaths a minute, that swing lines up with the reflex that steadies your blood pressure, and the heart rate swings grow to many times their size at rest. Lehrer and Gevirtz describe this happening in almost everyone, often within a minute, and put the average best rate nearer 5.5 breaths a minute (Lehrer and Gevirtz, 2014).
A 2022 review by Laborde and colleagues pooled 223 studies of slow breathing. Vagal HRV was higher during the breathing (172 studies), right after a single session (16 studies) and after programmes of repeated practice (49 studies) (Laborde and colleagues, 2022). The pace matters more than the name of the technique. In 84 students, breathing at 6 a minute with no holds raised HRV more than box breathing or 4-7-8 breathing (Marchant and colleagues, 2025).
One caution from the same 2014 review: slower breathing and longer out-breaths can make the swings bigger on their own, possibly without any change in vagal activity Vagal tone The ongoing influence of the vagus nerve on the heart, which slows it down. It is usually estimated from HRV measures such as RMSSD and HF power. More about Vagal tone . A very high number during the exercise shows the exercise working; your resting HRV may not have changed at all.
About 4 seconds in and 6 seconds out gives six breaths a minute. Breathe gently, not deeply, and go back to normal breathing if you feel light-headed. Can breathing exercises raise HRV? covers the patterns and how long to practise. For a full-screen pacer you can set to your own pace, open the breathing pacer.
HRV biofeedback
HRV biofeedback is slow breathing with a screen that shows your heart rate rising and falling, so you can find the pace where the swing is largest. A meta-analysis of 58 randomised trials found small to moderate benefits for anxiety, depression, anger and other outcomes (Lehrer and colleagues, 2020). A 2026 review of 18 studies that delivered it by app or home device, with 1,352 people, pooled the 10 that measured HRV and found a medium effect (g = 0.44) against control conditions (Vann-Adibe and colleagues, 2026). Does HRV biofeedback work? shows how to find your own resonance rate; the chest-strap test can pace you through several rates and rank them.
Cold water on the face
Cold on the face triggers the diving reflex, which slows the heart through the vagus nerve Vagus nerve The main nerve of the parasympathetic nervous system, running from the brainstem to the heart, lungs and gut. Its fast signals to the heart drive most beat-to-beat HRV. More about Vagus nerve . In 8 healthy volunteers, putting the face in cold water raised high-frequency HRV more than bending over an empty basin, whether they breathed or held their breath (Kinoshita and colleagues, 2006). In 13 men, 5 minutes of cold water on the face after hard exercise sped up the return of vagal HRV compared with sitting quietly (Al Haddad and colleagues, 2010). A wider meta-analysis of 24 studies of cold water and cold air found that RMSSD rose and stayed higher for up to about 15 minutes (Jdidi and colleagues, 2024).
That is a short-term response. We found no trial showing that a daily cold splash raises resting HRV. Do saunas or cold plunges raise HRV? covers the same pattern for whole-body cold.
The popular “ice on the neck” tip has one lab study behind it. In 61 adults, 16-second pulses of cold from a small device on the side of the neck raised RMSSD and slowed the heart compared with the same device kept at its starting temperature. Of the three places tested, neck, cheek and forearm, the effect was significant at the neck (Jungmann and colleagues, 2018).
Humming and singing
Humming is often sold as a way to vibrate the vagus nerve. Two small pilot studies from one Korean research group tested it. In 16 healthy adults, humming and paced breathing at 5 seconds in and 5 seconds out both raised HRV above rest, with no difference between them (Woo and Kim, 2025). In 24 adults, humming slowed breathing to near six breaths a minute and raised RMSSD and SDNN; humming with noise-cancelling earphones, so that people could not hear the sound, made no difference. The authors concluded that the effect seemed to come from the long out-breath rather than the vibrating sound (Kim and colleagues, 2026).
Singing points the same way. In a Swedish study of healthy 18-year-olds, singing a slow mantra with breaths only between phrases made heart rate rise and fall with the song, and the singers’ hearts sped up and slowed down together. The authors tied this to the slow, guided breathing that singing demands (Vickhoff and colleagues, 2013).
So if you enjoy humming or singing, it is a pleasant way to breathe slowly. The evidence so far points to the breathing, not the hum, as the active part.
Gargling, ear pressing and other tricks
Lists of vagus nerve hacks often include gargling and rubbing or pressing the ears. We found no study that measured HRV after gargling.
The ear idea has an anatomical basis. A branch of the vagus nerve supplies the skin of parts of the outer ear, and a 2020 review judged the concha and the inner tragus suitable spots to stimulate it (Butt and colleagues, 2020). Pressing a point there has been tested. In a single-blind trial, 114 healthy volunteers in Vietnam had acupressure or a sham at one point in the ear; SDNN and RMSSD rose more with the real pressure, and the rise lasted at least 5 minutes (Trinh and colleagues, 2024). That is one trial from one research group, measured over half an hour. Rubbing the ears, as the online tips suggest, has not been tested.
Vagus nerve stimulation devices
Implanted vagus nerve stimulators are a medical treatment, approved in the US for drug-resistant epilepsy and for depression in some patients (Johnson and Wilson, 2018). The devices sold to the public work differently: they pass a small current through the skin of the ear or neck, known as transcutaneous vagus nerve stimulation.
As a way to raise HRV, they have not delivered. A living meta-analysis, updated as new studies appear, pooled 16 single-blind studies in healthy people and found strong evidence of no effect: acute ear stimulation did not change vagal HRV compared with a sham (g = 0.014) (Wolf and colleagues, 2021). A 2026 review of 6 randomised trials of these devices also found no significant effect on HRV (Souza and colleagues, 2026). In research settings they seem safe: across 51 studies with 1,322 people, the most common side effects were skin irritation (18.2%) and headache (3.6%) (Redgrave and colleagues, 2018).
Researchers are still testing them for other uses. For your HRV, they are not a do-it-yourself fix, and anyone with a heart condition, a pacemaker Sinus node The heart's natural pacemaker, a small patch of cells that starts every normal heartbeat. Left alone it fires about 100 times a minute; the vagus nerve holds it lower at rest. More about Sinus node or epilepsy should ask a doctor before trying one.
What to do with your own data
- Pick slow breathing first. It has the most evidence and costs nothing. Our suggestion for a fair trial is ten to fifteen minutes on most days for at least four to six weeks.
- Never measure during the exercise. Take your daily reading at your usual time, in your usual way, so it stays comparable.
- Compare averages, not days. After a few weeks, compare your 7-day average with your average before you started. The lesson on building your HRV baseline shows how.
- Expect a small change. Age and genes set much of your HRV. Regular aerobic exercise, enough sleep and less alcohol also move resting HRV, as how to increase HRV explains.
When to check with a doctor
Slow breathing is safe for most people. If you have a heart or lung condition, or breathing exercises make you dizzy or anxious, ask your doctor before you practise regularly. Cold water on the face slows the heart, and holding your breath slows it further, so ask first if you have a slow or irregular heartbeat or have fainted before. Breathing and other exercises can support treatment for anxiety or low mood but should not replace it.