Can you actually train your vagus nerve?

What does the vagus nerve actually do?

The vagus nerve runs from the brainstem down through the neck and chest into the abdomen, carrying signals between the brain and the internal organs. Most of its fibres send information upward from the body to the brain. It is the main pathway of the parasympathetic system, the branch that slows the heart and supports digestion and recovery.

Because of that role, the vagus nerve gets treated online as a dial you can turn to feel calm. The anatomy is real and the framing is far simpler than the biology.

One number you will meet everywhere is heart rate variability, the variation in time between heartbeats. Higher variability is generally read as more parasympathetic activity, so it is the measure most vagus studies use. RMSSD is the part of it most specific to vagal control, and it is the number to watch when a study makes vagal claims.

Do vagus nerve stimulators work?

The evidence is genuinely split. A 2022 crossover trial of 28 healthy adults found that ear stimulation raised RMSSD, SDNN and high-frequency power against a sham. A larger 2025 trial of 78 adults found increases in overall variability on three protocols, and no change at all in RMSSD, the vagal marker specifically.

Forte and colleagues ran the smaller study in PeerJ, stimulating the cymba conchae of the left ear for ten minutes against a sham placed on the helix. Participants could not tell the two apart by sensation, which strengthens the comparison. The sample was 82 per cent women and the effects were modest.

Atanackov and colleagues published the larger trial in Biomedicines in 2025, testing six combinations of frequency and pulse width across seven sessions per person. Overall variability rose on some settings and followed an inverted U shape, so more stimulation performed worse than moderate stimulation. The vagal marker itself stayed flat.

That distinction is the crux. A device can shift a general autonomic measure while leaving the parasympathetic component unchanged, and only the second would justify the language used to sell these products.

What does the pooled evidence say?

It leans negative. A Bayesian meta-analysis by Wolf and colleagues concluded there is strong evidence for no immediate effect of ear stimulation on vagally mediated heart rate variability. That is a stronger statement than an inconclusive result, because Bayesian analysis can support the absence of an effect rather than only failing to find one.

A 2023 study tested whether stimulation timed to the out-breath would add anything to slow breathing at six breaths a minute. It did not. Slow breathing raised variability on its own, the stimulation added nothing on top, and the authors reported substantial Bayesian evidence for that null.

Worth noting what that second finding implies. The slow breathing worked. The device did not improve on it.

What about humming, gargling and cold water?

These are the methods most people mean by vagus training, and they are the least studied. No controlled trial base supports humming, gargling, or singing as vagal interventions. Cold water on the face triggers a genuine reflex that slows the heart, and that reflex is well documented while its lasting benefits are not.

The gap between how confidently these are recommended and how little they have been tested is the widest in this whole area. That does not make them useless. It means anyone telling you what they do to your nervous system is going past the evidence.

Slow breathing is the exception. It reliably raises heart rate variability during practice, it appears in the study above as the thing that actually worked, and it costs nothing.

So what is worth doing?

Practise slow breathing at around six breaths a minute, which is the intervention with consistent measured effects on variability. Treat ear stimulators as unsettled, and treat any product promising to tone your vagus nerve as making a claim the literature has yet to support.

If you want to track something, use heart rate variability trends over weeks and pay attention to RMSSD specifically. Single readings swing with sleep, alcohol, illness and posture, so a day-to-day number tells you very little.

Sleep, exercise and recovery move variability more reliably than any breathing technique or device. Those are the levers with the weight behind them.