Breathing exercises occupy an awkward position. They're free, they take minutes, they have no obvious downside, and the claims made for them are so uniformly enthusiastic that the whole category starts to sound like folklore. Every technique is presented as calming. None is ever presented as less calming than another.
Which is why one particular study is worth knowing about, because it does the thing this field almost never does: it puts several practices side by side, under the same conditions, and lets them differ.
The study
Balban and colleagues, working out of Stanford, ran a randomised controlled trial published in Cell Reports Medicine in 2023. They recruited 111 healthy participants and assigned each to one of four practices, done for five minutes a day over roughly a month:
- Cyclic sighing — exhale-weighted: a double inhale followed by a long, slow exhale.
- Box breathing — balanced: equal counts for inhale, hold, exhale, hold.
- Cyclic hyperventilation — inhale-weighted, with a retention phase, in the style of certain popular protocols.
- Mindfulness meditation — passive observation of the breath, as the comparison arm.
Daily mood, anxiety and physiological measures were tracked throughout. The design is what makes it useful: five minutes is a dose real people will actually do, and putting all four arms in one trial means the comparison isn't being assembled from separate studies with different populations and instruments.
What it found
All four arms improved mood and reduced anxiety. That's worth stating plainly — nothing here failed, and the meditation comparison arm was not a placebo in any meaningful sense.
But cyclic sighing came out ahead. It produced the largest daily improvement in positive affect — roughly a third greater than the mindfulness arm on the study's mood measure — and, notably, the advantage grew across the month rather than plateauing early.
The more interesting result is physiological. Cyclic sighing was the only arm that significantly reduced resting respiratory rate — and not merely during the exercise. Participants were breathing more slowly throughout the day, outside their practice sessions. Something about five daily minutes of exhale-weighted breathing was shifting a baseline, not just producing a temporary state.
Why the exhale seems to be the active ingredient
Line the three breathing arms up by their exhale-to-inhale ratio and the results order themselves. The most exhale-weighted practice performed best; the most inhale-weighted performed least well among the breathing arms. That's suggestive rather than conclusive from a single trial — but it has a plausible mechanism behind it, and it's one that's been characterised for decades.
Your heart rate isn't steady across a breath. It rises slightly as you inhale and falls as you exhale, a pattern called respiratory sinus arrhythmia, driven by rhythmic changes in vagal influence on the heart. Inhalation transiently withdraws that parasympathetic brake; exhalation restores it. Deliberately lengthening the exhale therefore extends the portion of each cycle in which parasympathetic influence dominates.
Do that repeatedly and you're not just feeling calmer — you're rehearsing a specific autonomic pattern. Which fits neatly with the resting-respiratory-rate finding: the practice appears to nudge a set point rather than only producing a temporary state.
The double inhale that makes cyclic sighing distinctive has its own rationale. Sighs occur spontaneously in mammals and are thought to help reinflate collapsed alveoli; the second sip of air appears to recruit lung volume that the first breath didn't, making the long exhale that follows more effective. It's a naturally occurring reflex, performed on purpose.
The honest caveats. One trial, 111 healthy participants, a month long. People with moderate to severe psychiatric conditions were excluded, so this does not speak to clinical anxiety treatment. Blinding is essentially impossible in a breathwork study — participants know what they're doing. And self-reported mood is a soft endpoint, which is precisely why the objective respiratory-rate finding carries disproportionate weight here.
So is box breathing bad?
No — and this is where the coverage of this study tends to overreach. Box breathing improved mood too. It was outperformed, not falsified.
It's also worth remembering what box breathing was designed for. Its reputation comes from high-pressure operational settings, where the goal is not maximum relaxation but regulated alertness — staying composed and functional under acute stress. A deliberately balanced pattern with holds is a reasonable design for that job. Its symmetry, which looks like a weakness through the lens of this trial's mood endpoints, is arguably the point.
There's a third pattern worth adding, which the trial didn't test: resonance breathing — slow paced breathing at roughly six breaths a minute, near the frequency where cardiovascular oscillations amplify and HRV measures peak. Steffen and colleagues, among others, have studied it for exactly this reason. It's the pattern with the strongest HRV-biofeedback literature behind it and tends to suit longer sessions, where cyclic sighing suits short ones.
| Practice | Shape | Best suited to |
|---|---|---|
| Cyclic sighing | Double inhale, long exhale | Fast down-regulation; evening wind-down; 2–5 minutes |
| Resonance breathing | ~6 breaths/min, even | Longer sessions; deliberate HRV work; 5–10 minutes |
| Box breathing | Equal counts, with holds | Composure under acute pressure; staying alert |
| Cyclic hyperventilation | Inhale-weighted, retentions | Activation rather than calming — not a wind-down tool |
The last row deserves emphasis: inhale-weighted, retention-based protocols are stimulating by design. Used before a workout or a cold exposure, that may be exactly what someone wants. Used at 10pm to get to sleep, it's the wrong tool, and the fact that it appears in the same "breathwork" category as the others is a genuine source of confusion.
The unglamorous conclusion
Five minutes a day, done most days, beat a more sophisticated-sounding practice done the same amount. The benefit accumulated over weeks. There was no version of this where a single heroic session did the work.
Which means the practical problem isn't picking the optimal technique — the gap between the best and worst breathing arm is much smaller than the gap between doing something and doing nothing. It's remembering to do it, at a moment when it fits, in a dose short enough that you don't negotiate your way out of it.
How CalmTide approaches it
CalmTide's Practice hub is built around that last point. It offers cyclic sighing (Balban 2023) and resonance breathing (Steffen 2020) — the two patterns with the most direct evidence for down-regulation — in 2, 5 and 10-minute doses, and suggests which fits the moment: evening wind-down, post-workout, or an elevated-load day when your own data says the day has been expensive.
That context is the part a standalone timer can't do. A five-minute session means something different after a night of poor sleep and two hard training days than it does on a well-recovered Sunday, and CalmTide picks from what your recovery signals are actually saying rather than from a fixed schedule.
Completed sessions save to Apple Health as Mindful Minutes, and the mindfulness component feeds back into your HPA Balance Score — so over weeks you can see whether the practice is showing up in your own numbers rather than taking it on faith. Everything runs on-device: no account, no cloud.
The right practice, at the right moment
CalmTide suggests cyclic sighing or resonance breathing based on what your recovery data is doing today — in 2, 5 or 10-minute doses, saved to Apple Health.
Explore CalmTideFrequently asked questions
What is cyclic sighing?
A double inhale through the nose — a full breath in, then a second shorter sip on top — followed by a long, slow, unforced exhale through the mouth, repeated for a few minutes. Also called the physiological sigh.
Is cyclic sighing better than box breathing?
In the trial that compared them directly, it produced the largest mood improvement and was the only practice to significantly slow resting breathing outside sessions. Box breathing also worked — and its balanced design suits staying composed under pressure rather than winding down.
Why does a long exhale calm you down?
Heart rate rises on inhalation and falls on exhalation (respiratory sinus arrhythmia), reflecting rhythmic vagal influence on the heart. Lengthening the exhale extends the phase in which parasympathetic influence dominates.
How long do you need to practise for it to work?
The trial used five minutes a day for about a month, with benefits accumulating over that period. A single session can shift how you feel now; the baseline changes came from daily repetition.
References & further reading
- Balban, M.Y., et al. (2023). Brief structured respiration practices enhance mood and reduce physiological arousal. Cell Reports Medicine, 4(1):100895.
- Steffen, P.R., et al. (2020). Integrating breathing techniques into psychotherapy to improve HRV: which approach is best? Frontiers in Psychology.
- Lehrer, P.M., & Gevirtz, R. (2014). Heart rate variability biofeedback: how and why does it work? Frontiers in Psychology, 5:756.
- Yackle, K., et al. (2017). Breathing control center neurons that promote arousal in mice. Science, 355(6332).
Related reading: What actually moves your cortisol? · The science behind CalmTide
This article is for general education and is not medical advice. CalmTide provides wellness estimates based on published research and is not a medical device. Breathing exercises are generally low-risk, but retention- and hyperventilation-based practices should not be performed in or near water, while driving, or by anyone who is pregnant or has a cardiovascular or respiratory condition without medical guidance. If you're dealing with persistent anxiety, speak to a qualified healthcare professional.