Cyclic sighing is the physiological sigh done on repeat. One sigh — inhale, second inhale on top, long slow exhale — is something you do in the moment. Chain those sighs together for five minutes and it stops being a one-off and becomes a practice, and the practice is the version that has been tested.
The distinction matters because most articles collapse the two. They are not the same intervention. The five-minute version is what a Stanford team led by David Spiegel and Andrew Huberman put in a randomized trial in 2023, and it’s the version with the outcome data attached.
The protocol
Every cycle has three parts and takes about eleven seconds. You repeat it for five minutes — roughly 27 cycles.
- Inhale through the nose, about 2.5 seconds. Comfortable, not maximal. You are filling to roughly 80%, leaving room for what comes next.
- Second short inhale on top, about 1.5 seconds. A quick top-up through the nose at the peak. This is the part people skip, and it’s what makes the breath a sigh rather than a slow breath.
- Exhale slowly through the mouth, about 7 seconds. Unforced, all the way to empty. The exhale should be the longest phase by a wide margin.
The exact seconds are not sacred. What is structural is the ratio: a short primary inhale, a shorter top-up, and an exhale roughly as long as both inhales combined plus half again. If 2.5/1.5/7 leaves you short of air, scale the whole pattern down and keep the proportions. If you find yourself straining on the exhale, you inhaled too much on step one.
Nose in, mouth out. The mouth exhale is deliberate — it lets you regulate the outflow rate by narrowing your lips, which is much harder to do through the nose.
What the Stanford trial actually found
Balban and colleagues randomized 108 participants into four arms: cyclic sighing, box breathing, cyclic hyperventilation with retention (the Wim Hof-adjacent arm), and mindfulness meditation, each done for five minutes a day over 28 days. Participants rated their mood and anxiety before and after every session, and wore a WHOOP strap that logged respiratory rate, resting heart rate, heart rate variability and sleep.
Every arm, meditation included, reported less anxiety and less negative mood after each session than before. This was not breathwork versus nothing. The differences were in positive mood and breathing rate. Cyclic sighing was the only breathing arm whose gain in positive mood was significantly larger than meditation’s, and the only one whose respiratory rate, logged daily, fell significantly more over the month than meditation’s. Its mood gain also grew the more days people had practised.
What it didn’t show matters as much. Heart rate variability and resting heart rate didn’t change over the month in any arm, and neither did the WHOOP sleep measures. And the study was not powered to compare the three breathing patterns with each other — the authors say so — so “cyclic sighing beat box breathing” is not a finding.
“Using a mixed-effects model, we show that breathwork, especially the exhale-focused cyclic sighing, produces greater improvement in mood (p < 0.05) and reduction in respiratory rate (p < 0.05) compared with mindfulness meditation.”
Three more caveats, because most coverage of this study omits them. First, 108 people over 28 days is a small, short trial, and the authors describe it as exploratory: it was registered after the fact, not before. Second, the comparison arms were all active interventions that also helped; the headline “beats meditation” holds for positive mood, and the margin was modest. Third, the paper declares that Huberman became an advisor to WHOOP, the wearable the study used, in June 2022. If you already have a meditation practice you like, this is not a reason to abandon it.
Why the long exhale, and what’s still open
Cyclic sighing differs from the other two breathing arms in two ways: the exhale is much longer than the inhale, and the inhale comes in two parts. The trial can’t say which of the two drove its result.
The long exhale has separate support. Van Diest et al. (2014) had 30 people breathe with the exhale either much longer or much shorter than the inhale. With the long exhale they reported more relaxation, and at six breaths a minute it came with more high-frequency HRV power. The double inhale has less. Li et al. (2016), the Stanford and UCLA labs that mapped the sigh circuit in mice, describe spontaneous sighs as occurring “every few minutes to reinflate alveoli.” Neither study tested whether a deliberate top-up, repeated for five minutes, does anything similar.
Cyclic sighing vs the single physiological sigh
Use the single sigh when something has just happened. One to three cycles, thirty seconds, eyes open, mid-conversation if necessary. It is a pause, not a practice.
Use cyclic sighing when nothing in particular has happened. Five minutes, once a day, ideally at the same time. That is the version the trial tested, and in it the mood gain grew with the number of days practised.
Don’t borrow the trial’s results for the single sigh: the evidence above is for five minutes a day over a month, not for one breath in the moment. There is a fuller treatment of the single breath on the physiological sigh page.
Cyclic sighing for sleep
There’s no trial evidence that it helps sleep. The Balban study logged sleep with WHOOP and found no change in any arm. For the last minutes before bed we’d reach for 4-7-8 or a plain extended exhale at a 1:2 ratio. Both run slower than cyclic sighing’s eleven-second cycle and skip the top-up.
If you want cyclic sighing in your evening, put it earlier: the wind-down hour, not the pillow. Five minutes after dinner, before the screens, is a reasonable slot.
Cyclic sighing in HRV Breathe
Cyclic sighing is one of the five techniques in the app. The pacer runs the 2.5 / 1.5 / 7 timing, with the blob over-swelling on the top-up so you can see the second inhale rather than counting it. On Apple Watch you get the same pattern as haptic taps, which is easier to follow with your eyes closed.
One honest note about measurement. Five minutes of cyclic sighing will not produce a reliable before-and-after HRV number, and we don’t pretend otherwise — wrist HRV samples arrive every few hours, not on demand, so a single pre/post pair is mostly noise. What you can track is whether you feel better afterwards, which the app asks you after every session, and what your breathing does over weeks. For the pattern the HRV research centres on, see resonance frequency breathing.
Who should skip it
Cyclic sighing has no breath holds and no deliberate hyperventilation. Still: if you have a respiratory condition, are pregnant, or have a history of fainting with breathwork, clear it with a clinician first. If any breathing exercise makes you dizzy, stop and breathe normally. Lightheadedness is not a sign it’s working; take smaller inhales or stop.