Box Breathing: The Technique Works, the Numbers Are Made Up

Short answer: slow breathing has a genuine, replicated physiological effect, and the 4-4-4-4 pattern specifically has never been shown to be better than any other slow pattern. The systematic review of this literature screened 2,461 abstracts, found 15 usable studies, and explicitly reported no consensus on an optimal ratio.

So the technique is worth using. The precision is theatre. And one variable that the box pattern deliberately ignores is the one with the most evidence behind it.

What is actually established

Zaccaro and colleagues reviewed the psycho-physiological correlates of slow breathing. Their definition of "slow" is the first useful thing: below 10 breaths per minute. Normal resting respiration sits around 12 to 20, so almost any deliberate slowing crosses the threshold.

That is a real and coherent picture. Slow breathing shifts autonomic balance toward the parasympathetic side, and people feel calmer. Nobody serious disputes it.

Slow breathing techniques act enhancing autonomic, cerebral and psychological flexibility in a scenario of mutual interactions.

Zaccaro, Piarulli, Laurino, Garbella, Menicucci, Neri and Gemignani, Frontiers in Human Neuroscience, 2018

Where the four seconds came from

Not from a laboratory. The box pattern reached the public through accounts of US Navy SEAL training, usually via a single former instructor's book, and the "four" is a teaching convenience: it is memorable, it is symmetrical, and it produces a rate of about 3.75 breaths per minute, which is comfortably slow.

None of that makes it wrong. It makes it arbitrary, and the review is explicit that the studies it pooled used varied ratios with no consensus on optimal timing. There is no trial showing 4-4-4-4 beats 5-5-5-5, or 4-7-8, or simply breathing at six per minute without holds.

Which means the confident instruction you will find everywhere — that the counts must be equal, that the holds are essential, that four is the correct number — is presenting a mnemonic as a protocol.

The variable that does have evidence

Here is what the box pattern leaves out, and it is the most supported detail in the whole area.

Exhaling longer than you inhale is the part with mechanistic support. Heart rate rises slightly during inhalation and falls during exhalation — respiratory sinus arrhythmia, the same phenomenon underlying HRV. Lengthening the exhale lengthens the phase in which the parasympathetic system dominates.

Box breathing makes the inhale and exhale equal, which discards that lever entirely.

So which should you use

The honest ranking, given that no head-to-head trial exists:

If you want the simplest thing that satisfies the evidence: breathe at about six breaths per minute. Roughly five seconds in, five out. No holds, no counting past five, nothing to get wrong. This is the rate most often used in the HRV literature and it clears the slow-breathing threshold easily.

If you want the exhale advantage: make the out-breath longer than the in-breath. Four in, six out. Or 4-7-8 if you like a named protocol.

If you like box breathing: keep using it. It is slow, it is structured, the holds give you something to do with your attention, and adherence is the variable that actually determines whether any of this helps you. A technique you will do beats a marginally better one you will not.

What the holds are probably doing is not physiological so much as attentional. Counting through four phases occupies working memory more completely than counting two, which may be why people find it easier to stop ruminating during box breathing than during simple slow breathing. That is a reasonable hypothesis and not a finding, and we are flagging it as such.

What to expect, honestly

Slow breathing is a state intervention. It changes how you feel over the next few minutes. The review's outcomes were measured during and shortly after breathing sessions, not months later.

That is genuinely useful. It is also different from what is often implied. Five minutes of box breathing will not treat an anxiety disorder, and the studies were not designed to test whether it could. The NCCIH's position on relaxation techniques generally is that they can help with some conditions as part of broader care, not as a replacement for it.

That last row matters if you track HRV on a wearable. Breathing exercises raise HRV while you do them, which is why a guided breathing session shows a spike. Whether that translates into a higher overnight baseline is a separate question, and it is not what the acute studies measured. We covered what those numbers can and cannot support in what is a good HRV for your age.

When not to use it

Slow breathing is low-risk, not no-risk, and the breath-holding component specifically has situations where it is the wrong tool.

Breath-holds and raised blood pressure. Holding the breath transiently raises intrathoracic pressure and blood pressure. For most people that is trivial; with uncontrolled hypertension or certain cardiac conditions it is a reason to drop the holds and breathe slowly without them.

Pregnancy. Extended breath-holding is generally advised against. Slow breathing without holds keeps the benefit and removes the question.

Panic and hyperventilation. This is the counter-intuitive one. During a panic attack, many people are already over-breathing, and being told to take a big deep breath makes it worse. The useful direction is smaller, slower breaths with a long exhale, not deeper ones. Box breathing's instruction to inhale for a full four seconds can work against this if it is interpreted as "inhale as much as possible".

Asthma and COPD. Breathing techniques can help, and which technique depends on the condition. This is worth asking a respiratory physiotherapist about rather than picking from a search result.

While driving or in water. Obvious once stated, routinely ignored. Slow breathing with holds can produce light-headedness.

How it compares to the alternatives people actually try

None of these has been tested against the others in a way that would let you rank them, so this is a comparison of mechanism and practicality rather than of effect size.

The physiological sigh deserves a note because it is the odd one out: two inhales through the nose, the second shorter and stacked on the first, then a long slow exhale. It is not a practice so much as a single manoeuvre, and its appeal is that it takes about five seconds rather than five minutes, which makes it the only one on this list you will realistically do mid-meeting.

Again, the honest framing: all of these clear the sub-10-breaths-per-minute threshold that the review associates with the documented effects. Beyond that, choose by what you will actually repeat.

How to actually do it

Sit or lie down. Breathe through your nose. Let the belly move rather than the shoulders, which is the one piece of technique that consistently matters and is usually mentioned last.

Then pick a pattern and hold it for five minutes, not thirty seconds. The studies in the review ran sessions of meaningful length; the effect is not instantaneous.

If you feel light-headed, you are breathing too deeply rather than too slowly — a common error, since "deep breathing" and "slow breathing" get used interchangeably and only the second is what the evidence is about. Reduce the volume, keep the rate.

And if the holds make you tense, drop them. There is no study telling you not to.

The broader question of what does and does not calm a stressed nervous system is covered in what the evidence actually supports.

This article summarises published research for general information and is not medical advice. Breath-holding is not advisable in pregnancy, with uncontrolled high blood pressure, or with some cardiac and respiratory conditions. Persistent anxiety warrants seeing a doctor.