The 90-Second Fix for Doomscrolling, from Patañjali

The Scroll

Ninety seconds of one particular breath is the best-evidenced exit from a doomscroll I found, and the instruction is sixteen hundred years old. You know the moment. You picked up the phone to check one thing, and some long stretch later you couldn’t say what you actually read. Jaw tight. Breath gone shallow without you noticing it change. That last part — the breathing — is where this whole thing turns.

By the end you’ll have the whole of it: seven breaths, a specific ratio, about ninety seconds, plus an honest account of how thin the evidence underneath it really is. But the number attached to every other version of this needs dealing with first. They all say five thousand years. That’s wrong, and the way it’s wrong tells you what kind of video you’re usually watching.

Sixteen Hundred Years, Not Five Thousand

The Yoga Sūtras of Patañjali are not five thousand years old. Philipp Maas’s 2013 survey of the dating scholarship puts the text at around four hundred of the common era. Other scholars argue for the centuries just before it, so the honest range is roughly sixteen hundred to twenty-two hundred years. Late antiquity, not prehistory. I’m being fussy because a channel that inflates a date will inflate a finding, and you should hold me to that.

Here’s what the text actually is. Under two hundred short lines, arranged in four chapters, and the second line defines the entire project: yogaś citta-vṛtti-nirodhaḥ. Citta is mind-stuff, the whole apparatus of attention and memory. Vṛtti means turnings — the shapes the mind keeps taking. Nirodha is the stilling of them. So from line two onward, this is a manual about a mind that will not stop generating shapes.

Which is a fair description of a feed. Every swipe hands the mind another shape to take. And that isn’t me stretching the text to fit, because a few lines later Patañjali lists the five kinds of turning: correct knowledge, error, imagination, sleep, and memory. The feed supplies all five, in rotation, for as long as you’ll hold it. The text has a category for what’s happening to you.

The Diagnosis, Then the Remedy

The diagnosis comes before the remedy, and it’s more specific than I expected. Line thirty lists nine obstacles to a steady mind. Line thirty-one then names four things that ride along with a scattered one: suffering, low mood, unsteadiness in the limbs, and disturbed inhalation and exhalation. Śvāsa-praśvāsa — breathing in and out — is listed as a companion. Not a cause, not a cure. The text says the two show up together, and that’s the hinge: if they travel as a pair, you can grab the one you can reach.

Three lines later comes the instruction. Pracchardana-vidhāraṇābhyāṃ vā prāṇasya — the mind may also be steadied by the expulsion and the retention of breath. That’s Edwin Bryant’s 2009 reading, and he’s careful to mark it as one option among several. What commentators argue over is which half does the work: the long expulsion, or the held pause after it. Modern practice has quietly picked the expulsion, and I want to flag that as a choice rather than a translation.

There is a reason the exhale is the half that survived. Your heart rate isn’t flat across a breath — it climbs on the way in and drops on the way out. Lehrer and Gevirtz’s 2014 review of heart-rate-variability biofeedback describes that swing, and describes a breathing rate near six a minute for most adults where the swing is at its largest. Make the exhale longer and you’re leaning on the falling half.

Now the research, strongest first. Zaccaro and colleagues, 2018, reviewed the slow-breathing literature and found consistent shifts in heart-rate variability, with softer and more mixed findings on how people actually felt. Then Balban and colleagues, 2023, in Cell Reports Medicine: a hundred and eight people, five minutes a day for a month. The exhale-heavy practice — a double inhale and a long sigh out — outperformed mindfulness meditation on mood and on resting breathing rate.

So, the protocol. Phone face down on the table, not in your hand. In through the nose for a count of four, out through the mouth for a count of eight. Seven of those. Don’t hold anything at the bottom, don’t force the last of the exhale, and don’t do it while walking. Twelve seconds a breath, seven breaths, eighty-four seconds. It works — and it also doesn’t do the thing you probably want it to do.

Where It Breaks

Here’s where the old answer and the modern one part company. Line thirty-four is not a reset button. It sits inside a list — lines thirty-three through thirty-nine — of seven alternative methods for steadying the mind, and the target of the whole list is nirodha, the stilling of the turnings. Not being calm enough to resume scrolling. I use it as a circuit breaker, and I should be plain that the text would consider that a very small use of it.

And the evidence is thinner than the confidence around it. Balban’s trial wasn’t blinded — you obviously know which breathing you were told to do — the mood outcomes were self-reported, the participants were healthy volunteers, and it ran for a month. Zaccaro’s review says much the same in review language: small studies, heterogeneous, hard to control. Nobody has tested the sutra’s version. So: decent evidence for slow, exhale-weighted breathing, and none at all for this particular count of seven.

Two more limits, then the useful part. If you have a heart or breathing condition, or you’re pregnant, this is a question for a doctor rather than for me — I’m summarising research, not giving medical advice, and breath retention in particular is where I stop and say ask someone qualified. And it does not fix the habit. Ninety seconds interrupts a state. The phone is still there, and your hand still knows the way back.

The Smallest Version

So, the smallest version that fits a working day. One round — seven breaths, four in, eight out — is eighty-four seconds, and you take it at the moment you notice you’ve lost the thread, not on a schedule. Four in and eight out is twelve seconds a breath, which is five a minute, just under the six-a-minute band where Lehrer and Gevirtz put the largest heart-rate swing. No app, no cost, and it fits inside a lift ride.

Run it twice a day at most, for two weeks, and judge it on one marker: after the seventh breath, can you say out loud what you originally picked the phone up to do? That’s the test. Not whether you feel calmer — you’ll flatter yourself on that — but whether the thing you wanted before the scroll started is still retrievable. Mine came back more often than not, which is the only reason I kept it.

What I’d Read Next

That’s the honest shape of it. A line from around the fourth century, one option out of seven, with modern work that supports the mechanism and never quite reaches the practice. I kept it because the text is careful about its own claims and the research points the same direction. That’s a lower bar than proof, and I’d rather say so than dress it up.

The next one in this series takes the other six methods on that list — lines thirty-three, thirty-five, thirty-six, thirty-seven, thirty-eight and thirty-nine — and asks what, if anything, has actually been tested. Some of them have nothing behind them at all, and I’ll say which ones.

The other six methods are the next video in this series — I’d start there.