How to measure meditation (and what no sensor can tell you)

Nobody keeps going to a gym that never lets them see whether they got stronger. So it makes sense that people who meditate want some proof it's doing something.

The trouble is the tools. The obvious ones either measure the wrong thing, or measure the right thing badly. I spent a long time with the research while building Dhyaan, and this guide is what I took away: which signals deserve your attention, which ones will mislead you, and how to look at your own numbers without turning a quiet practice into an exam.

What can a device measure during a sit?

Ordinary devices can pick up four things. Each one is real, and each one has a blind spot.

Stillness is the most useful signal

Your attention and your body move together more than you'd think. Lab studies keep finding that people whose minds wander more also fidget more. Eye movements get smaller as people settle into focused meditation, and experienced meditators move their eyes less than beginners.

So the moment you stop adjusting is a decent outside marker for the moment you arrived. The times you moved and came back tell you something, and so does the longest stretch you stayed put.

It has limits. You can sit like a statue while replaying an argument in your head. And very still sometimes means very sleepy, so drowsiness needs its own check, which I cover in falling asleep during meditation.

Heart rate shows how far your body settled

In a calm seated practice heart rate usually drifts down a few beats a minute, most of it in the first several minutes. Across controlled trials, meditation lowers resting heart rate too. It can also climb during hard concentration, loving-kindness practice or a wave of emotion, so a rise doesn't mean you failed. The full picture is in what happens to your heart rate when you meditate.

Breathing is the quiet long-term signal

In one study, long-term meditators breathed noticeably slower at rest than non-meditators, and the gap widened with the hours they'd spent on retreat. Breathing is hard to read without a chest strap. When you're very still, though, the tiny rise and fall of each breath can show up in head or wrist motion.

HRV is the most misunderstood

Heart-rate variability jumps whenever you breathe slowly, so a high reading during a sit mostly reports on your breathing. A 2021 meta-analysis of 19 randomised trials also found no reliable rise in resting HRV from mindfulness training. I go through the evidence, and how to read the number on your watch, in does meditation increase HRV?

What no sensor can measure

Nothing reads attention directly. Researchers who study the mind moment to moment during meditation still depend on people pressing a button when they notice they've drifted. The most heavily marketed consumer attempt at a meditation score, an EEG headband, went through a randomised trial, and its scores didn't track people's progress.

A single number for how well you meditated does two kinds of damage. It turns practice into performance, and you start sitting rigidly to win. Then, when the number stays flat while you're quietly changing, it tells you to quit.

How to read your own numbers

Words come first. Before you look at anything, say or write how the sit felt. Scattered, settled, absorbed, whatever fits. Your own report is data too, and if you see the numbers first they'll colour it.

Then compare like with like. A silent sit, a paced-breathing session and a round of japa produce very different heart and motion patterns. Set a sit only beside your own past sits of the same kind and length.

Single sits are noisy, so look at four to eight weeks at a time. You want to see if you're arriving sooner, coming back more easily and settling further. Check how much of the sit was actually seen, too. A watch that slipped or AirPods that died halfway give you half a reading, and a good tool should tell you that.

Last, let the number be wrong. If you felt restless and the motion data says you were still, both can be true. Restlessness can live entirely on the inside.

The signs that show up off the cushion

What people most want from meditation happens in ordinary life. Getting over an argument faster. Noticing a tight jaw before it becomes a headache. Falling asleep less tangled in tomorrow. No device sees any of that, and one line a week in a notebook will. I list five signs worth checking in how to know if your meditation is working.

How Dhyaan handles it

We built Dhyaan around these limits. After the bell it asks how the sit felt, and only then reads the sit back to you: when you arrived, how often you came back, your longest still stretch, how your heart settled. AirPods supply head stillness and likely drowsiness. Apple Watch supplies hands and heart. The steadiness reading, Sthir, is only ever compared with your own past sits.

The full method, blind spots included, is on how Dhyaan measures. If you'd like to feel it from the outside first, try the 60-second stillness test.

Three questions I get asked

Can a smartwatch tell if I'm meditating?

It can tell you were still and what your heart did. It can't separate meditation from quiet rest or from dozing, so your own report needs to sit beside the numbers.

Is a meditation headband more accurate?

EEG headbands pick up the brain's electrical activity, which is closer to the mind than heart rate is. Their scores still haven't been shown to track meditation progress in a randomised trial, and more data hasn't produced a more trustworthy verdict.

What's the single best thing to track?

Whether you sat. In the research on doses and habits, turning up on most days predicts change better than any reading taken during a sit. If you only track one thing, track that, and read how long should you meditate for how much is enough.