Do Moon Phases Affect Sleep? What the Research Actually Says

The belief that the full moon disturbs sleep is old enough to be built into language — "lunatic" comes from luna, the Latin word for moon. For a long time science dismissed it as folklore. More recent research has made the picture genuinely more interesting.

What studies have found

A well-known 2013 study in the journal Current Biology re-analysed sleep-lab data collected for an unrelated purpose, so no one — including the researchers — had the moon in mind at the time. Around the full moon, participants took longer to fall asleep, slept about twenty minutes less on average, and showed reduced deep sleep. Because the study was done in a windowless lab, the effect could not be explained simply by moonlight coming through a window.

The specifics of that 2013 paper, led by Christian Cajochen, are worth having. Thirty-three volunteers had been studied under tightly controlled conditions. Around the full moon, the researchers reported that participants took about five minutes longer to fall asleep, slept roughly twenty minutes less, showed a marked reduction in the deep-sleep brain activity measured by EEG, and had lower levels of melatonin. It was a striking set of results, and it was reported widely.

Then came the replications, and this is where the story gets instructive. In 2014 a group including Michael Cordi and Björn Rasch went looking for the same effect in a much larger dataset — over a thousand participants — and did not find it. They titled the paper, pointedly, with a reference to the file drawer problem: the tendency for null results to go unpublished while positive ones get written up. Their argument was that a field where only the interesting findings surface will appear to show an effect whether or not one exists. Several other reanalyses since have likewise come up empty.

The most interesting recent contribution runs the other way. A 2021 study in Science Advances, led by Leandro Casiraghi, tracked sleep using wrist monitors across three Toba/Qom Indigenous communities in Argentina — one with no access to electricity, one with limited access, one fully urban — and, separately, several hundred university students in Seattle. Across all of those groups, sleep started later and lasted less long in the three to five nights before a full moon. The effect was largest in the community without electric light and smallest in the urban one, but it was detectable even there.

That pattern is what makes the finding hard to dismiss. If the effect were purely cultural expectation, it should not have been strongest in the group with the least exposure to the folklore and the most exposure to actual moonlight. And the timing is a point in its favour: the nights before a full moon are precisely the nights when bright moonlight is available in the early evening, before people would otherwise be asleep. That is a plausible mechanism rather than a mystery.

The honest summary is that there may be a small effect for some people, that it is not dramatic, that it is probably about light rather than anything more exotic, and that it remains genuinely contested.

What the moon does not do

It is worth separating the sleep question from the much broader folklore, because the broader claims have been tested thoroughly and have not held up. The belief that the full moon drives admissions to emergency rooms, psychiatric crises, crime rates and birth rates is persistent and, as far as the evidence goes, wrong. A 1985 meta-analysis by James Rotton and Ivan Kelly pooled dozens of studies and found no reliable relationship, and later work has largely agreed.

Why does the belief survive such a comprehensive debunking? Mostly because of how memory works. If you have a chaotic shift at work and someone remarks that there is a full moon, the coincidence is memorable and gets stored. The equally chaotic shifts under a waning crescent, and the quiet nights under a full moon, are unremarkable and are not. Psychologists call this illusory correlation, and it is strong enough that surveys have repeatedly found large proportions of nurses and police officers who believe in the lunar effect despite working in the very settings where the data say it does not exist.

The tidal argument, which comes up often, does not survive scrutiny either. The moon does raise tides in oceans, and the human body is largely water — but tidal force depends on the difference in gravitational pull across an object, and across something the size of a person that difference is vanishingly small. A wall standing a short distance away exerts a greater gravitational pull on you than the moon does.

Why there might be an effect

• Light: before electricity, the days around the full moon were meaningfully brighter at night, and human sleep may still carry a trace of that ancient sensitivity.
• Expectation: if you believe the full moon disrupts sleep, you may sleep worse simply because you expect to — a real and measurable effect.
• Internal rhythms: some researchers suspect humans may have a faint internal monthly cycle, though this is far from settled.

How to sleep well around the full moon

Whether the effect is astronomical or psychological, the fixes are the same as for any restless night: keep the bedroom genuinely dark, avoid screens before bed, keep a consistent sleep and wake time, and go easy on caffeine and alcohol. If you notice you sleep worse near the full moon, knowing when it falls lets you protect those nights in advance.

If the Casiraghi findings are pointing at the real mechanism, the most targeted intervention is the dullest one: block the light. Proper blackout curtains or a sleep mask address the only pathway anyone has proposed that has a solid physical basis. This is also the cheapest hypothesis to test on yourself — if masking the light removes your full-moon restlessness, you have your answer.

One further suggestion, aimed at the expectation effect rather than the astronomy. If you find that checking the phase makes you anticipate a bad night, stop checking before bed and look in the morning instead. Keep a rough note of how you slept and compare the two afterwards. Over a few months you will have a small dataset on yourself, which is worth considerably more to you than any population-level average — the studies above measure what happens on average, and you are not an average.