Do sleep masks actually work?
Short answer: a mask is a cheap, portable way to make a room darker, and if light is what wakes you or keeps your sleep shallow, it removes that variable more reliably than curtains do. What we are not going to claim is that wearing one improves your sleep quality. The studies are small, most were run on hospital wards rather than bedrooms, and the best-known bedroom study found no difference in sleep structure or in how well people said they slept. A mask solves a light problem, and whether it works for you is mostly a question of fit.
We sell one, disclosed further down. This article is about what the category never discusses: where masks leak, and why shape beats fabric.
Nobody publishes a test method for "100% blackout"
Every listing carries the phrase — total blackout, 100% light blocking, zero leakage — and none of them defines the test. No light source, no distance, no measurement at the eye, no face shape, no sleeping position, no pass criterion. Curtain fabric has blackout classifications you can look up; sunglasses have transmission categories. "100% blackout" on a sleep mask has nothing behind it: every brand uses it, us included, and none of us has published a method.
So the useful question is not how dark a mask is in the abstract. It is where light gets past it on your face.
Where masks actually leak
Three places, and all three are about your anatomy rather than the product.
The bridge of the nose. The main one. Your nose creates a gap the mask has to bridge, and it varies enormously between faces. A shallow or unadjustable nose section sits proud of the skin and leaves two triangles of light pointing down at your cheeks. The fixes are a moulded nose wing, an adjustable bridge, or enough softness to collapse into the gap.
The cheekbone. Contoured masks seal along bone rather than soft tissue, which is what makes them work, but cup depth has to match your orbit. Too shallow and it presses your lashes; too narrow and it rides on the brow and leaves a gap underneath, which is the leak you notice at dawn.
The strap, when you turn over. A mask that seals on your back is a different object once half your face is in a pillow: the pillow pushes the lower edge up, the strap slides, and the seal opens on the side you are lying on. This is the failure people describe as "it moved in the night", and it depends on strap width and placement on the skull more than on the mask front.
Test it in a bright room in the middle of the day, not at bedtime. Mask on, face a window, look for glow; then lie down, roll onto each side and press your face into the pillow the way you actually sleep. If you can see the shape of the window, that mask does not fit you.
Contoured versus flat is the distinction that matters
A flat mask is fabric resting directly on your eyelids. A contoured mask has moulded cups that leave space in front of the eyes, so you can blink, open your eyes underneath it, and move through the night with nothing touching the lids.
That matters more to most people than the material does. Eyelid pressure is the commonest reason someone tries a mask, dislikes it and concludes masks are not for them, and it is a property of shape rather than of silk versus cotton. It also gets worse the longer you wear it.
What light at night actually does
Closing your eyes does not make it dark. Katsuhisa Ando and Daniel Kripke measured light attenuation by the human eyelid in the 1990s: the eyelid attenuates light rather than blocking it. Mariana Figueiro and Mark Rea followed up in BMC Research Notes in 2012, delivering light through the closed eyelids of sleeping participants, where it suppressed melatonin and shifted melatonin onset. Those were tiny studies, six and seven people, using narrow-band green light. What they establish is that "my eyes are shut, so it doesn't matter" is wrong.
A 2022 PNAS study from the Northwestern group of Ivy Mason and Phyllis Zee had healthy adults sleep one night in dim light and one night in a room lit to around 100 lux. On the lit night heart rate was higher, sympathetic activity was raised, and insulin resistance was higher the next morning. Participants noticed none of it. Note what that is not: one laboratory night in healthy young adults, measuring cardiometabolic markers rather than sleep quality, with no mask involved. Whether removing the last of the light from your bedroom changes anything you can feel is a far weaker claim, and nobody has shown it.
What the mask studies themselves show
Viviana Greco and colleagues at Cardiff published two experiments in Sleep in 2023, with 89 and 33 participants, comparing a week wearing an eye mask against a week without. People learned word pairs slightly better and reacted slightly faster on an attention task, but the effects were small and there was no difference in sleep macrostructure, total sleep or self-reported sleep quality. The same journal then ran a reanalysis by Stephen Rhodes arguing the original models understated the uncertainty; on reanalysis the plausible effect was roughly zero to two extra words out of eighty.
The hospital evidence is more encouraging and lower quality. The Cochrane review by Rui-Feng Hu and colleagues in 2015 pooled trials of earplugs and eye masks in intensive care, found more hours of sleep and less delirium, and rated that evidence low quality. A 2017 trial by Alexandre Demoule and colleagues in Critical Care used polysomnography in 64 patients: no significant difference in deep sleep, though fewer prolonged awakenings. About a third did not keep the devices on all night, and some found the mask claustrophobic. Intensive care is bright, noisy and constantly interrupted, so a benefit there describes that environment rather than your bedroom — and the adherence problem is the finding most likely to apply to you.