Most advice about sleeping on a plane is about gear: the right neck pillow, the right headphones, the right seat. Very little of it has been tested on passengers. When I looked for studies that measured how much people sleep in an economy seat, I found none. Every flight study with a sleep tracker in it was of pilots. What does exist is lab work that rebuilds parts of the cabin, and it says some useful things about seat angle, about that drink before you nod off, and about blocking noise and light.
This post is about the hours in the seat. The bigger question of when to sleep so your body clock lands in the right place belongs to our post on why flying east is worse and what helps with jet lag.
The short version
- Recline as far as you can. In a lab test, sleeping sitting upright gave 29% less total sleep and 79% less dream (REM) sleep than lying flat; a seat tilted back to 40° came close to flat on total sleep.
- Skip the drink if you plan to sleep. At air pressure matching a cabin at 8,000 ft (2,438 m), people who drank before bed had median blood oxygen of 85.3% and spent about 201 minutes below 90%, with less deep sleep than on the ground.
- Use earplugs and a mask together. In 14 healthy adults sleeping through recorded hospital noise and light, the pair raised dream sleep from 9.3% to 12.9% of the night.
- Expect dry air and steady noise. Cabin humidity averaged 6% on 26 long-haul flights, and wide-body cabins run at roughly 70 to 75 dB(A) at cruise.
- Do not expect a full night. Even pilots taking scheduled rest breaks on long flights slept only about 3 to 4 hours.
What the cabin does to sleep
Before the tips, it helps to know what you are up against, because most of it you cannot change.
| Cabin factor | What was measured | Who or what was tested | Can you change it? |
|---|---|---|---|
| Seat angle | Upright: 29% less sleep than flat | 6 adults, daytime naps in a lab | Partly: recline, pick your seat |
| Thin air | Oxygen saturation fell 4.4 points at 8,000 ft | 502 adults in a pressure chamber | No |
| Thin air plus alcohol | Median oxygen 85.3%, less deep sleep | 17 healthy adults in a pressure chamber | Yes: don’t drink |
| Humidity | Average 6%, range 1 to 27% | 26 flights on one jet type | Barely |
| Noise | About 70 to 75 dB(A) at cruise | Five wide-body jet types | Yes: earplugs or headphones |
| Light | Mask plus earplugs raised dream sleep | 14 adults in a simulated hospital room | Yes: a mask |
Seat angle: the closer to flat, the better
The one study that measured sleep at different seat angles was not about passengers. Roach and colleagues in Australia wanted to know what rest seats do for long-haul crews, so in 2018 they had six healthy adults sleep in a seat in the lab, wired up to measure their sleep stages, at three back angles: 20° from upright, 40°, and fully flat at 90°. Everyone slept only four hours the night before, then got a four-hour nap in the seat in the afternoon, once at each angle.
The upright seat lost on almost every count. Against the flat seat, it gave 29% less total sleep, 30% less deep sleep and 79% less REM sleep, the stage most linked to dreaming. The reclined seat did far better: similar total sleep and deep sleep to flat, though still 37% less REM.
Six people is a very small study, the naps were in the daytime after a short night, and nobody was on a plane. But the direction is clear, and it fits what anyone who has tried to sleep bolt upright already suspects. A typical economy seat reclines only a little, so the practical lesson is to use all the recline you have, and, if you can choose, a seat where reclining will not start a fight. Neck pillows are meant to fix part of the upright problem by holding the head still, but I found no study testing whether one helps anyone sleep.
The drink before you sleep
This is the finding I would most like every long-haul passenger to see. Planes do not keep the cabin at ground-level air pressure. They pressurize it to match an altitude of up to about 8,000 ft (2,438 m), which means each breath carries less oxygen. In a 2007 study in the New England Journal of Medicine, 502 adults spent 20 hours in a pressure chamber set to different cabin altitudes, and at 8,000 ft their blood oxygen fell by an average 4.4 percentage points. Discomfort rose at 7,000 to 8,000 ft and showed up after 3 to 9 hours. Worth knowing: the lead author worked for Boeing.
Sleep makes the drop worse, and in 2024 a German team added alcohol. In the study by Trammer, Elmenhorst and colleagues in Thorax, healthy adults slept for four hours, from midnight to 4 a.m., either in a normal sleep lab (23 people) or in a chamber matching 8,000 ft (17 people). Each person did one night after drinking alcohol and one without. Their average blood alcohol was 0.043%, about half the 0.08% US driving limit, so a couple of drinks rather than a bender.
| Condition | Median blood oxygen | Minutes below 90% | Median heart rate |
|---|---|---|---|
| Ground, no alcohol | 95.9% | 0 | 63.7 |
| Ground, alcohol | 95.0% | 0 | 77.0 |
| Cabin altitude, no alcohol | 88.1% | 173 | 72.9 |
| Cabin altitude, alcohol | 85.3% | 201 | 87.7 |
With both alcohol and thin air, deep sleep fell to a median 46.5 minutes, against 84 and 67.5 minutes in the ground-level groups. Look at the second column too. Even without a drink, healthy sleepers at cabin pressure spent most of a four-hour sleep below the 90% mark doctors use as a warning line. The drink pushed oxygen lower and heart rate up by about 15 beats a minute on top of that.
The limits are real. These were healthy volunteers, the sleep window was only four hours, and the two settings had different people in them. Older passengers and people with heart or lung disease were not tested at all. If that describes you, whether you should fly and how is a question for your doctor, not for this post.
Noise and light: plugs and a mask
A cabin at cruise is loud. In a 2022 study, researchers recorded the noise on five wide-body jets with a calibrated phone app from an aisle seat, and the averages at cruise sat in the low to mid 70s dB(A), with all five types within 3 dB of each other. The low rumble is louder still on a scale that counts low tones, which is why foam earplugs and noise-cancelling headphones feel so different.
Nobody has tested earplugs and masks for sleep on a real flight, so the best stand-in is a 2010 study that put 14 healthy adults through a recording of intensive care noise and 100 lux of light, about the brightness of a dim living room, while measuring their sleep. Wearing earplugs and an eye mask together raised REM sleep from 9.3% to 12.9% of the night and cut the time it took to reach REM from 147 to 106 minutes. Total sleep rose from 455 to 475 minutes, but that was not a clear result. The authors note that their sample was small and that the design cannot say whether the mask or the plugs did the work.
I wear a sleep mask because my bedroom is not fully dark, so I am biased in favor of them. The honest picture is mixed. In a 2023 study where young adults wore an eye mask at home, the mask helped next-day memory and alertness, but a headband recording in 29 of them found no change in their sleep stages. On a plane, with cabin lights, screens and window shades, a mask has more light to block than in a bedroom. That is my reasoning, not a test result.
Dry air
Cabin air is very dry. Swedish researchers who took measurements on 26 long-haul Boeing 767 flights found an average humidity of 6%, with a range of 1 to 27%, at an average temperature of 72°F (22°C). For comparison, our post on the best humidity for sleep suggests 40 to 50% in a bedroom. I live in Florida and have used a humidifier for nasal comfort at night, and the effect was small, so I would not promise much from a saline spray or a damp cloth. But a dry nose and throat is one more reason you wake up on a plane, and drinking water costs nothing.
A plan for the seat
- Decide whether to sleep at all. On a short stay, staying on home time may be smarter; the jet lag post covers that choice.
- Pick a seat you can recline, away from the galley and toilets if you can.
- Skip alcohol in the hours before you try to sleep.
- Put in earplugs or noise-cancelling headphones, and wear a mask.
- Recline fully and support your head so your neck can relax.
- Count any sleep as a win. In a 2013 study of 70 airline pilots on scheduled in-flight rest breaks, estimated sleep was about 3 to 4 hours per long flight, so a broken few hours in economy is normal.
What this post does not know
It does not know how much passengers sleep in economy versus business class, because no study I found measured it, and it does not know whether neck pillows help. It also leaves out sleeping pills, melatonin and blood clots on long flights. In a survey of 460 Qantas long-haul passengers, 15% used a medicine to sleep on board. Whether that is safe for you, at what dose, and alongside what else you take, is a question for a doctor or pharmacist. The same goes for flying while pregnant, with children, or with a heart or lung condition.
This article is for general information only and isn't medical advice. If you have a sleep disorder or a health concern, talk to a qualified professional. Full disclaimer.