Using Apple Watch
How detrimental is noise for your sleep? Andrej Karpathy raised this exact question in a tweet. The answer to Karpathy’s question is, well, empirical, with a rough experiment suggested by Paul Graham:
Consumer devices like Apple Watch measure many sleep metrics, including sleep stages, heart rate variability, and heart rate. The Apple Watch’s microphone also measures noise in fine-grained frequency, which can be filtered down to times when you’re asleep. Given that we’re a medical practice that makes extensive use of wearable data to improve patient care, and advance the science of heart health (where sleep and sleep apnea are major inputs), so we were curious to quantify the actual effect of noise on sleep.
TL;DR: across every metric we examined, bedroom noise shows a consistent, dose-dependent impact on sleep quality.
Our panel shows a steady, almost linear erosion of REM minutes from the low-40 dB range through the mid-50s. Once noise climbs past roughly 58–60 dB (about the loudness of normal conversation), the line kinks sharply downward—REM time falls by an additional ~15 minutes in that single step. Deep-sleep minutes follow a similar pattern, holding fairly steady below 50 dB, slipping modestly through the low-50s, then dropping 6–7 minutes once the room crosses that same upper-50s threshold. Taken together, the curves suggest a threshold effect near 60 dB where restorative stages start to collapse rather than decline gradually.
Total sleep duration mirrors the stage-specific trends: nights under 55 dB hover around 6¼–6½ hours, but once noise breaches the upper-50s the average night shortens by nearly an hour—nearly all of it coming from REM and deep stages rather than lighter sleep.
Physiology responds in the opposite direction. Sleeping heart rate remains flat in the quietest bins, rises a beat or two in the low-50s, then jumps 4–5 bpm in the high-50s/low-60s. Sleeping HRV shows the mirror image: minor dips below 50 dB, followed by a pronounced 15–20 % drop once noise crosses that same ~60 dB mark. These shifts imply the body’s overnight stress load spikes at the same acoustic threshold that disrupts sleep architecture.
Last, our own composite score—a weighted mix of duration, stages, and physiology—tracks the individual metrics closely. Scores stay in the high-70s below 55 dB but slide into the mid-60s once noise exceeds ~60 dB, confirming that the upper-50s/low-60s zone is the pivotal point where overall sleep quality degrades quickly.
The red dashed lines mark the single largest step-down for each series, and most of them cluster in a narrow band around 55–60 dB. Below that range, incremental quieting still buys modest gains, but above it the penalties accelerate: REM and deep minutes shrink by roughly a quarter, total sleep contract by about an hour, HR climbs by several beats per minute, and HRV flattens out.
Practically, that suggests a threshold effect: keeping bedroom sound levels beneath the low-60s dB (roughly the volume of normal conversation) is a pivotal target for preserving restorative sleep stages and the physiologic calm that goes with them.
FYI, there's been tons of research of the effects of noise on sleep from different sources. There have been studies ranging from in-lab experiments, to in-home experiments with artificial and natural sources. If you're interested, some resources:
Institute of Noise Control Engineering Digital library: https://www.inceusa.org/publications/ince-digital-library/ (papers older than 10 years old are available free)
Federal Interagency Committee on Aviation Noise: https://fican1.wordpress.com/findings/ (focuses on aviation noise)
Acoustical Society of America Lay Language Papers: https://acoustics.org/lay-language-papers/ (search for "sleep" -- the ASA has a full library of more detailed research but the documents cost money unless you're an ASA member)
World Health Organization guidelines on noise - https://iris.who.int/bitstream/handle/10665/343936/WHO-EURO-... (doesn't get into specifics on research on sleep, but does refer recommended limits to sleep disturbance)
NIH has done a bunch of research on sleep disturbance from noise, you would need to search through their library
edit (one more): TRB/National Academies https://nap.nationalacademies.org/search/?rpp=20&ft=1&term=n...
A lot of the stuff that posters are asking for have in fact been done, it just takes some digging through the research sites to find them. There's a lot of variation in the data, the hypothesis is that sleep sensitivity varies a lot based on various physical factors (age being a big one).
>age-related deafness
Interestingly, there's been some suggestion that hearing loss is not inevitable with age, but is mostly just the accumulation of noise-related hearing loss in a loud industrial society.https://canadianaudiologist.ca/a-new-perspective-on-chronic-...
https://www.icben.org/2017/ICBEN%202017%20Papers/SubjectArea...
https://www.nature.com/articles/s41370-024-00660-3
I think partly the issue is that how we measure noise doesn't match how noise causes injury. Your cochlea acts as a spiral resonant tube, essentially a "physical FFT," concentrating energy at a particular frequency onto a particular location in the spiral. Too much (local!) energy damages the hair cells, causing conductive hearing loss.
But because we calculate A-weighted decibels by summing all frequencies and then checking if we're above the injury threshold (vs checking whether we exceed the injury threshold at any frequency), using A-weighted decibels can't accurately determine damaging noise levels. If all the energy is concentrated at Middle A it will cause more damage than spreading the energy out across the spectrum, even if the A-weighted decibels come out equal.
It's a somewhat subtle, wrong order-of-operations problem. There's also a separate problem that A-weighting is designed to normalize for perception at various frequencies, not hearing damage.
I've tried searching the literature to find out whether this is either 1)wrong, or 2)generally known within the fields of audiology and occupational hygiene, but so far I've come up empty.
--
I recall an HN poster long ago saying how they wore earplugs daily to achieve "super hearing." It occurs to me that all they were doing was actually protecting their ears from damage. :-|
I've tried searching the literature to find out whether this is either 1)wrong, or 2)generally known within the fields of audiology and occupational hygiene, but so far I've come up empty.
FWIW, I've also heard the same, but don't remember where off the top of my head. It's at least potentially true, but the conventional wisdom among acousticians/noise control engineering is that age-related hearing loss is mostly to increasing age rather than external factors.
Oh, I meant about the wrong order-of-operations problem with decibels, which I have never seen anyone talk about. If you've heard of it please let me know.
The links discuss the evidence that hearing loss isn't inevitable with age, including examples of pre-industrial societies with quiet environments that when tested showed no hearing degradation with age.
The controversy seems to be mostly about how much of that effect was caused by good diet vs lack of exposure to loud sounds. I tend to think both are needed to be fully protective, eg to take an extreme example alcohol is known to cause damage to hearing cells even without exposure to loud sounds.
I expect, with apologies to Tolstoy, "All dysfunctional hearing is different, whereas all healthy hearing is the same."
I'd expect the relationship goes the other way around. Most old people I know sleep lightly comparatively
I’ve spent years “fixing” my sleep. Things I recommend:
* a sleep tracker. I love AutoSleep - https://apps.apple.com/us/app/autosleep-track-sleep-on-watch...
* Any kind of white noise. I use an air purifier at home, and a little pocket size white noise machine when I travel.
* find and eliminate any noises in your home. Computers, fridges, squeaky doors, etc.
* find and eliminate any lights. Especially stupid power and status LEDs. Bedroom should be pitch black at night. Electrical tape works well for this.
* Blackout curtains
* cheap eye mask
* Magnesium Glycate supplement
* Earplugs- https://www.amazon.com/gp/aw/d/B0015TBGR6
* And finally, a recent addition that I’ve fallen in love with is sleep ear buds: https://www.amazon.com/dp/B0DT9GBFQ2
Although, I would wait on getting sleep ear buds, Anker just announced a new model with ANC and a similar price.
Any advice on how to eliminate noisy neighbours that would randomly drop their shit at random intervals in the middle of night and let their child to stomp around every late evening?
Unfortunately, the best way to cure this is "don't live underneath other people".
I second this. If you really have to live in a flat, top floor is always the superior option.
I used to live in a wooden construction apartment, with neighbors above me. They themselves stomped a lot, both late at night and early in the morning, somehow.
Then they had a kid, which grew up and of course would run around the house at any hour. I was getting terrible sleep and it was driving me crazy.
All I could really do was move, and I made sure the next building was reinforced concrete, and that I would be on the top floor. The unit also ended up being on one end so only one wall shared with a neighbor.
It was instant relief and absolutely worth the move. It sucks though because obviously not everyone can live on the top floor of a building. If you're a heavy sleeper then I guess this is less of a problem, so I hope we have enough heavy sleepers in society so there's less competition for top floors. Probably wishful thinking though because the top floor is desired for other reasons unrelated to sleep.
Another thing to avoid are hardwood floors. I live in a concrete high rise but that wasn't enough to deal with a neighbor upstairs who would stomp around in shoes on the — admittedly very nice looking — white oak floor. It's basically a giant drum. Even vinyl is quieter.
That would be my no 1 go-to, if I didn't have to remain alert for my 3 year old sleeping next door (he can seep through the noisy neighbors though!)... (btw our neighbours are underneath and they cook at night, and talk very loudly)
Unfortunately, ear plugs would not help against low frequency bass noise, which this noise typically is.
Ear plugs should reduce intensity of sounds across the board. Yes, they’re less effective against low frequency noise, but reducing the intensity any amount goes a long way – the logarithmic scale is real!
I also personally find high frequency noise to be more disruptive: a car speeding by making wind noise is a lot worse than a large truck rumbling by slowly. Lower frequency is lower energy, after all
You know, I work in hospitals all day and people are always stomping around and moving heavy equipment. But you never hear anything coming through the ceiling.
Contact your state representative and ask them to do something about the building codes. It's a completely voluntary problem. American buildings are noisy because we decided that was okay.
You assume I'm in US, while I'm not. Not even EU. I know building owner personally and the only factual way is to evict those people. The only problem is that building owner is leasing them an apartment, and you can't evict people basing on household noises, even if they are driving me insane.
>Contact your state representative and ask them to do something about the building codes. It's a completely voluntary problem. American buildings are noisy because we decided that was okay.
Congratulations on finding an equally terrible and bad for everybody apartment-living analog to the sort of suburbanite Karen opinion that underpins meddling HOAs and busybody municipal codes.
If you want a quieter apartment pony up to live in a nicer one. Don't force the rest of society to shoulder the expense.
You should probably read The Market for Lemons. It's incredibly difficult to predict from the buyer's perspective how noisy an apartment will be. This is [one reason] why we have regulations.
> Any advice on how to eliminate noisy neighbours
... Not from a Jedi. :p
No need for fancy light sabers, put them to the sword.
I've also been doing small improvements to my sleep over the years and share some of your recommendations! What's worked best for me have been:
1. Good mattress and pillow (this is a 101 kind of thing).
2. Having the right temperature (same as above).
3. White noise. I also use a white noise machine and the White Noise app on iOS (which I bought for like $1 years ago and still use it). Ironically, I don't use white noise but brown noise, which is slightly more soothing to me.
4. Air purifier, not necessarily for the white noise it produces because it's pretty silent, but because that way I guarantee that I get at least 8 hours of purified air a day.
5. Black out curtains.
6. Eye masks. I actually wear two at once -- the first one to block any lights, the second one to press against my eyelashes, which I find it better for sleeping.
7. Magnesium Glycinate. I've tried different supplements, different forms of magnesium, and even different brands, and the one that I currently use has worked wonders for me. Made me go from being a light sleeper to a medium-heavy sleeper.
Haven't tried earplugs, but I don't really feel the need to. Another thing I tried (but never noticed a real difference) was using a small piece of tape on my lips to avoid opening my mouth.
> Ironically, I don't use white noise but brown noise, which is slightly more soothing to me
In case you or others didn't know, iOS has a built in noise generator with different presets under the accessibility options. Sometimes I use it with AirPods and active noise cancellation and it pretty much guarantees I can get a solid block of sleep, at least as long as the batteries last in one go.
The feature is called "Background Sounds"
What is the purpose of hearing the white noise? Is it to shield from other sounds? Or to distract from thinking? Are you focussing on it or not?
The brain can easily tune it out and it masks another intermittent sounds.
I may suggest custom ear buds. Best $200 you'll spend. The original method is a wax form, but now there is a 3D scanning tool to map the shape of your ear.
2 euro silicone ear plugs from Aliexpress easily beat the custom ear plugs I had.
I have the same experience. Custom ones literally made my ears ache. Probably the harder material. Note they were made out of 3D scan of my ears..
Currently using Hansaplast Lärmstopp.
... or just buy silicone ones from the pharmacy (the ones for swimmers).
I keep some of these on me at all times in case I end up next to any loud speakers.
Same thing - I've been adjusting my sleep since first Sleep Cycle came for iPhone. I started with a "smart" alarm and learned alarms are harmful. I didn't need an alarm, I just needed 7-8h of sleep. I was going to bed 1-2AM.
I'm doing all of the above with the following differences:
- I use AirPods pro. They are not that comfortable, but leaving a Audible low in background helps falling asleep and also getting back to sleep rather than picking random problems to solve and eventually waking up. It was "life-saver" when I started my first company before learning to manage stress.
- I used these 3M sleeping ear buds and they were the most confortable https://www.amazon.com/3M-Disposable-Earplugs-Quiet-Sleep/dp...
- Extra pillow over head, helps a lot (I have no idea why)
What does AutoSleep offer above and beyond Apple’s Health app?
I have (and recommend) a white noise machine at home, but when I travel, I use the awesome MyNoise app, which lets you EQ the noise precisely so you can target the specific annoyance in your environment.
> What does AutoSleep offer above and beyond Apple’s Health app?
In my experience. After initial setup, it is way more accurate.Apple commonly says I get more hours of sleep than I know I do. Apple counts just being relaxed as sleeping. Whereas I can tune AutoSleep to correctly identify the difference between me sleep and just being super relaxed. Thus I get a much more accurate image of what is working
Also the apple sleep graph is actually terrible. Almost impossible to understand when wake up are occurring.
I will definitely check it out then as I am frustrated by the same problem where Apple shows me as asleep when I am meditating myself back to sleep. It even shows me as sleeping when I have pushed the crown to see the time!
Would like to second ear plugs, it took some testing to find the right softness/size because I have very small ear canals, but these slim-fit earplugs have been life-changing for my sleep: https://a.co/d/38Kftbm
The earplugs don't completely cut out environmental noise, but they dampen it enough that I don't find myself being awakened by any bumps in the night.
And snipping down the tail ends of the earplugs with scissors helps reduce pressure if you're a side sleeper.
I have a street in front of my bedroom and I'm wondering how well ANC earbuds work in that case. Will it be completely silent or will there still be some kind of whoosh-sound?
A large quantity of disposable earplugs is so cheap that I'd recommend you try that before investing time and money into ANC earbuds.
My spouse snores loudly, and a $15 container of 50 pairs has lasted more than 2 years, and saved our ability to co-sleep (I was only a couple nights away from moving to our guest room). We're also on a street with a very permissive speed limit, and we're under the approach path of the nearby airport, and it's never a problem.
https://www.amazon.com/dp/B0051U7W32 I got these, $15 for 50 pairs; the GP comment is $40 for 200 pairs which is a better value but more investment if you wind up dissatisfied
It won’t be completely silent, if only because ANC always comes with low-level white noise.
I second the disposable earplugs recommendations.
Getting blackout curtains and killing off any stray light from clocks was huge for me.
Sleep eye mask
Appreciate the list. How do you eliminate fridge noise (besides replacing the fridge entirely)?
I can't stand earplugs, and think of ear buds as counter-productive. I've used (silken) eye masks, but have no use for them anymore, because successfully blacked out everything.
Don't know what to make of white-noise, do purring cats count? Or really relaxed snuggling with sexy gals after having fucked each others brains out? Or both?
I don't know how to state this well but I find the fact that nearly all new apartment buildings in California are only allowed to be built on busy streets very frustrating. I'm guessing it's part nimby-ism and part zoning but every time I see a new apartment complex it's next to the busiest streets in the city (LA, SF). It's like many things, if you want health you must be rich. If you can't afford a house then F.U. You get a loud apartment with constant traffic noise.
I *almost* feel like it should be illegal to build them next to the freeway like
Those ones just north of SFO next to the 110: https://maps.app.goo.gl/sAkUMhmnutZ1jFd27
The ones in downtown LA next to the 110: https://maps.app.goo.gl/4mBVLo12hLR4EwU16
The ones north of LAX at Howard Hughes: https://maps.app.goo.gl/Z8bnB7wR5jDBddhg6
I don't actually think I think they should be illegal but dang it, I wish there were more options for new apartments. In LA there are lots of older apartment buildings in quiet residential neighborhoods built 50+ years ago.
https://maps.app.goo.gl/KMLav1zLiQZK6T8L6
https://maps.app.goo.gl/aUH52rBAbaHExjzB6
They are no longer allowed to be built AFAICT.
> only allowed to be built on busy streets
Is this true or a misleading observation?
I can see some argument about it, increasing traffic in areas that aren't equipped for it if the building is large enough. But it might just be market forces, too: in larger cities, it's often more desirable to be near a major artery. Earplugs (if the noise bothers you) vs a longer commute.
It is quite literally illegal to build anything but detached single family homes on 95.8 percent of the total residential land area in California / 75% of the residential land area in most major American cities.
"Residential land area"? Much of California's land has no zoning at all. Of course, it's not near any cities either (or near any road noise).
It is illegal to build housing without a permit, and you will not be issued a permit on much of California’s land.
Residential land area = land you can build housing on
https://belonging.berkeley.edu/single-family-zoning-californ...
Daily reminder that our zoning policy is deeply racist and evil
I don’t know about “only allowed”, but this is a thing:
https://sd11.senate.ca.gov/news/senate-passes-senator-wiener...
Check out the location of these condos in downtown Toronto built alongside the Gardiner Expressway, a very busy 90km/h (~55mph) highway: https://maps.app.goo.gl/TGLCBJGbrSEkAm7W6
The satellite view shows many units in multiple buildings directly next to this highway with no noise protection: https://maps.app.goo.gl/JA4jMPSSYsDBq5my5
Years ago I lived near a busy four-lane road with what seemed like a reasonable 60km/h speed limit. However, when it rained, the noise from the car tires on the wet road made it difficult to hear music from my radio. It was very unpleasant to leave the balcony doors open, as the constant din starts to work its way into your brain and make it very difficult to focus.
https://cayimby.org/legislation/sb-79/
They will be allowed by right near any type of transit stop if SB79 goes through!
Streets are much quieter at lower speed limits. There is definitely room for improvement here from the traffic engineers.
To a point. The problem is that the kind of people who are always screeching about traffic speeds are never satisfied and below a certain threshold the only way to go get slower is 4-way stops. And once those go in you get to listen to every motorcycle and delivery truck accelerate after the stop.
Say what you will about the suburbs but the grass and landscaping on each and every property is like noise cancelling foam being applied compared to the hardscaping downtown.
This is why I moved to suburbs tbh
Compared to the city it's so quiet and peaceful. My sleep is much better
I was looking for an apartment on the SF Peninsula a couple years back. Nearly all apartment buildings constructed in recent years are right next to the freeway or right next to the CalTrain train tracks. Lots of air and noise pollution, and it feels like borderline hostility towards renters. Lived there for a short while and ultimately moved out of the Bay Area because of the poor options for renters in that area (although it gets better the further south you go) and of course the sky-high housing prices.
That's not a problem with the location. That's a problem with building materials. I've lives on the busiest street in town, with well insulated walls and windows that block out street noise. And I've lived in poorly -constructed suburban houses that admit noise from across the neighborhood.
I disagree. The location is definitely a factor. If you live within 300 feet (about 100 meters) of a multi-lane highway, then depending on the wind patterns, you are definitely getting more exposure to pollutants that come from cars, trucks, or their brakes and tires. But no one in California is ready to talk about addressing this because it would require radically changing the status quo.
So you’re saying people living in apartments are required to never open their windows?
I’m all for better construction but I still want to be able to open my windows without having to let in tons of noise and car exhaust