Getting a good night’s sleep is paramount to health; and yet, whenever we reach for a sleep aid, we are told there are risks involved. The active ingredient in Benadryl, diphenhydramine, is commonly used to induce sleep, but there are concerns (based on really imperfect evidence) that it could increase the risk of developing dementia in the long term. Then there are earplugs, but ear-nose-and-throat specialists are usually not fond of them because they can push earwax down the ear canal and because putting them in position with grubby hands could cause an ear infection.
Now, white noise machines and apps are standing trial for crimes against sleep. They were supposed to improve sleep, not worsen it. Yet, a new and thorough study—though limited in its duration—shows that, if you’re trying to drown out the din of the city by playing white noise at night, you may aggravate the problem.
Noise is noise, and loud noise is bad
Noise is big business. On Apple’s app store for iPhone, top results for “white noise” have tens of thousands of ratings each, while  when searching “white noise” has an astounding 349 million views. It is subtitled “Soothe crying infant,” which gives you an idea how many sleepless parents are turning to noise for a good night’s sleep.
I keep referring to “white noise” but the truth of the matter is that noise comes in different colours, so to speak. White noise has the same loudness across all frequencies, meaning that the bass region (think of the sound a bass drum makes) is as loud as the treble region (think of the sound that a cymbal makes). It’s the sound of a predigital television set tuned to a dead channel, to quote William Gibson. The noise is even and aggressive.
Pink noise is different. Whereas white noise is uniform, pink noise is not. It is most intense in the bass region, and as the frequency goes up to higher pitches, its intensity goes down. It’s the sound of a waterfall: booming and more natural than the artificiality of white noise. (Wikipedia has audio samples of both Ěý˛ą˛Ô»ĺĚý.)
Playing these sorts of noises at night is meant to help people fall asleep and avoid being woken up by sudden environmental noises, like car horns, loud trucks, and inebriated folks making their way home. How these continuous noises are supposed to accomplish that has been the subject of speculation. It may be that there is something inherently lulling, almost hypnotic, about this kind of constant noise, like how some believe that the sound of rain promotes sleep.
It may also be that the noise masks sudden sounds. Think about it. If you’re asleep and your bedroom is very quiet, your brain will perceive the sound of a motorcycle blasting through outside your house as being very loud because the difference in loudness between your bedroom and the bike is large. But if your bedroom is already noisy thanks to an app on your phone playing pink noise, that motorcycle’s sound will not be łŮłó˛ąłŮĚýmuch louder. It’s the difference in loudness that wakes you up: shrink that difference and you may stay unconscious.
The third hypothesis is that playing noise diligently before going to bed trains your brain to associate the sound with sleep. In this line of thinking, there is nothing inherently sleep-promoting about the noise; rather, you develop a Pavlovian response when hearing the noise and it’s a cue for your brain to relax.
Compared to pharmaceuticals, using continuous noise looks harmless, but as one of our student contributors pointed out in an article we published in 2024, noise is still noise, and if the noise is too loud, there are consequences. When it comes to noise exposure on the job, the Canadian Centre for Occupational Health and Safety recommends  for an eight-hour shift in any 24-hour period. (The American equivalent, OSHA, sets their threshold at .) If you are three feet away from someone (just shy of a metre—basically, very close) and you need to raise your voice to be heard, your environment is probably louder than 85-87 decibels. Constant exposure to loud noise is a stressor and it can .
Why do we bring this up? Because white noise can be louder than this threshold. The white noise sources tested by researchers for their claimed ability to help with sleep range , comparable to a motorcycle engine. For parents who use continuous noise to help their infant sleep, there are  that this could negatively impact their speech development because of how loud the sound is.
But let’s say you keep the volume low enough. Will continuous noise help you sleep? For the longest time, it looked like the answer was “yes,” but that may have been because the studies themselves were so bad, they should have been kicked out of bed.
Shallow sleep
Inadequate studies get published regularly because researchers don’t have enough time or money to conduct larger, more rigorous studies and because an academic scientist’s career is unfairly judged by the number of publications they put out. Hence, a lot of so-so, preliminary pilot studies making their way into the literature, and research into continuous noise does not escape from this irritation.
´ˇĚý of 38 studies of continuous noise used as a sleep aid and published in 2021 concluded that, even though it looked like continuous noise was helpful, the studies’ quality was very low and they were so different from one another that they could not easily be compared. Some studies had found a deleterious impact on sleep of continuous noise (see  from 1993, for example, that highlighted the problem in middle-aged volunteers), but the overall picture was rosy—if it could be believed. Remarkably, some of the authors who penned this systematic review decided to do something about the sad state of the evidence and they conducted  of pink noise’s impact on sleep, comparing it to earplugs.
They brought in 25 healthy young adults to spend seven nights in a sleep laboratory. The protocol was pretty strict: acoustically isolated bedrooms, constant temperature, a noise-free first night to adapt to the new setting, and scheduled times for the lights to turn off and on again. Participants’ sleep was monitored via polysomnography, which measures brain waves, blood oxygen levels, breathing, heart rate, and limb movements. And they had surveys to complete every evening and every morning.
The twist here is that every night was different, and the order in which these conditions were triggered was changed from one group of participants to the next. One night was silent. Another featured a mix of disruptive sounds (helicopters, cars, crying baby, fire alarm) played over high-accuracy speakers. Yet another featured the same mix but participants had to wear cheap, soft foam earplugs during the night.
Then we get into the continuous noise. One night had no disruptive sounds, but instead a 50-decibel pink noise was piped in from ceiling speakers. And finally, participants spent two nights hearing the disruptive ˛ő´ÇłÜ˛Ô»ĺ˛őĚý˛ą˛Ô»ĺĚýthe pink noise, the latter at either 50 decibels or 40 decibels (which sounds half as loud as 50 dB).
The theory was that both the ear plugs and the pink noise would work against the car and helicopter sounds and improve the quality of the participants’ sleep. But that is not what happened.
The earplugs »ĺľ±»ĺĚýwork quite well. When disruptive noises were played on their own, participants lost about 23 minutes of deep sleep, which is important for brain recovery and the consolidation of memories; but when they wore earplugs at the same time, they recovered 17 of those minutes. Not a perfect shield against the noise, but a really good one. Plus, two thirds of the participants rated the plugs as either very comfortable or somewhat comfortable.
The pink noise, however, made thing worse overall. When used to mask environmental noises, it made sleep worse than if environmental noises were played on their own. And if you’re playing pink noise not to mask the hubbub of the city but simply to get lulled into the arms of Morpheus, do know that, in this study, pink noise on its own stole from participants over 18 minutes of REM sleep on average compared to their night without the pink noise. Those 18 minutes were instead spent in a much more superficial—and less restorative—part of sleep.
This study, while rigorous, is far from perfect. It recruited a small number of people who were all young and healthy, and they each got to experience a particular scenario once. It’s possible that the brain adapts to pink noise and that sleep improves in the long term. This study did not check for this important possibility.
But at this point, we shouldn’t take the hype around continuous noise devices and apps at face value. It sure ˛ő´ÇłÜ˛Ô»ĺ˛őĚýlike white or pink noise would benefit sleep by harmlessly masking the sounds around us; but until you test a hypothesis, all you have is your gut assessment.
And now we find ourselves back at the start, with  having trouble either going to sleep or staying asleep and being told that no sleep aid comes consequence-free. Pills create dependency and will worsen things long term (and some may even increase the risk for dementia—more studies needed); constant ear plug use may lead you to an ENT’s office to dislodge impacted cerumen, and young parents can’t risk not hearing their baby cry at night; and now white noise apps may be too loud if we crank our phones up and they might reduce the quality of our sleep anyway.
It’s easy to say don’t bring screens into the bedroom. Log off an hour before going to bed. Keep lights low. Most people with insomnia, I suspect, have done all of this already. A well-done study is how we saw clear cracks in the white noise hypothesis. Subsequent studies might show us a better solution.
Take-home message:
- If you use a continuous noise generator (e.g. white noise or pink noise), ensure that the volume is not too loud, as some devices as well as smartphone apps are capable of delivering a very loud sound that is problematic in the long term.
- A new study tested pink noise on its own and as a way to mask disruptive environmental sounds at night and found that the pink noise made sleep worse and more superficialbr.
- Better and longer studies are needed to see if these results can be replicated or if the brain habituates to pink noise and sleep improves after a while.
