By F. Perry Wilson, MD, MSCE

Usually in this space, I try to pull together data to help find the signal in the noise of the medical literature. Today I’m flipping that right on its head. This time, we’re focusing squarely on noise itself. Pink noise, in fact. A new study suggests that the rather calming, soft static-y sound of pink noise may actually be hurting your sleep.

During med school, I was living up in Washington Heights in New York City and, though my exhaustion from the busy clinical service meant I fell asleep almost as soon as my head hit the pillow, it was difficult to sleep through the night — or the day, as was often the case.

The noise of the city — the sirens, the shouting, the airplanes and helicopters — filtered up through my window and disturbed my slumber. Like many people, my solution to this problem was to purchase a $15 noise machine and stick it by the bed. Back then it was a “white noise” machine, and, subjectively, I felt that it worked. It drowned out the intermittent noises outside in favor of a more continuous, innocuous drone.

That was 20 years ago, and science has progressed. White noise is passé at this point. No, the new hotness is “pink noise” and you don’t need a noise machine anymore at all; no, there are countless apps in the app stores that promise to deliver a continuous pink noise to help you fall asleep.

Before we figure out whether pink noise actually helps sleep, let’s define our terms quickly. White noise is probably the easiest. White noise is sound with equal intensity at all frequencies. It’s like the static on an old TV.

Pink noise, on the other hand, has an equal intensity across each octave of frequency, which, practically, means that the higher frequencies are quieter.

But a lot of that research compared pink noise to no noise at all, and that doesn’t feel like the real world to me. This is why I was delighted to see this paper, “Efficacy of Pink Noise and Earplugs for Mitigating the Effects of Intermittent Environmental Noise Exposure on Sleep,” from Mathias Basner and colleagues at UPenn, who rigorously evaluated how pink noise affects sleep, especially when it is used to cover up other sounds.

Here’s how the study worked.

Twenty-five intrepid individuals, all healthy and between the ages of 21 and 41, volunteered to spend a week sleeping in UPenn’s sleep lab.

Each night provided a different “soundscape,” so to speak, in random order. Here were the experimental conditions:

Then there was “environmental noise.” Every 4-6 minutes (the interval was varied randomly so individuals couldn’t predict when the noise would happen), the participant would hear a car, or a train, or a jet, or a helicopter, a fire alarm, a baby crying, a low sonic boom, or, because this is 2026, a drone. Yeah, not a pleasant night’s sleep.

One night they got just pink noise. Just a constant sound.

Then there were the more interesting nights: one where the environmental noise happened but pink noise was playing all along (this is the Perry-in-Med School condition), and one with the environmental noise, but the participants wore earplugs.

That was the setup. During these sessions, a ton of data was collected: surveys about mood; subjective experiences of sleep quality; tests of cognitive and cardiovascular function; and, critically, polysomnography — those electrodes on the scalp that can tell us exactly what stage of sleep someone is in.

I get nervous when I see a study with a small sample size measuring a zillion parameters, but to the researchers’ credit, they appropriately adjusted their statistical significance parameters to account for multiple comparisons. Gold star — I’d expect nothing less from a Penn study.

Let’s start with some obvious comparisons. Compared to a night without any noise, how did all those sirens and jets and drones affect sleep? Adversely.

In particular, exposure to that environmental noise reduced deep sleep by about 25 minutes, in favor of lighter, stage 2 sleep. This is probably not good, as deep sleep is more regenerative.

Overall, though, even with all that noise, the total sleep time was pretty similar: just 5 minutes less.

What about pink noise? Compared to the control situation (no noise at all), it doesn’t look good. Deep sleep was unaffected, but REM sleep (dreaming sleep) was reduced by 23 minutes, also in favor of stage 2 sleep. REM is another critical sleep stage, important for memory and emotional regulation.

From these data, it seems clear that sleeping in a quiet room is just better. But not all of us have that luxury. Can pink noise at least attenuate the effects of the environmental noise?

Nope — you get the worst of both worlds. In this study, exposure to environmental noise (the jets and cars and stuff) plus pink noise led to significantly less deep sleep and significantly less REM sleep. In this situation, both lighter stage 2 sleep and wake time were increased; people slept on average about 15 fewer minutes.

The noise machine is not helping. In fact, it may be hurting.

But as it turns out, there is a relatively simple solution: earplugs. On nights when the participants were exposed to those environmental noises but were also wearing earplugs, they slept just as well as on the no-noise nights. Same deep sleep, same REM.

The bottom line is that those noise machines may not be helping you — not on quiet nights because they reduce REM sleep, and not on loud nights because they don’t do a good enough job drowning out the environmental noises. If you can’t find a quiet place to sleep, the second best thing is to make a quiet place to sleep, and earplugs are, apparently, pretty good for that.

So, in the end, whether the noise is white, blue, brown, or pink, silence, it turns out, is golden.