For about a year I filed all of this under one heading. Fan noise, rain recordings, the hissing static people upload in ten hour blocks: same category, pick whichever.
Then I ran a boring experiment: same task, a different sound each day for a week. White noise had me quietly irritated by the forty minute mark and I could not have told you why. Brown noise I repeatedly forgot was playing, which is the highest compliment you can pay a background sound.
The colors are not a branding exercise. They describe something measurable, and once you know what, choosing stops being guesswork.
Noise colors describe how a sound's power is distributed across frequencies. White noise carries equal power at every frequency, which makes it bright and hissy. Pink noise rolls off as frequency climbs, at the rate that gives every octave the same total power, which is why it sounds fuller and more balanced to human ears. Brown noise rolls off more steeply again, loading most of the energy into the bass, which is why it sounds like a waterfall or a motorway two streets away.
That is the entire taxonomy. Everything after this is preference, task, and volume.
White noise is flat: equal power per hertz, all the way up. You would reasonably expect flat to sound neutral. It does not. It sounds like a hiss, weighted heavily toward the top end.
The reason is that human hearing works in octaves, not hertz. Each octave is a doubling of frequency. The octave from 100 Hz to 200 Hz is 100 Hz wide. The octave from 5,000 Hz to 10,000 Hz is 5,000 Hz wide, fifty times more territory. Since white noise puts the same power into every hertz, the high octaves collect far more total energy than the low ones, simply because they are wider. Flat on paper, treble heavy in your ear.
Pink noise corrects for exactly this. Its power falls off in inverse proportion to frequency, about a 3 decibel drop per octave, and the result is equal power in every octave. It is the version that actually sounds balanced, and the standard signal for tuning a room in audio engineering.
Brown noise falls off twice as fast, around 6 decibels per octave. It is named after Robert Brown, the botanist behind Brownian motion, not after the color, because the signal is mathematically related to a random walk. The practical effect is that the treble is mostly gone, leaving a deep, dense rumble with no sharp edges in it.
| | Character | Best for | Who tends to dislike it | |---|---|---|---| | White | Bright, hissy, present. Untuned radio | Masking sharp intermittent sounds: doors, keyboards, a dog, hard surfaces | Anyone sound sensitive, and most people past an hour. The high end fatigues | | Pink | Fuller and softer than white, still clearly audible. Steady rainfall without the detail | Masking well while staying tolerable for long stretches | People who want the sound to vanish. Pink stays noticeable | | Brown | Deep, rumbling, almost physical. Waterfall, jet cabin, rain on a roof | Long sessions, restlessness, masking traffic or a neighbour's bass | People who find low frequencies oppressive, and anyone on cheap earbuds, where it turns to mud |
Brown noise went from an audio engineering footnote to a popular search term in the last few years, largely through ADHD communities on TikTok and Reddit, where people described it as the first thing that made their head feel quiet.
The lived reports are real and there are a lot of them, and when many people independently describe the same specific experience, something is happening. But a wave of people saying a thing helped is not evidence that it reliably helps, and the gap between those two statements is where most sound-and-focus content quietly cheats.
The plausible explanation is that brown noise carries no information. No consonants, no transients, no melody, nothing that rewards a listen. If your attention grabs at anything with structure, a sound with no structure gives it nothing to grab.
The research on noise and cognitive performance is genuinely mixed, and the effects are neither large nor consistent. Some people perform better on some tasks with background noise. Other people perform worse. The direction varies by individual, by task, and by volume, and the studies in this area tend to be small. You will find social media claims stated with a confidence the underlying literature does not support.
One strand is worth naming honestly, because it is the origin of most brown noise claims. A research group led by Göran Söderlund has proposed what they call a moderate brain arousal model, arguing that children with attention difficulties may be under-aroused, and that adding noise can push a noisy neural system into a more responsive range through a phenomenon called stochastic resonance. It is a real published line of work with a real mechanism, and it predicts something interesting: noise should help the people who struggle most with attention while hurting those who do not, which broadly matches what you hear anecdotally.
It is also a small literature with mixed replication, conducted mostly with white noise rather than brown, and it does not license the claim that brown noise fixes ADHD. I use noise daily and I still think the evidence for the sharper claims is thinner than the internet suggests. I would rather say so than dress it up.
I keep using it anyway because there is a second mechanism, and that one is not contested.
Masking is straightforward acoustics. A steady broadband sound raises the noise floor of the room, so quieter intermittent sounds stop standing out against it. Your neighbour's door still closes. It stops being an event.
This matters more than the arousal argument because attention is not tuned to loudness, it is tuned to change and to information. A constant rumble carries no news. A voice, a notification, a door, a sudden silence: those do, and your orienting response goes to them whether you consent or not.
The best documented case is speech. Cognitive psychologists have studied the irrelevant sound effect for decades: background speech disrupts tasks that involve holding verbal material in mind, such as writing or reading closely, and it does so even when you are not listening and cannot report what was said. Speech you can almost make out is the worst kind, meaningful enough to pull at you and unclear enough to keep pulling.
Two conclusions fall out of that, and they are the ones I would keep if I forgot everything else here:
If speech is not your problem, textured sound with slow variation may suit you better across a whole day, which is roughly the argument in why coffee shop noise works for some tasks and fails badly for others.
Most people play this too loud, and the drift is insidious, because noise raises your own hearing threshold. You set a comfortable level, your ears adapt, it seems quieter, you nudge it up. An hour later you are somewhere you would have called unpleasant at the start.
Turning the volume down usually improves the effect, which sounds backwards until you remember the goal is an uneventful room, not a drowned one.
If you want the sound without a browser tab full of ads and a video that stops when the tab sleeps, Deep Focus bundles White, Pink and Brown alongside Forest, Rain, Ocean, Thunder, Wind, Coffee Shop and Crackling Fire. They play natively rather than streaming, which matters for one unglamorous reason: the sound is not one more thing living in the browser you were trying to close.
Speakers are better whenever they are an option. The sound sits in the room rather than in your head, volume creep is slower, and you can run them all day without the fatigue of clamped headphones. The catch is that they cannot mask a person talking in the same room without getting loud enough to bother everyone else.
Headphones win for shared spaces and for speech specifically, and noise cancelling pairs well with noise colors: cancellation handles the low steady hum, the color covers the intermittent mid-range cancellation is bad at. One honest warning, though. Over-ear headphones plus brown noise plus an eight hour day is isolating, and I have finished days oddly disconnected, having heard nothing but a synthetic waterfall.
Skip the comparison content, including this one, and run the test.
A minority finish this and find all three worse than quiet, which is a legitimate result, not a failed experiment.
They differ in how power is spread across frequencies. White is flat and sounds bright and hissy, pink falls off so each octave carries equal power and sounds balanced, and brown falls off twice as steeply as pink and sounds deep and rumbling. Brown has the least high frequency content, which is why it is the least fatiguing for most people over long sessions.
Many people with ADHD report that it helps, which is worth taking seriously, but the research does not support a confident claim. The most relevant published work, the moderate brain arousal model, suggests noise may help people with attention difficulties more than others, and it is a small literature with mixed replication that mostly used white noise. Treat it as worth trying, not as an established treatment.
For verbal work the rule that matters is not the color but the absence of words, since speech and lyrics compete with the same system you use to produce sentences. Among the colors, pink and brown are easier to sustain across a real study session, while white is the strongest choice for blotting out people talking nearby.
Many people do, and steady sound can help by stopping sudden noises from standing out against the quiet. Two cautions: keep it low, because exposure across a whole night adds up, and use a speaker rather than sleeping in earbuds.
Yes, and this is underreported. Some people perform worse with background sound, particularly on demanding tasks and particularly if they are already sound sensitive. If you have persisted with noise because the internet said to, run a few sessions in silence as a control before concluding it helps.
Not the same, but related. Rain, waterfalls and ocean share a similar tilt toward the low end with slow natural variation on top, which is why they feel less flat than a synthetic color. Some find that variation more sustainable and others find it distracting, which is the subject of a separate piece on ambient sound.
The part I got wrong for years was assuming there was a best one. There is only the right one for the room you are in and how long you intend to sit there. Get the volume down and the words out, and the color is a small decision.
Get the latest productivity tips and Deep Focus updates delivered to your inbox
