Nightjar's evidence labels, explained
Why a sound app labels its evidence
Sound for sleep and calm sits between two loud camps: sites that sell beats and tones as remedies for almost anything, and sites that dismiss the whole thing. The research is somewhere in the middle and differs a lot from one sound to the next. Slow breathing has hundreds of studies; a 741 Hz tone has none. Nightjar gives every part of every track a label, so you can see which is which before you press play.
The five labels
| Label | What it means | An example in Nightjar |
|---|---|---|
| Backed by trials | Trials in people point the same way. | Slow breathing, around six breaths a minute. Not a sound, and the only note with this label. |
| Early evidence | A few small or low-certainty studies. | Singing bowls, as a part of the Singing Bowls track and the steady tones. |
| Mixed | Studies disagree. | Binaural beats and isochronic pulses; noise beds for sleep. |
| Tradition | Said for a long time. Rarely tested. | The solfeggio tones, such as 528 Hz and 741 Hz. |
| Comfort sound | No research reviewed. A sound many people like, not a treatment. | Rain, fire, the sea, crickets, a train, slow ambient music. |
Backed by trials: slow breathing
Breathing slowly, around six breaths a minute, reliably raises heart-rate variability: a 2022 review in Neuroscience and Biobehavioral Reviews pooled 223 studies. In a 2023 meta-analysis of 12 randomised trials with 785 adults in Scientific Reports, breathwork modestly lowered self-reported stress. Nightjar’s note adds the caveats: six a minute is an average and your best pace may differ, and for sleep the trials relied on how people said they slept. No sound in Nightjar carries this label, because no trial has tested what Nightjar plays. See slow breathing and calm.
Early evidence: singing bowls
In small studies people felt calmer after singing-bowl sessions. A 2020 review in Complementary Therapies in Medicine found only four studies and could not recommend bowls yet; a 2025 review of 19 clinical studies, 9 of them randomised, in Integrative Medicine Research said they “may have potential”. The trials that describe the bowls used live ones, and most compared them with silence or a waiting list.
Mixed: beats and noise beds
Whether binaural beats change your brainwaves is unsettled: of 14 brain-recording studies in a 2023 review in PLOS ONE, 5 found it, 8 did not and 1 was mixed. Small trials before surgery and dental work found less anxiety, but listeners know what they are hearing. Noise for sleep is Mixed too: a 2021 review of 38 studies in Sleep Medicine Reviews rated the evidence that it improves sleep as very low certainty. See do binaural beats work? and does white noise help you sleep?
Tradition: the solfeggio tones
Nine of the ten steady tones come from a modern “solfeggio” list. Each note quotes the tradition as a tradition (“Said to…”) and says what was found, which for most is nothing: no study has tested what a 174 Hz tone does. Where a claim nothing supports is common, it appears only next to its denial. 432 Hz is the exception among the tones and is labelled Mixed, because two small pilot studies compared it with 440 Hz. See solfeggio frequencies.
Comfort sound: rain, fire and most of the library
Rain, fire, the sea, a stream, a fan and the slow soundscapes have no research reviewed behind them. Many people find them calming; that is comfort, not a treatment. It is the honest label for most of what anyone listens to at night.
How a track gets its one label
A track is made of parts: perhaps a beat, a noise bed, a few sounds and a soft pad. Each part has its own note. The track shows one label, from the part research has looked at most: the beat or tone if there is one, otherwise the noise bed, otherwise the sounds. A track with nothing to read is a comfort sound.
So Soft Landing is labelled Mixed, from its beat, though its rain is a comfort sound and its soft pad has a comfort note too. Skyline Rain, rain heard through a closed window, is a Comfort sound. Every part’s note is there to read, with who found it and when.
What the labels never say
No note says a sound “cures”, “heals”, “treats” or “prevents” anything, nothing is called “proven”, and no track is labelled Backed by trials unless the trial was of what the track plays. A tradition is reported as a tradition. A claim such as “repairs DNA” appears only beside the words “Nothing shows that.”
Where Nightjar helps, and where it doesn’t
The labels let you choose with your eyes open: a beat because you like it and the research is Mixed, rain because it is a comfort and that is enough. Research on binaural beats is mixed, and the app says so. The Lab runs fair A/B nights in random order, and its results say not sure yet until there are five nights of each.
The labels cannot tell you what a sound will do for you, and they are only as good as the last reading of the research. They summarise studies of other people, often small ones; your own nights are a different and useful kind of evidence.
Sources
- Laborde and colleagues, Effects of voluntary slow breathing on heart rate and heart rate variability: a systematic review and a meta-analysis, Neuroscience and Biobehavioral Reviews, 2022
- Fincham and colleagues, Effect of breathwork on stress and mental health: a meta-analysis of randomised-controlled trials, Scientific Reports, 2023
- Stanhope and Weinstein, The human health effects of singing bowls: a systematic review, Complementary Therapies in Medicine, 2020
- Cai and colleagues, Therapeutic effects of singing bowls: a systematic review of clinical studies, Integrative Medicine Research, 2025
- Ingendoh, Posny and Heine, Binaural beats to entrain the brain? A systematic review, PLOS ONE, 2023
- Riedy and colleagues, Noise as a sleep aid: a systematic review, Sleep Medicine Reviews, 2021
Last checked 13 November 2026.
Questions
Why does Nightjar show labels that make its own tracks look weak?
Because the research is what it is. Most sleep and focus sounds have weak or no evidence behind them, and a label that only appeared when it flattered a track would tell you nothing.
Does Comfort sound mean a track does not work?
It means no research was reviewed for that part, usually because there is none. Rain or a fire can still be a sound you love falling asleep to; the label just says that is comfort, not a measured effect.
How often are the labels checked?
The notes behind them carry the date they were last read against the research. At the time of writing that date is 30 Oct 2026, and a note changes when a new study changes what can honestly be said.