A falsification-oriented evidence synthesis of continuous broadband noise, environmental masking, sleep architecture and the limits of the “sleep-promoting noise” hypothesis.
Publication type: Research Report / Evidence Synthesis
Version: 1.0
Date: 3 October 2026
Institution: The Null Institute
Author: Ilya Nikitin, Ph.D., Founder and Director, The Null Institute
Status: Evidence synthesis; not externally peer reviewed
Abstract
Background. White noise and pink noise are widely used as sleep aids. Two different mechanisms are often blended together: masking unpredictable environmental sounds and directly improving sleep physiology in otherwise quiet conditions.
Objective. To test the stronger claim that continuous broadband noise itself improves human sleep, and to distinguish that claim from the narrower hypothesis that steady sound can protect sleep by masking intermittent environmental noise.
Methods. We used a 2022 systematic review as an evidence-map anchor and then examined later controlled human studies and key earlier experiments capable of separating masking from intrinsic sleep promotion. We treated multiple publications from the same participant sample as one experimental dataset and separated continuous overnight noise from phase-locked acoustic stimulation.
Results. The literature is heterogeneous and does not establish a general sleep-enhancing effect of continuous white or pink noise. Small controlled studies support a masking mechanism in some noisy environments. A 2026 seven-night polysomnographic crossover study in 25 healthy adults found that continuous pink noise reduced REM sleep and that earplugs protected sleep from environmental noise more effectively. A separate 2026 crossover pilot in 12 healthy adults found that 45 dB pink noise attenuated some traffic-noise-related sleep fragmentation and metabolic changes. A 30-day trial in 60 older adults reported improved subjective sleep after an intervention labelled white noise, but the audio consisted of nature-sound recordings and the design did not isolate spectral white noise from expectation, preference or environmental masking. Evidence that pink noise is superior to white noise is not established.
Conclusion. The surviving hypothesis is conditional rather than universal: continuous broadband sound can sometimes reduce the disruptive contrast of intermittent environmental noise, but adding sound to an already quiet sleep environment has not been shown to reliably improve sleep and may alter sleep architecture. “Masking external noise” and “promoting sleep” should be treated as different claims.
1. Research question
The popular question is whether white noise or pink noise helps people sleep. That formulation hides the key causal distinction.
Does continuous broadband sound improve sleep itself, or does it mainly help when it masks a noisier and more irregular acoustic environment?
Those mechanisms make different predictions. If broadband noise is intrinsically sleep-promoting, it should improve sleep relative to quiet. If masking is the main mechanism, benefit should be strongest when disruptive external sound is present, while adding continuous noise to quiet conditions may do little or may itself disturb sleep.
2. Evidence strategy and provenance
This report is an AI-assisted evidence synthesis, not a prospectively registered systematic review or meta-analysis.
The investigation began on 27 September 2026 and was refreshed on 3 October 2026. Evidence collection followed two stages. First, the 2022 systematic review by Capezuti et al. was used as an evidence-map anchor. It included 34 studies involving 1,103 participants and reported that the evidence was too heterogeneous and inconsistent to support auditory stimulation strongly as a general sleep intervention [1]. Its abstract reports searches of nine databases; the full methods text names eight database platforms. We therefore preserve the authors’ reported search scope without using the database count as an evidentiary claim.
Second, we searched for controlled human evidence published after the review and for earlier primary studies that could discriminate between the masking and intrinsic-sleep-promotion hypotheses. Searches combined terms relating to white noise, pink noise, broadband noise, sleep, environmental-noise masking, REM sleep, sleep architecture, earplugs and hearing safety.
Primary reports and durable records were preferred over secondary summaries. Sources were checked through PubMed/PMC, journal publisher pages, the U.S. National Transportation Library/ROSA-P archive and relevant public-health sources.
Publications were reconciled at the study level rather than counted independently merely because they appeared as separate papers. In particular, the 2026 Sleep paper by Basner et al. and the later Sleep Health analysis use the same registered 25-participant trial, NCT05774977, and are treated here as one experimental dataset [3,4].
Brief phase-locked acoustic pulses delivered at selected phases of slow oscillations were analysed separately from continuous broadband sound. These are materially different interventions.
3. What the older evidence actually established
The 2022 systematic review found 34 adult studies and 1,103 participants across white noise, pink noise and mixed auditory interventions [1]. Nineteen studies reported at least one positive sleep finding, but study designs, sound spectra, timing, environments, populations and outcomes varied substantially. A meaningful meta-analysis was not possible. The review’s own conclusion was that there was no strong evidence supporting auditory stimulation as a sleep intervention.
This matters because a simple count of “positive studies” overweights small, heterogeneous studies and does not answer the causal question. A sound can appear helpful because it replaces a worse acoustic condition rather than because the sound itself promotes sleep.
4. A clean demonstration of masking
Stanchina et al. exposed healthy sleepers to recorded intensive-care-unit noise [2]. Four participants completed all three polysomnographic conditions, while a fifth completed the first two. The arousal index was 13.3 per hour during baseline sleep, 48.4 per hour during ICU noise and 15.7 per hour when mixed-frequency white noise was added to the ICU recording.
The sample is far too small for a broad clinical claim. But mechanistically the experiment is informative: arousal probability tracked the change from acoustic background to peak sound more closely than the absolute peak. Raising the steady background reduced the contrast of the intermittent events.
That is evidence for masking. It is not evidence that white noise makes sleep intrinsically deeper in a quiet room.
5. The strongest modern falsification: continuous pink noise versus quiet
Basner et al. studied 25 healthy adults across seven laboratory nights using polysomnography [3]. Conditions included a noise-free control, 93 intermittent environmental-noise events at 45–65 dBA, continuous pink noise at 40 or 50 dBA, foam earplugs and combinations of these exposures.
Environmental noise reduced N3 sleep. Continuous pink noise reduced REM sleep. Adding pink noise to environmental noise produced some dose-dependent improvements in event-related fragmentation and N3 loss, but overall sleep architecture deteriorated rather than improved. Earplugs mitigated nearly all environmental-noise effects more effectively, except at the highest noise level.
This is difficult to reconcile with a universal “pink noise promotes sleep” hypothesis. It is compatible with a narrower model in which continuous sound may mask discrete events while simultaneously imposing its own physiological cost.
A later publication from the same NCT05774977 trial examined event-specific arousals and awakenings [4]. Pink noise reduced some aircraft-noise-triggered events in a dose-responsive manner but remained less effective than earplugs. Because the participants and protocol are the same, this paper strengthens the mechanistic interpretation of the trial but does not constitute an independent replication.
6. Independent pilot: traffic noise plus pink noise
Vincens et al. provide an independent controlled test [5]. Twelve healthy adults slept for five consecutive nights under quiet, traffic-noise, 45 dB continuous pink-noise and combined conditions. Traffic-noise events increased a physiological index of sleep fragmentation and were accompanied by changes in several blood metabolites. Adding pink noise attenuated some of these acute effects.
The result supports the masking branch, but the authors explicitly describe the study as exploratory and caution that the sample was small and homogeneous. It does not demonstrate that continuous pink noise improves sleep relative to an already quiet night.
7. A contrary 30-day trial, and why it does not settle the question
A 2026 parallel randomized trial in 60 community-dwelling adults aged 65 years or older reported improved subjective sleep quality and reduced fatigue after 30 nights of an intervention described as white noise [6]. PSQI scores declined from 11.5 to 9.2 in the intervention group while remaining approximately unchanged in controls.
This is relevant contrary evidence and should not be discarded. However, it does not cleanly test conventional spectral white noise. Participants chose among recordings described as river, forest, rain and sea waves; the intervention was delivered for 30 minutes at bedtime, background environmental noise was not controlled, participants could not be blinded, and the main outcomes were self-reported. Ten of 30 intervention participants specifically said the sound helped mask disturbing environmental noise.
The trial therefore supports the possibility that preferred bedtime sound can improve subjective sleep in some older adults. It does not distinguish spectral white noise from relaxation, expectancy, preference or masking.
8. Is pink noise better than white noise?
We found no strong basis for a general superiority claim. The older literature contains very few clean head-to-head comparisons, and the 2022 systematic review concluded that heterogeneity prevented a useful pooled estimate [1]. Differences in spectrum are often entangled with differences in loudness, device, timing and acoustic context.
The evidence therefore supports a practical distinction more strongly than a colour ranking: continuous masking sound versus quiet, and masking sound versus intermittent environmental noise.
9. Why “pink noise improves deep sleep” can mean something else entirely
Some experiments use brief pink-noise pulses delivered at precise phases of endogenous slow oscillations rather than continuous sound. Papalambros et al., for example, used a real-time phase-locked algorithm in 13 cognitively healthy older adults and found transient increases in slow-wave activity and improved overnight word-pair recall [7].
This is a neurostimulation protocol. It should not be used as evidence that leaving continuous pink noise playing through the night has the same effect.
10. Safety and long-term uncertainty
Short-term studies do not provide a strong basis for claims about nightly use over months or years. Questions about adaptation, tolerance, chronic sleep-architecture effects and long-term auditory exposure remain under-studied.
NIOSH uses 85 dBA averaged over an eight-hour workday as an occupational recommended exposure limit; NIOSH explicitly states that this is not a recommended general-environment or bedroom level [8]. For sleep applications, the evidentially conservative principle is therefore to avoid unnecessary intensity and use the lowest level that achieves the intended masking effect.
Infant devices deserve additional caution. Testing of 14 infant sleep machines found that all exceeded 50 dBA at maximum volume and 30 cm, while three exceeded 85 dBA [9]. This is a device-output study, not evidence that ordinary low-level use causes hearing loss, but it shows that some devices are capable of producing levels high enough to make exposure conditions important.
11. Competing explanations
| Hypothesis | Expected observation | Current evidence |
|---|---|---|
| Continuous broadband noise directly promotes sleep | Improvement versus quiet across objective sleep measures | Weak / contradicted in key modern PSG trial |
| Broadband noise masks intermittent disturbance | Benefit mainly when irregular external noise is present | Supported in small controlled studies |
| Pink noise is generally superior to white noise | Matched head-to-head advantage across conditions | Not established |
| Timed pink-noise pulses improve slow-wave physiology | Phase-specific EEG effects under closed-loop stimulation | Supported, but it is a different intervention |
| Any bedtime sound can improve subjective sleep through non-acoustic mechanisms | Preference/expectancy effects without objective architecture change | Plausible; incompletely separated |
12. Limitations
- This was a targeted evidence synthesis, not a prospectively registered systematic review.
- The search was refreshed but not designed to guarantee exhaustive capture of every auditory-sleep study.
- Many studies are small, use different spectra and sound levels, and mix objective and subjective outcomes.
- Long-term nightly exposure is poorly studied.
- Most rigorous laboratory experiments involve healthy adults, limiting generalisation to insomnia, children, older adults with multimorbidity and other clinical populations.
- Preference, expectancy and masking are difficult to separate in unblinded home studies.
- The absence of demonstrated long-term harm is not evidence of long-term safety.
13. Conclusion
The original broad hypothesis does not survive intact.
Continuous white or pink noise is not established as a universal sleep enhancer. The more defensible model is conditional: steady broadband sound can sometimes reduce disruption from intermittent environmental noise by lowering acoustic contrast. That mechanism can coexist with costs imposed by the continuous sound itself.
The 2026 polysomnographic evidence is especially important because it directly separates these effects: pink noise reduced some event-related fragmentation while also reducing REM sleep and worsening the overall sleep structure in some conditions [3]. A separate small pilot supports partial masking of traffic noise [5]. A home trial in older adults provides contrary subjective evidence, but its intervention and design do not isolate spectral white noise [6].
The question is not simply whether sound helps sleep. It is whether adding continuous sound produces a net benefit relative to the acoustic environment it replaces.
That reframing explains why apparently conflicting studies can all be partly correct.
Methods and evidence provenance
This report used a falsification-oriented search strategy. Evidence inconsistent with the initial sleep-enhancement hypothesis was retained rather than treated as an outlier. Study-level provenance was checked to avoid double-counting multiple publications from the same experimental sample. No quantitative meta-analysis was performed because study populations, spectra, exposure levels, duration, background acoustic conditions and outcomes were too heterogeneous for a single pooled effect to be informative.
Where a preferred full-text route was inaccessible, claims were checked against alternative durable records such as PubMed, the journal version of record or the National Transportation Library archive. No substantive conclusion in this report depends on a secondary news report.
AI assistance
AI systems were used for literature discovery, source triage, extraction, comparison of competing explanations, adversarial review and drafting. AI-generated text was not treated as evidence. Factual claims used in the report were traced back to identifiable source material. Final responsibility for source selection, interpretation and publication remains with the human author.
Funding and competing interests
Funding. No external project-specific funding was received for this report.
Competing interests. The author has no financial or commercial interest in white-noise or pink-noise devices, sleep-sound applications, hearing-protection products or other products discussed in this report.
Relevant disclosures in the underlying literature were recorded during appraisal. The Basner et al. NCT05774977 study was funded by the U.S. Federal Aviation Administration through ASCENT; the paper reported no financial conflicts and disclosed that M. Younes developed and holds a patent on the ORP technology used in analysis [3]. The independent Vincens et al. pilot was funded internally through a Sahlgrenska Academy International Starting Grant; M. Younes disclosed a patent, licensing, shareholding, royalties and consulting fees relating to ORP technology [5]. These disclosures were not treated as grounds for excluding either study.
Data, code and ethics
No new participant-level dataset was collected or analysed for this evidence synthesis, and no new statistical analysis code was produced. Public study records and datasets remain available from their original repositories where provided by the investigators. This report analyses published and publicly available information; no new human participants were recruited and no identifiable personal data were accessed.
Peer review and version status
This report has undergone internal adversarial evidence review but has not undergone external peer review. Version 1.0 reflects evidence identified through 3 October 2026. Material corrections or substantive changes in interpretation should be recorded through version updates and a public change history.
References
- Capezuti E, Pain K, Alamag E, Chen XQ, Philibert V, Krieger AC. Systematic review: auditory stimulation and sleep. J Clin Sleep Med. 2022;18(6):1697–1709. doi:10.5664/jcsm.9860.
- Stanchina ML, Abu-Hijleh M, Chaudhry BK, Carlisle CC, Millman RP. The influence of white noise on sleep in subjects exposed to ICU noise. Sleep Med. 2005;6(5):423–428. doi:10.1016/j.sleep.2004.12.004.
- Basner M, Smith MG, Cordoza M, et al. Efficacy of pink noise and earplugs for mitigating the effects of intermittent environmental noise exposure on sleep. Sleep. 2026;49(5):zsag001. doi:10.1093/sleep/zsag001.
- Basner M, Smith MG, Cordoza M, et al. The efficacy of pink noise and earplugs for mitigating sleep disruption induced by different environmental noise sources. Sleep Health. 2026;12(4):702–706. doi:10.1016/j.sleh.2026.03.009. Secondary analysis of the same NCT05774977 trial as reference 3.
- Vincens N, Nause A, Basner M, et al. Pink noise reduces impact of traffic noise on sleep and the blood metabolome: a cross-over pilot study. Commun Med. 2026;6:114. doi:10.1038/s43856-026-01380-5.
- Vahhabzadeh Mousavi SA, Salehian M, Mazlom SR. The effect of white noise on sleep quality and fatigue in community-dwelling older adults: a randomized controlled trial. BMC Geriatr. 2026;26:883. doi:10.1186/s12877-026-07311-2.
- Papalambros NA, Santostasi G, Malkani RG, et al. Acoustic Enhancement of Sleep Slow Oscillations and Concomitant Memory Improvement in Older Adults. Front Hum Neurosci. 2017;11:109. doi:10.3389/fnhum.2017.00109.
- National Institute for Occupational Safety and Health. Understanding Noise Exposure Limits: Occupational vs. General Environmental Noise. CDC/NIOSH.
- Hugh SC, Wolter NE, Propst EJ, Papsin BC. Infant Sleep Machines and Hazardous Sound Pressure Levels. Pediatrics. 2014;133(4):677–681. doi:10.1542/peds.2013-3617.