Pular para o conteúdo
Mil Dias·

White noise for babies

What the evidence says — and the limits no one tells you

White noise helps many babies fall asleep faster — a real, replicated effect since 1990. But more recent research shows that machines marketed for babies frequently exceed safe volume limits. Not a remedy, not a poison: it's a tool with a narrow useful window.

4 min read
Última atualização: May 9, 2026

White noise is a continuous broadband sound — all audible frequencies at roughly equal intensity, like the static of an off-air TV or the hum of a distant fan. The version preferred for babies tends to be brown noise (deeper) or pink noise (intermediate), closer to what the fetus heard in the womb — maternal blood flow and heartbeats, around 85-95 dB.

Marketing claims aside, the scientific evidence for use with babies is real, but small, and surrounded by cautions few manufacturers mention.

What the evidence shows

It works to induce sleep — well-documented immediate effect. In Spencer's seminal study at St. George's Hospital in London, 80% of newborns exposed to continuous white noise fell asleep within 5 minutes, against 25% in the control groupSpencer et al. 1990. Subsequent studies and NICU clinical use confirm the short-term effect: the sound masks awakenings from ambient noise and mimics the uterine environment.

But the volume of most machines is dangerous. In 2014, Hugh and colleagues at Toronto's Hospital for Sick Children tested 14 white noise machines marketed for babies. All of them, at maximum volume 30 cm from the crib, exceeded the 50 dB limit recommended for neonates over prolonged periods. Three exceeded 85 dB — the OSHA adult occupational limit, above which prolonged exposure causes hearing lossHugh et al. 2014.

There's experimental signal warranting caution with continuous use. In rats, Chang and Merzenich at UCSF showed that exposure to continuous white noise during the critical period delayed tonotopic maturation of the auditory cortexChang & Merzenich 2003. This is an animal model — direct extrapolation to humans isn't possible — but it's why audiologists advise moderation with prolonged use, especially in the first months.

Myth

White noise causes autism or speech delay (viral myth on social media).

Evidence

There is no evidence in humans that proper white noise use causes autism. The caution studies are in rats with continuous exposure at high volume. The real and demonstrated risk is hearing damage from excessive volume — not autism.

Safe-use protocol

Five rules that follow directly from the evidence:

  1. Volume ≤ 50 dB. Practical reference: if you need to raise your voice to talk over the sound, it's too loud. Free decibel meter apps on phones work well to check.
  2. Machine ≥ 2 meters from the crib. Never inside it. The farther, the lower the intensity reaching the ear.
  3. For falling asleep, not the whole night. Most defensible pattern: turn on at bedtime, turn off or fade after 20-30 minutes. A timer solves it.
  4. Broadband and continuous, no beats. Pure white, brown, or pink noise. Avoid tracks with pulsing at high volume, electronic beats, or music with lyrics — the calming effect depends on spectral continuity.
  5. Doesn't replace co-regulation or safe sleep. Empty crib (no pillows, loose blankets, toys), back-sleeping, in the parents' room until 6-12 monthsAAP 2022. White noise is additional, not a substitute.

When it actually helps

  • Noisy home (apartment with neighbors, older siblings, busy street) — masking awakenings from sudden noise is the best-supported use.
  • Colic and difficult late-afternoon period — combined with holding, skin-to-skin, and movement.
  • Naps in unfamiliar settings (travel, grandparents' house) to reduce awakenings from strangeness.
  • Transition from the "noisy maternity room" phase to the relative silence of home in the first weeks.

When it's not worth it (or can backfire)

  • As the only sleep strategy, without addressing routine, environment, and regulation.
  • At high volume all night — even an adult shouldn't sleep at that intensity.
  • To "train" the baby to ignore the adult — not a valid or evidence-based goal.
  • If the baby sleeps well without it. There's no preventive benefit in adding a tool that isn't needed.

About dependency

Some babies form an association between white noise and sleep and come to need it to fall asleep. This is not "wrong" — it's an association like any other (sucking, rocking, dark). It only becomes a problem if the family wants to stop and the transition is difficult. Strategies to reduce use: lower volume gradually over 1-2 weeks, or shorten the time before turning off.

References

  1. Spencer, J. A. et al. (1990). White noise and sleep induction. Archives of Disease in Childhood, 65(1). doi:10.1136/adc.65.1.135
  2. Hugh, S. C. et al. (2014). Infant sleep machines and hazardous sound pressure levels. Pediatrics, 133(4). doi:10.1542/peds.2013-3617
  3. Chang, E. F. & Merzenich, M. M. (2003). Environmental noise retards auditory cortical development. Science, 300(5618). doi:10.1126/science.1082163
  4. American Academy of Pediatrics — Task Force on Sudden Infant Death Syndrome (2022). Sleep-Related Infant Deaths: Updated 2022 Recommendations for Reducing Infant Deaths in the Sleep Environment. Pediatrics, 150(1). doi:10.1542/peds.2022-057990

Related articles