Delta Waves: Slow Rhythms, Sleep and Listening

What are delta waves?

Delta waves are the slowest of the main brainwave bands: rises and falls in the brain's electrical activity at under four cycles a second. The clinical glossary of the International Federation of Clinical Neurophysiology (IFCN) sets the delta band at 0.1 to under 4 Hz, with 0.5 Hz as the practical lower limit, and a single delta wave at a quarter of a second to two seconds long.

They are recorded with EEG, electroencephalography: electrodes on the scalp pick up the summed activity of millions of neurons beneath. Each band above delta is faster. Theta runs from 4 to under 8 Hz, alpha from 8 to 13 Hz, beta from above 13 to 30 Hz, as the glossary's 2019 correction puts it, and gamma above 30 Hz. The guide to brainwave frequencies takes each band in turn, and theta, the next band up, has a guide of its own.

Delta0.5 to under 4 HzTheta4 to under 8 HzAlpha8 to 13 HzBetaabove 13 to 30 HzGammaabove 30 to 80 HzDeep sleep waves0.5 to 2 Hz, over 75 µVThe falling session3.5 slowing to 1.5 Hzcycles a second0.52832

The bands, slowest first

Band	Range
Delta	0.5 to under 4 Hz
Theta	4 to under 8 Hz
Alpha	8 to 13 Hz
Beta	above 13 to 30 Hz
Gamma	above 30 to 80 Hz
Deep sleep waves	0.5 to 2 Hz, over 75 µV
The falling session	3.5 slowing to 1.5 Hz

In deep sleep, delta waves are also the largest waves in the recording, several times taller than the faster activity of waking or dreaming, and that size is part of how deep sleep is defined.

Where does the name delta come from?

Hans Berger recorded the first human EEG in 1929 and named its main rhythm alpha a year later, as Babiloni and colleagues recount. The neurophysiologist W. Grey Walter is credited with delta. In a 1936 paper in The Lancet he used it for slow rhythms recorded near brain tumours, and within a year the name was being used for the slow waves of deep sleep. Edwards and Chang's history of the EEG spectrum sets out the sequence; theta followed in 1944.

How are delta waves measured and scored?

A sleep laboratory records the EEG all night, and a trained scorer reads it in 30-second pages called epochs, a convention the American Academy of Sleep Medicine kept when it rewrote its rules in 2007 (Silber and colleagues). Each epoch gets one stage: awake, N1 or N2 (light sleep), N3 (deep sleep) or REM, rapid eye movement sleep.

An epoch is scored as N3 when slow delta waves of 0.5 to 2 Hz, more than 75 microvolts (µV) from peak to trough and measured over the front of the head, fill at least a fifth of it. The IFCN glossary gives the wave rule, and Ishii and colleagues give the fifth. Older studies split deep sleep into stages 3 and 4; the 2007 rules merged the two into N3, and Moser and colleagues compare the two systems on the same recordings.

Deep sleep (N3)140.3 µVREM sleep27.7 µVseconds0246810

The two ten-second traces

Stage	Minutes after falling asleep	Peak to trough	Delta power over its 30 s, µV²
Deep sleep (N3)	68	140.3 µV	556
REM sleep	202.5	27.7 µV	25

That recording comes from Sleep-EDF, an open collection of nights recorded in people's homes from 1987 to 1991 and published on PhysioNet. Its electrodes sit on the forehead and crown, not where today's rules place them, so its microvolts are not a direct reading against the 75 µV line. The contrast between the two stages is the point.

When do delta waves happen in a night?

Mostly early. Sleep runs in cycles of 80 to 100 minutes, four to six a night, according to the US National Heart, Lung, and Blood Institute. Deep sleep "occurs in longer periods during the first half of the night", in the words of the NIH's guide to sleep.

The reason is sleep pressure. In the two-process model, a major framework in sleep research for three decades, a pressure to sleep builds while you are awake and is paid off while you sleep, alongside the body clock (Borbély and colleagues' reappraisal). Delta activity is its clearest trace. After 40.5 hours without sleep, eight young adults showed significantly more delta power in their recovery night, and on ordinary nights it fell from each sleep cycle to the next (Borbély and colleagues). Towards morning, slow-wave activity is lower and large slow waves are fewer (Riedner and colleagues).

Delta power
0 to 1,200 µV²AwakeREMLightDeep (N3)hours asleep0 h1 h2 h3 h4 h5 h6 h7 h

Hour by hour

Hour asleep	Deep sleep (N3), min	REM, min	Mean delta power in sleep, µV²
1st hour	9.5	0	187
2nd hour	31.5	13.5	311
3rd hour	15	0	179
4th hour	0	42	44
5th hour	4	4	45
6th hour	0.5	19.5	38
7th hour (44.5 min)	4	0	22

Download the night (CSV)

The same shape holds across all 153 nights in the Sleep-EDF collection: in the typical night, 60 per cent of deep sleep came in the first third of the time asleep.

Does deep sleep change with age?

Yes, and by a lot. A meta-analysis of 65 studies covering 3,577 people aged 5 to 102 found that the share of the night spent in slow-wave sleep falls with age through adulthood, although after 60 only sleep efficiency kept falling significantly.

0306090120150180 min

30405060708090100age in years; each square is one person, each blue line a group median

Medians a night, by age group

Ages	People	Deep sleep, min	Total sleep, h
25 to 34	20	67.8	7.2
49 to 60	22	23.5	7.0
66 to 74	20	23.6	6.9
85 to 101	16	9.1	7.2

Download all 78 people (CSV)

The Sleep-EDF nights show it plainly. The 20 people aged 25 to 34 had a median of 68 minutes of deep sleep a night. The 16 aged 85 to 101 had 9 minutes, and four of them had none at all, while total sleep stayed close to seven hours in every group. The study's own authors found that slow waves lose both size and continuity with age, and that women's slow waves were about 40 per cent larger than men's (Mourtazaev and colleagues).

Part of the fall is the ruler. Deep sleep is scored against a fixed 75 µV line, older brains make smaller slow waves, and the scoring task force found no reason to change that line at any age, so some slow activity in older sleepers falls below it. Not all of the fall is the ruler, though: continuity, which does not depend on size, declines too.

What are the benefits of delta waves?

The benefits research links to delta waves come from the brain's own slow waves in deep sleep, measured in sleepers rather than listeners:

• Paying off sleep pressure. Delta power tracks how long you have been awake and falls as the night goes on.
• Memory. When researchers strengthened the slow rhythm of early sleep, with an oscillating electrical current (Marshall and colleagues) or with short sounds timed to each slow wave (Ngo and colleagues), volunteers retained more of what they had learned before sleep.
• Fluid flow. In deep sleep, slow waves are followed by waves of blood flow and then of cerebrospinal fluid, the clear fluid around the brain (Fultz and colleagues). The study shows the three moving together, not what that coupling does.

These are findings about the slow waves the sleeping brain makes for itself. None of them shows that playing a delta-range sound does the same.

Can a sound put the brain into delta?

Two very different kinds of sound get called delta audio. The first is closed-loop stimulation in a sleep laboratory: a computer follows the sleeper's EEG and plays a short sound in time with each slow wave. Played in step, those sounds strengthened the slow rhythm; played out of step, they did nothing (Ngo and colleagues, above). The second is a recording you press play on, such as a binaural beat, which keeps its own pace whatever the brain is doing.

A binaural beat is two steady tones a little apart in pitch, one in each ear. The beat is not in the sound itself. Neurons in the brainstem that compare the two ears produce it, so it is heard only on headphones and only with tones below about 1,000 Hz (Chaieb and colleagues, after Gerald Oster's 1973 account). When the two tones are less than 3 Hz apart, the beat is heard as a tone that seems to move from ear to ear (Ingendoh and colleagues).

Left ear108 Hz, steadyRight ear111.5 slowing to 109.5 HzThe beat, first hour3.5 a secondThe beat, ninth hour1.5 a secondseconds012

Two ears, one beat

What	Rate
Left ear	108 Hz, steady
Right ear	111.5 Hz slowing to 109.5 Hz over nine hours
The beat in the first hour	3.5 a second
The beat in the ninth hour	1.5 a second

Whether the brain then follows the beat, the idea called entrainment, is unsettled. Ingendoh and colleagues' 2023 systematic review of 14 EEG studies found five in line with it, eight against it and one mixed, and concluded that the question "remains open". Three of the 14 used delta beats, and the review lists no delta entrainment among its results; one of the three found a rise in alpha instead. The review does cite an earlier study, outside its sample, that reported entrainment to delta and theta beats. A 2019 meta-analysis of 22 studies found a medium average effect of binaural beats on tests of memory and attention and on other measures, varying with the beat, how long people listened and when (Garcia-Argibay and colleagues). It measured no sleep. Do binaural beats work? follows that wider research.

What do studies of delta beats at bedtime show?

There are only a few, and they point in different directions:

• In a sleep laboratory, healthy young adults slept three nights. On the third, 15 heard a 3 Hz binaural beat on a 250 Hz tone from the moment light sleep began until deep sleep arrived, and 8 heard a silent sham. The beat group gained more deep sleep over their previous night than the sham group did (Jirakittayakorn and Wongsawat). It was one night, with small and uneven groups, and the beat began only once they were asleep.
• At home, 221 people fell asleep to an hour of pink noise carrying a 4 Hz binaural beat, the same beat in both ears, or no beat at all. How long they took to fall asleep did not depend on which they heard (Shumov and colleagues, in Russian). 4 Hz is the top edge of delta.
• In 20 healthy students, a delta beat on a 174 Hz tone did not shorten the time it took to fall asleep. On the night they heard it, the authors report more deep and REM sleep and fewer awakenings, and call for further studies (Dabiri and colleagues). There was no sham night to compare with.
• The closest sham-controlled trial of listening before bed used theta beats, not delta. 43 adults with mild sleep complaints listened to music for 30 minutes before sleep for two weeks, with or without the beats, and the beats added no clear benefit to the music alone (Bang and colleagues).

A PubMed search of reviews from 2023 to 2026, run for this guide, found no systematic review or meta-analysis of binaural beats for sleep. The nearest, a 2026 review of music and binaural beats in young adults, draws its sleep evidence from a music trial. A 2025 review of music and sleep puts it plainly: "Innovative approaches, such as brain-wave music and binaural beats, show promise but require further validation."

The stronger evidence for sound at bedtime is about music. A Cochrane review of 13 trials with 1,007 adults who had trouble sleeping found, with moderate certainty, that people who listened to music every day rated their sleep quality higher than people who did not; the three trials that measured sleep objectively found little or no difference. Steady background noise is widely used, but a systematic review of noise as a sleep aid rated the evidence for it as very low quality and warned that it may affect sleep and hearing for the worse.

What is the falling delta session?

Harmonance's Falling Delta Session is a nine-hour binaural beat made of two pure tones and nothing else. The left ear hears 108 Hz throughout. The right ear starts at 111.5 Hz and glides in one straight line to 109.5 Hz, so the beat between them starts at 3.5 Hz, near the top of the delta band, and slows to 1.5 Hz by the ninth hour. After about two and a quarter hours it drops below 3 Hz, where a beat starts to feel like movement rather than a pulse. At nine hours it stops.

The idea is a sound that slows as the night does. That is a design choice, not a mechanism: no study has tested this recipe, and the studies above give no reason to expect the brain to follow it. Its 108 Hz tone sits well inside the range where a binaural beat can be heard, and the studies above used tones of 174 to about 250 Hz; no best tone for a delta beat has been agreed.

The uploaded nine-hour master, measured

Moment	Left ear	Right ear	Beat
First second	108.00 Hz	111.50 Hz	3.50 Hz
Halfway, 4 h 30 min	108.00 Hz	110.50 Hz	2.50 Hz
Last second	108.00 Hz	109.50 Hz	1.50 Hz

Harmonance measured the uploaded master itself, not only the recipe. Its loudness reads -23 LUFS, and its peaks stay more than 21 dB below the loudest a recording can go.

How do you listen at bedtime?

• Use headphones. The beat exists only when each ear gets its own tone. Through a speaker the two tones mix in the air and make an ordinary wobble in loudness instead, a monaural beat; binaural beats and isochronic tones explains the difference. Soft over-ear headphones or flat earbuds are easier to lie on.
• Keep it quiet. The World Health Organization's safe listening advice is to keep a device below 60 per cent of its maximum, and its listening standard with the ITU sets 80 dB for 40 hours a week as the adult limit. Nine hours a night is 63 hours a week, so for a session this long, quieter than conversation is the sensible level. The 2018 laboratory study played its beat at 60 dB, about the level WHO gives for normal conversation.
• Keep a steady routine. The NHS suggests getting up at the same time every day, relaxing for at least an hour before bed and keeping the bedroom dark and quiet (NHS advice on sleeping habits). If you are still awake after about 20 minutes, it suggests getting up and doing something relaxing, such as listening to quiet music (Every Mind Matters).
• Notice, then decide. If the beat feels intrusive, turn it down or stop. A pure tone with no beat, or quiet music, is a fine alternative.

If you are here because you cannot sleep tonight, Can't sleep? gathers gentle options in one place.

What have listeners said?

Comments are not evidence, and these come from a different recording: the channel's original 1.618 Hz delta beat, played on a 432 Hz tone. The site's 1.618 Hz golden ratio beat plays the same beat on a 108 Hz tone. They are quoted whole, without names.

I slept to this and it was great

While I was reading the description of this video the sound I hear are like going left to right then I focused and now I can only hear a almost steady with small fluctuations

It sounds different at different volumes, is there a particular sound that's preferable to another? Is a continuous sound better or a pulsating one?

I just found this to be irritating. Like tinnitus.

The second describes the moving quality that research reports for slow beats. The third asks what many listeners ask. On volume, the advice above is clear: keep it low. A steady tone or a pulsing one comes down to which you find easier to rest with.

Sources and limits

The night and age figures come from one open collection of healthy people recorded from 1987 to 1991, scored by the older Rechtschaffen and Kales rules on forehead electrodes. Harmonance merged stages 3 and 4 into N3, measured delta power from the recordings and averaged each person's nights; each chart's numbers can be downloaded beside it. The studies of delta beats are few, small and mixed, and none tested the falling session. The comments come from a different recording.

1. Kane and colleagues, IFCN glossary of EEG terms: the frequency bands, delta waves and deep sleep.
2. Kane and colleagues, correction to the IFCN glossary: the beta band.
3. Babiloni and colleagues, on Hans Berger and the alpha rhythm: the first human EEG and the naming of alpha.
4. Walter, 1936, The Lancet: the paper that introduced delta.
5. Edwards and Chang, syllabic and fluctuation ranges in speech: a history of the EEG band names.
6. Silber and colleagues, the visual scoring of sleep in adults: 30-second epochs and the amplitude rule.
7. Ishii and colleagues, slow-wave sleep from macro to micro: the rule for scoring N3.
8. Moser and colleagues, AASM and Rechtschaffen and Kales compared: stages 3 and 4 merged into N3.
9. Sleep-EDF Database Expanded, PhysioNet: the recordings behind the charts, used under the Open Data Commons Attribution licence.
10. Kemp and colleagues, the slow-wave microcontinuity of the EEG: the paper PhysioNet asks users of Sleep-EDF to cite.
11. Mourtazaev and colleagues, age and gender in slow waves: the Sleep Cassette study's own findings.
12. NHLBI, sleep phases and stages: cycle length and the number of cycles.
13. NINDS, Brain Basics: Understanding Sleep: deep sleep in the first half of the night.
14. Borbély and colleagues, the two-process model reappraised: sleep pressure and the body clock.
15. Borbély and colleagues, sleep deprivation and EEG power: delta power after long waking and across the night.
16. Riedner and colleagues, slow waves early and late in the night.
17. Ohayon and colleagues, sleep across the lifespan: slow-wave sleep and age.
18. Marshall and colleagues, boosting slow oscillations during sleep: memory.
19. Ngo and colleagues, closed-loop sound in sleep: sounds timed to slow waves.
20. Fultz and colleagues, slow waves, blood flow and cerebrospinal fluid.
21. Chaieb and colleagues, auditory beat stimulation: how a binaural beat is made.
22. Oster, Auditory Beats in the Brain, Scientific American: the classic description.
23. Ingendoh and colleagues, binaural beats to entrain the brain?: the systematic review of EEG studies.
24. Garcia-Argibay and colleagues, meta-analysis of binaural beats: memory, attention and other measures.
25. Jirakittayakorn and Wongsawat, a 3 Hz binaural beat during sleep.
26. Shumov and colleagues, binaural and monaural beats at sleep onset, in Russian.
27. Dabiri and colleagues, a delta binaural beat and EEG in sleep.
28. Bang and colleagues, theta beats with music before sleep.
29. Elnazer, music and binaural beat interventions for young adults.
30. Pan and Wang, elements of music for sleep.
31. Jespersen and colleagues, Cochrane review of music and sleep.
32. Riedy and colleagues, noise as a sleep aid.
33. WHO, safe listening: volume and weekly listening time.
34. WHO and ITU, safe listening devices and systems: the 80 dB, 40-hour adult standard.
35. NHS, sleeping habits.
36. NHS Every Mind Matters, sleep tips.
37. The original 1.618 Hz delta beat on Harmonance's YouTube channel: the listener comments.