- Delta waves are EEG rhythms below 4 Hz; the international clinical glossary (Kane 2017) defines the delta band as 0.1 to under 4 Hz, in practice from 0.5 Hz.
- Deep sleep (stage N3, slow-wave sleep) is scored when waves of 0.5–2 Hz and at least 75 microvolts fill at least 20 % of a 30-second epoch.
- In young and middle-aged adults, deep slow-wave sleep makes up about 20 % of the night and comes mostly in its first half.
- After 40.5 hours without sleep, delta power in the first recovery night was clearly above normal (Borbély and colleagues, 1981, 8 young adults).
- A meta-analysis of 65 studies and 3,577 people (Ohayon and colleagues, 2004) found that the share of slow-wave sleep falls with age from childhood on.
- Generalised delta activity in an awake adult is considered abnormal on a clinical EEG, most often a sign of encephalopathy.
What are delta waves?
Delta waves are the slowest rhythm that an ordinary EEG records: waves that rise and fall fewer than four times a second. In a healthy adult they appear mainly in deep sleep, where a clinical review calls them "one of the few instances where delta waveforms are considered normal" (Alcala-Zermeno and colleagues, 2025).
An EEG records voltage changes of a few to about a hundred millionths of a volt from electrodes on the scalp, usually placed by the international 10–20 system. The signal comes mainly from the synaptic currents of large groups of cortical neurons working in step, not from single cells (Buzsáki, Anastassiou and Koch, 2012). Even researchers say much is still unknown about exactly how scalp EEG arises (Cohen, 2017).
A brain-wave band is a frequency slice of one recording, not a mode the brain switches into. An EEG never shows one wave at a time: slow and fast rhythms run together, and the slower ones are larger, because power in a brain recording falls steadily as frequency rises (He, 2014). That is why delta, the slowest band, is also the tallest on the page when it is present.
Delta waves frequency: how many hertz?
The usual answer is 0.5–4 Hz. The International Federation of Clinical Neurophysiology glossary (Kane 2017) sets the delta band at 0.1 to under 4 Hz, but in practice it starts at 0.5 Hz, because ordinary EEG does not record very slow shifts; anything below 0.1 Hz is called infraslow activity. Clinical reviews give delta as 0.5–4 Hz.
Band edges differ between sources. A review of 184 resting-EEG studies (Newson and Thiagarajan, 2018) found that the most common definition of each band was used by only 30–50 % of studies, and delta started anywhere from 0 to 2 Hz and ended between 3.5 and 6 Hz.
| Definition | Delta range | Note |
|---|---|---|
| IFCN clinical glossary (Kane 2017) | 0.1 (in practice 0.5) to under 4 Hz | The standard used on this site's chart |
| Most common in 184 studies (Newson 2018) | 1.3–3.5 Hz | Used by a minority of studies |
| Range found in 184 studies (Newson 2018) | 0–6 Hz | From the lowest start to the highest end |
| Deep sleep scoring (N3) | 0.5–2 Hz, at least 75 µV | Only the slowest, largest waves count |
| Slow oscillation (Achermann and Borbély 1997) | below 1 Hz, peak 0.7–0.8 Hz | Behaves differently from delta above 2 Hz |
Researchers separate the slow oscillation, below 1 Hz, from delta waves of about 1–4 Hz. In all-night recordings of 8 healthy young men, deep sleep showed a distinct peak at 0.7–0.8 Hz. The usual drop in delta from the first to the second NREM episode did not appear below 2 Hz, so the two behave as different rhythms (Achermann and Borbély, 1997). When articles talk about "delta" in sleep, they often mean both together.
Delta waves and the stages of sleep
A sleep lab divides the night into 30-second pages, called epochs, and gives each one a stage. NREM sleep has three stages (N1, N2, N3); the older Rechtschaffen and Kales system of 1968 had four, and its stages 3 and 4 were merged into N3. In N2, delta appears as single K-complexes, large two-phase delta waves over the centre of the head, often with a sleep spindle on top. In N3 they fill the recording.
Light sleep (N1 and N2) fills more than half of the night, deep slow-wave sleep about a fifth and REM sleep about a fifth to a quarter, in cycles of roughly 90 minutes. Most people have four to five cycles per night. The first REM period comes after about 90 minutes and is short; REM periods grow longer towards morning, while most slow-wave sleep comes in the first NREM period (Steiger and Pawlowski, 2019). The hypnogram above is a schematic of this pattern, not a real recording.
| Stage | Main EEG signs | Typical share of the night (young or middle-aged adult) |
|---|---|---|
| Awake, relaxed, eyes closed | Alpha 8–12 Hz at the back of the head | Not counted |
| N1 | Alpha fades; theta 5–8 Hz; vertex sharp waves; slow rolling eye movements | 5–10 % |
| N2 | Sleep spindles 11–16 Hz (at least 0.5 s); K-complexes | About 50 % |
| N3 | Waves of 0.5–2 Hz, at least 75 µV, in at least 20 % of a 30 s epoch | About 20 %, more in the first half of the night |
| REM | Low-amplitude mixed theta and alpha; sawtooth waves 2–6 Hz; rapid eye movements; muscle atonia | 20–25 %, more in the second half of the night |
The shares come from a recent clinical review (Alcala-Zermeno and colleagues, 2025), which adds that they change with age. In more than 11,000 recordings from the National Sleep Research Resource, N3 was 31.6 % of sleep in a children's cohort and 11.1 % in an older-adult cohort (Purcell and colleagues, 2017).
What is slow-wave sleep?
Slow-wave sleep is stage N3, the deepest stage of NREM sleep, defined by delta waves. By the scoring rules of the American Academy of Sleep Medicine, as summarised in a 2025 clinical review, an epoch counts as N3 when high-amplitude waves of 0.5–2 Hz and at least 75 µV fill at least 20 % of it. Slow-wave sleep and deep sleep are two names for the same stage.
The definition is a rule for scoring, so the numbers depend on which rule is used. When the same 72 recordings were scored by the AASM rules instead of the 1968 rules, they showed about 9.1 minutes more N3, 10.6 minutes more N1 and 20.5 minutes less N2 (Moser and colleagues, 2009). A deep-sleep figure from one lab or device is therefore not directly comparable with a figure scored another way.
Why delta waves grow after a long day: sleep pressure
The longer you stay awake, the stronger the slow waves of the next sleep. In a classic experiment, 8 young adults stayed awake for 40.5 hours; in the first recovery night their delta power was significantly above baseline. In normal nights, delta and theta power fell from one sleep cycle to the next (Borbély and colleagues, 1981).
The authors read low-frequency EEG power as a marker of a process that declines during sleep and whose starting level depends on how long you were awake before. Alexander Borbély built this into the two-process model of sleep regulation (1982): a homeostatic Process S, often called sleep pressure, interacts with the circadian Process C, the body clock. A 2016 reappraisal by Borbély and colleagues keeps the same two parts.
Process S explains the shape of a normal night. Pressure is highest at bedtime and declines during sleep, which fits the pattern of most deep sleep in the first cycles and more N2 and REM later. In this model, low-frequency EEG power serves as the marker of how much sleep pressure has built up.
How delta waves and deep sleep change with age
The share of deep sleep falls with age from childhood on. A meta-analysis of 65 studies covering 3,577 healthy people aged 5–102 found that the percentage of slow-wave sleep fell significantly with age in children and adolescents and again in adults. In adults, total sleep, sleep efficiency and REM share also fell, while N1, N2 and waking after sleep onset rose. After 60 the only change that stayed significant was less efficient sleep (Ohayon and colleagues, 2004). The abstract reports directions rather than exact percentages, and effect sizes depended on how strictly the participants were screened for illness.
Young children have more slow activity even when awake. A baby's main resting rhythm at the back of the head first appears at 3–4 months and runs at a mean 5.3 Hz up to age one, 6.8 Hz at 2–3 years, 7.9 Hz at 4–5 years and 8.7 Hz at 6–7 years, against 9.9 Hz in adults aged 16–50 (Alcala-Zermeno and colleagues, 2025). In 80 children aged 8–12, delta and theta decreased and alpha and beta increased with age (Clarke and colleagues, 2001).
In infancy and childhood, high-amplitude bursts of delta and theta at the transitions between sleep and waking, called hypnagogic hypersynchrony, are normal. In an awake adult, delta outside sleep is read very differently, as the last section explains, so the same EEG feature means different things at different ages.
Do delta waves help memory?
Experiments that changed slow waves during sleep also changed next-morning memory for word pairs. In a German study, weak oscillating currents at 0.75 Hz applied through the scalp during early NREM sleep increased slow-wave sleep and slow oscillations and improved retention of word pairs; stimulation at 5 Hz reduced slow oscillations and left memory unchanged (Marshall and colleagues, 2006).
Sound can do something similar, but only with precise timing. When short pink-noise clicks were played in phase with the up states of each slow oscillation, the slow oscillation, the spindle activity coupled to it and word-pair memory all grew; clicks played out of phase did nothing (Ngo and colleagues, 2013). In 13 adults aged 60–84, clicks timed by a real-time algorithm raised slow waves only during the stimulation blocks, not over the whole night, and overnight recall improved more than after a sham night, in step with the slow-wave increase (Papalambros and colleagues, 2017).
Sleep spindles, the faster bursts of N2, are part of the same story. After intensive word-pair learning, people produced more spindles than after a matched task without learning, most of all in the first 90 minutes of sleep, and spindle density correlated with recall (r = 0.56; Gais and colleagues, 2002). These are laboratory studies with small samples and computer-controlled stimulation; none of them used a continuous sound track.
Do slow waves wash the brain?
Slow waves are linked to pulses of brain fluid, but their cleaning role is not proven. In sleeping volunteers measured with fast brain imaging and EEG, each slow brain wave was followed by a change in blood flow and then a pulse of cerebrospinal fluid through the brain (Fultz and colleagues, Science, 2019).
The study showed coupling: neural slow waves, then blood-flow oscillations, then waves of fluid flow, in a regular sequence during NREM sleep. It did not measure waste removal, so headlines saying that delta waves "wash toxins out of the brain" go beyond what it found. The finding is a reason why researchers now study slow-wave sleep together with fluid flow, not proof of a cleaning effect.
Does delta waves music work for sleep?
No study has shown that music or tracks labelled "delta" move the brain into the delta band or deepen sleep. Many apps and videos sell delta music for sleep; the label describes a marketing idea, not a measured effect. What research does show is that the brain locks onto music in a different way: its slow rhythms follow the beat and the note rate, roughly 1–8 Hz, set by the tempo of the music (Doelling and Poeppel, 2015; Nozaradan and colleagues, 2011). Calm music may well help you relax; the band name on it adds nothing measurable.
The only sound methods that raised slow waves in studies were the ones described above: single short clicks timed by a computer to each slow wave, in a lab. A continuous track cannot do that, because it does not know when your next slow wave starts.
Delta binaural beats are a separate question. A 2023 systematic review found 14 EEG studies of binaural beats: five found the brain followed the beat, eight did not, one was mixed (Ingendoh and colleagues, 2023). A 2019 meta-analysis found a medium effect of binaural beats on memory, attention, anxiety and pain, which shows people respond to them but not that the brain wave changed (Garcia-Argibay and colleagues, 2019). The sleep studies of delta beats, and a generator to try them, are on the page about binaural beats for sleep.
Delta waves while awake: when they are a warning sign
In an awake adult, delta waves are not a sign of deep calm. A 2025 clinical review states that "generalized delta activity (0.5–4 Hz) outside of slow-wave sleep (N3) and theta activity (4–8 Hz) outside of drowsiness and light sleep (N1 and N2) is considered abnormal in an adult EEG" (Alcala-Zermeno and colleagues, 2025).
The review lists encephalopathy as the most common setting, with causes including brain injury, neurodegenerative disease, toxic or metabolic encephalopathy, meningoencephalitis and sedative drugs. Focal delta slowing, limited to one area, points to a local problem; the review's example is a stroke-like episode. Rhythmic delta over the temporal lobe (TIRDA) is linked with temporal lobe epilepsy in over 40 % of cases.
So more delta is not always better. It is normal in deep sleep and in young children, and abnormal in a waking adult. Only a clinical EEG, read by a specialist who knows the person's age, state and medicines, can say which is which.
Questions people ask
What frequency are delta waves?
Delta waves run at about 0.5–4 Hz, meaning fewer than four waves per second. The international clinical glossary defines the band as 0.1 to under 4 Hz, in practice from 0.5 Hz. Studies draw the edges differently: a review of 184 studies found delta definitions starting anywhere from 0 to 2 Hz and ending between 3.5 and 6 Hz.
Are delta waves the same as deep sleep?
Not quite. Deep sleep, stage N3 or slow-wave sleep, is defined by delta waves: waves of 0.5–2 Hz and at least 75 µV filling at least 20 % of a 30-second epoch. But delta waves also appear in N2 as K-complexes, in young children, and in some brain disorders in awake adults.
What is the difference between delta and theta waves?
Delta is below 4 Hz and theta is 4 to under 8 Hz by the clinical glossary. In adults theta rises as you drift into sleep (N1) and appears in REM sleep, while delta dominates deep sleep (N3). A baby's main resting rhythm sits at about 5 Hz, in what would be the theta range for an adult.
How much deep sleep is normal?
In young and middle-aged adults, deep slow-wave sleep is typically about 20 % of the night, according to a 2025 clinical review. The share varies with age: in one large data set N3 was 31.6 % of sleep in a children's cohort and 11.1 % in an older-adult cohort. Figures also depend on the scoring rules used.
Why do I get less deep sleep as I get older?
A meta-analysis of 65 studies found that the share of slow-wave sleep falls with age from childhood on, while light sleep and waking during the night increase in adults. After 60 the main further change was less efficient sleep. The studies describe the trend; they do not single out one cause.
Can you increase delta waves?
The best-documented factor is time awake: after 40.5 hours without sleep, delta power in the recovery night was clearly higher than normal. In labs, slow waves were raised with weak currents at 0.75 Hz or with short clicks timed by a computer to each slow wave. No study has shown that a continuous track or app does the same.
Does delta waves music help you sleep?
No study has shown that music labelled "delta" puts the brain into the delta band. The brain follows the beat and note rate of music, set by its tempo, not by the label. Calm music may help you relax, but the sound studies that raised slow waves used precisely timed clicks, not continuous music.
Are delta waves while awake dangerous?
In an awake adult, generalised delta on a clinical EEG is considered abnormal and most often points to encephalopathy, for example from injury, infection, a metabolic problem or sedating drugs. It is something a neurologist interprets in context, not something to self-diagnose from an app. In young children slow activity is normal.
Do slow waves clean toxins from the brain?
A 2019 study in Science showed that each slow wave during NREM sleep was followed by a change in blood flow and then a pulse of cerebrospinal fluid. That is coupling between brain waves and fluid flow. The study did not measure waste clearance, so the cleaning claim is not proven.
- Kane N., Acharya J., Beniczky S., Caboclo L., Finnigan S., Kaplan P. W., Shibasaki H., Pressler R., van Putten M. J. A. M. A revised glossary of terms most commonly used by clinical electroencephalographers and updated proposal for the report format of the EEG findings. Revision 2017. Clinical Neurophysiology Practice, 2017, 2, 170–185.
- Newson J. J., Thiagarajan T. C. EEG frequency bands in psychiatric disorders: a review of resting state studies. Frontiers in Human Neuroscience, 2018, 12, 521.
- Buzsáki G., Anastassiou C. A., Koch C. The origin of extracellular fields and currents: EEG, ECoG, LFP and spikes. Nature Reviews Neuroscience, 2012, 13 (6), 407–420.
- Cohen M. X. Where does EEG come from and what does it mean? Trends in Neurosciences, 2017, 40 (4), 208–218.
- He B. J. Scale-free brain activity: past, present, and future. Trends in Cognitive Sciences, 2014, 18 (9), 480–487.
- Iber C., Ancoli-Israel S., Chesson A., Quan S. F. The AASM Manual for the Scoring of Sleep and Associated Events: Rules, Terminology and Technical Specifications. American Academy of Sleep Medicine, 2007.
- Alcala-Zermeno J. L., Katyal R., Frauscher B., Schomer D., Sperling M. R., Strowd R., Tatum W. O., Wirrell E., Beniczky S., Nascimento F. A. Seminar in Epileptology: Normal awake and sleep patterns, interictal abnormalities, and ictal patterns on scalp EEG. Epileptic Disorders, 2025, 27 (5), 803–866.
- Steiger A., Pawlowski M. Depression and sleep. International Journal of Molecular Sciences, 2019, 20 (3), 607.
- Purcell S. M., Manoach D. S., Demanuele C., et al. Characterizing sleep spindles in 11,630 individuals from the National Sleep Research Resource. Nature Communications, 2017, 8, 15930.
- Moser D., Anderer P., Gruber G., et al. Sleep classification according to AASM and Rechtschaffen & Kales: effects on sleep scoring parameters. Sleep, 2009, 32 (2), 139–149.
- Achermann P., Borbély A. A. Low-frequency (< 1 Hz) oscillations in the human sleep electroencephalogram. Neuroscience, 1997, 81 (1), 213–222.
- Borbély A. A., Baumann F., Brandeis D., Strauch I., Lehmann D. Sleep deprivation: effect on sleep stages and EEG power density in man. Electroencephalography and Clinical Neurophysiology, 1981, 51 (5), 483–495.
- Borbély A. A. A two process model of sleep regulation. Human Neurobiology, 1982, 1 (3), 195–204.
- Borbély A. A., Daan S., Wirz-Justice A., Deboer T. The two-process model of sleep regulation: a reappraisal. Journal of Sleep Research, 2016, 25 (2), 131–143.
- Ohayon M. M., Carskadon M. A., Guilleminault C., Vitiello M. V. Meta-analysis of quantitative sleep parameters from childhood to old age in healthy individuals: developing normative sleep values across the human lifespan. Sleep, 2004, 27 (7), 1255–1273.
- Clarke A. R., Barry R. J., McCarthy R., Selikowitz M. Age and sex effects in the EEG: development of the normal child. Clinical Neurophysiology, 2001, 112 (5), 806–814.
- Marshall L., Helgadóttir H., Mölle M., Born J. Boosting slow oscillations during sleep potentiates memory. Nature, 2006, 444 (7119), 610–613.
- Ngo H.-V. V., Martinetz T., Born J., Mölle M. Auditory closed-loop stimulation of the sleep slow oscillation enhances memory. Neuron, 2013, 78 (3), 545–553.
- Papalambros N. A., Santostasi G., Malkani R. G., Braun R., Weintraub S., Paller K. A., Zee P. C. Acoustic enhancement of sleep slow oscillations and concomitant memory improvement in older adults. Frontiers in Human Neuroscience, 2017, 11, 109.
- Gais S., Mölle M., Helms K., Born J. Learning-dependent increases in sleep spindle density. Journal of Neuroscience, 2002, 22 (15), 6830–6834.
- Fultz N. E., Bonmassar G., Setsompop K., Stickgold R. A., Rosen B. R., Polimeni J. R., Lewis L. D. Coupled electrophysiological, hemodynamic, and cerebrospinal fluid oscillations in human sleep. Science, 2019, 366 (6465), 628–631.
- Doelling K. B., Poeppel D. Cortical entrainment to music and its modulation by expertise. Proceedings of the National Academy of Sciences USA, 2015, 112 (45), E6233–E6242.
- Nozaradan S., Peretz I., Missal M., Mouraux A. Tagging the neuronal entrainment to beat and meter. Journal of Neuroscience, 2011, 31 (28), 10234–10240.
- Ingendoh R. M., Posny E. S., Heine A. Binaural beats to entrain the brain? A systematic review of the effects of binaural beat stimulation on brain oscillatory activity, and the implications for psychological research and intervention. PLoS ONE, 2023, 18 (5), e0286023.
- Garcia-Argibay M., Santed M. A., Reales J. M. Efficacy of binaural auditory beats in cognition, anxiety, and pain perception: a meta-analysis. Psychological Research, 2019, 83 (2), 357–372.
The page has no player. Its interactive charts are schematic, not real recordings: a map of the five bands (delta 0.5–4 Hz, theta 4–8, alpha 8–13, beta 14–30, gamma 30–80 Hz, after the IFCN glossary) with a moving trace for each, slowed four times and with relative heights; a schematic EEG of six states with the relative share of each band; and a schematic hypnogram of a typical night with 4–5 cycles of about 90 minutes, most deep sleep early and longer REM towards morning.
