Brain waves: the EEG bands from delta to gamma

What are brain waves?

Brain waves are rhythms in the brain's electrical activity, recorded through the scalp by an electroencephalogram, or EEG. Each electrode picks up the combined activity of very large numbers of neurons working in step, and the signal is tiny: in adults, the alpha rhythm is "mostly below 50 µV", fifty millionths of a volt (IFCN glossary).

A brain wave is described by its speed, in hertz: the number of "complete cycles of repetitive waves or complexes in 1 s". That definition comes from the glossary the International Federation of Clinical Neurophysiology (IFCN) keeps for clinical EEG reports, which also defines a band as a range "of EEG frequency in a spectrum for a given recording". A recording holds activity at many speeds at once. So a band name describes one part of the signal, at one place on the head, not the state of the whole brain.

What are the five brain wave bands?

From slowest to fastest, they are delta, theta, alpha, beta and gamma. The glossary sets delta from 0.1 to under 4 Hz, theta from 4 to under 8 Hz, alpha from 8 to 13 Hz and gamma from over 30 to 80 Hz. Its 2017 printing put beta at 14 to 30 Hz, which left the step from 13 to 14 Hz in no band at all, and a correction in 2019 moved beta's lower edge to just over 13 Hz (glossary correction). For delta, the glossary adds that "for practical purposes lower frequency limit is 0.5 Hz", because a routine EEG does not record slower changes.

The five brain wave bands

Band	Range	Most typically recorded
Delta	0.1 to under 4 Hz	deep sleep, at 0.5 to 2 Hz
Theta	4 to under 8 Hz	drowsiness; over the front midline during mental tasks
Alpha	8 to 13 Hz	awake with the eyes closed, over the back of the head
Beta	over 13 to 30 Hz	awake, over the front and centre of the head
Gamma	over 30 to 80 Hz	mostly by electrodes placed inside the skull

A band is a range of speeds, not a state of mind. Inside alpha's range the glossary describes two different rhythms. The alpha rhythm sits over the back of the head, and its entry reads: "Best seen with the eyes closed, during physical relaxation and relative mental inactivity. Blocked or attenuated by attention, especially visual, and mental effort." The mu rhythm, at 7 to 11 Hz, sits over the centre of the head and is blocked by movement, by readiness to move and even by the thought of moving. Sleep spindles, short bursts at 11 to 16 Hz, straddle alpha and beta, and they come in sleep, not in alert thought. Age changes the picture too: in the resting EEG of 80 children aged 8 to 12, the share of delta and theta fell with age while the share of alpha and beta rose (Clarke and colleagues).

Why do sources draw the bands differently?

Because the edges are conventions, chosen for a purpose, and different fields chose differently. A Carnegie Mellon study that recorded children and adults reading, with a single-channel consumer EEG headset, describes the spectrum as "typically clustered into 5 frequency bands": delta at 1 to 3 Hz, theta 4 to 7, alpha 8 to 11, beta 12 to 29 and gamma 30 to 100 Hz (Chang and colleagues). Its alpha band stops at 11 Hz, two hertz short of the glossary's.

The brainwave frequency lists that circulate online inherit the same disagreement, and one of them writes it down. Its notes record where each of its own sources draws the bands: one source puts alpha at 9 to 14 Hz, another at 9 to 13 Hz and five others at 8 to 12 Hz, while beta ends at 27, 30, 36, 38 or 40 Hz depending on whom it follows (US patent 8,298,078). The chart sets seven of these sets of edges side by side.

Download the band edges (CSV)

Where do brainwave frequency lists come from?

One long list, which gives almost every number from 0.1 Hz upwards its own effect, is printed in full in a United States patent. Patent 8,298,078, applied for by the gaming machine maker WMS Gaming and granted in 2012, describes a "wagering game machine with biofeedback-aware game presentation": a gaming machine that sends chosen frequencies to the player as sound, light or vibration. To show what it might play, the patent reprints the list, from 0.1 Hz, "Organ/muscle resonances", to 42,800 Hz, with its compiler's notes and a bibliography. It introduces the list with one sentence: "It is believed that many frequencies can have significant physiological effects when presented to a person, such as a wagering game player." Its closing paragraphs give the aim, a player "playing longer and enjoying the gaming experience to a greater degree" (US patent 8,298,078).

The bibliography shows what the list rests on. It names 82 sources by code. Six are research articles named with a title, four of them from the Journal of Neurotherapy, such as "Academic Performance Enhancement with Photic Stimulation and EDR Feedback". Six more give only a researcher and a year, among them "Hans Berger, German Psychiatrist, 1929", and two describe unpublished studies at second hand. The rest are websites, magazines, message-board posts, software presets, device manuals and lists made for electrical devices. Of one set of lines, the compiler notes that they are "intended to be generated through an electromagnetic means", not heard; the Rife frequencies guide covers where those come from.

What the list's sources are

Kind of source	Count	For example
Research articles named with a title	6	four from the Journal of Neurotherapy
Unpublished studies, described second hand	2	students trying to have an out-of-body experience
Researchers named with a year and no title	6	"Hans Berger, German Psychiatrist, 1929"
People named with no work or year	4	a name followed by the word "Researcher"
Message-board posts and emails	6	posts to a message board about brainwave software
Device manuals, code tables and software presets	8	the owner's manual of an electrical stimulator
Frequency lists made for electrical devices	5	the Crane List of Rife-style frequencies
Websites, magazines, newspapers, books and the like	45	a personal page on a free web host

Download the sources by kind (CSV)

A single line shows how the claims pile up. The entry for 10 Hz, the beat this page plays, runs to more than 20 effects credited to more than a dozen sources: "safest frequency, especially for hangover & jet lag", "universally beneficial", "activate kidneys", "Learning a foreign language", "Adrenal Stimulant", and, among them, the plain fact that 10 Hz is the "dominant alpha frequency". The list gives no measurement for any of these. One of them, the bibliography notes, comes from experiments that used "pulsing lights rather than sound". Others rest on studies named by little more than their authors' surnames, one of them with the compiler's note "need to research this one for more info."

The same list travels in translation, and changes on the way. A six-page Czech version circulates as a PDF with no author, date or sources named. Checked line by line against the patent's list, 99 of its 100 entries match a line there, often word for word in translation; the one that does not is 0.3 Hz, where "the inner clock stops". A few numbers shift slightly, and some claims grow. The 10 Hz line becomes the safest frequency for stomach complaints and for giving up nicotine. The 20.215 Hz line, which the English list credits to one source that "implies it could mimic the effects of" LSD, loses its hedge and gains a promise that it does so "aniž způsobuje návyk", without causing addiction. The Tibetan Buddhist chants at 4.5 Hz become the meditation of a Tibetan Buddhist, and the compiler's doubts drop out.

What do the lists get right?

Where a line repeats a measurement, it holds up. The band names, their order and their rough ranges match the glossary's. "Deep dreamless sleep" for delta agrees with the glossary's deep sleep, dominated by delta waves of 0.5 to 2 Hz, and "Dream/sleep spindles" at 10 to 14 Hz is close to the glossary's spindles at 11 to 16 Hz.

Several lines are not about the brain at all. The list's "Earth Resonance" at 7.83, 14.1 and 20.3 Hz, and its "Schumann Resonance (5th frequency of 7)" at 33 Hz, echo real resonances of the cavity between the ground and the ionosphere, which lightning sets ringing at about 7.8, 14.3, 20.8, 27.3 and 33.8 Hz (NASA Goddard). They are electromagnetic waves, not sounds; the Schumann resonance guide covers what is known about them. The "lowest note on a piano" at 27.5 Hz is plain arithmetic: the piano's lowest A is 440 Hz halved four times. And the "bottom limit of normal hearing" at 16 Hz is near the measured figure. Hearing thresholds are standardised down to 20 Hz; sound below that can still be heard if it is loud enough, but a pure tone stops sounding like a tone at around 20 Hz (Møller and Pedersen).

Which lines have no study behind them?

Most of them. PubMed, the index of biomedical research kept by the US National Library of Medicine, holds 328 records with binaural beats in the title or abstract. None of them mentions LSD. None mentions hair growth: the two that mention hair are about hearing. None mentions pyramids, telepathy, astral projection or out-of-body experiences, and none mentions the kidneys, the bladder or the spleen, though the list gives organs frequencies of their own.

Four lines and their sources

Line	What it says, and the source it is credited to
1.05 Hz	"Helps hair grow+get its color back", credited to a personal web page on brainwave entrainment, 1999
4.5 Hz	"Shamanic State Of Consciousness/Tibetan Buddhist Chants", credited to a name with no work, passed on in a 2001 message-board post
10 Hz	"safest frequency, especially for hangover & jet lag", credited to "Octaves and windows", a 1988 item with no author
20.215 Hz	"LSD-25" ("he implies it could mimic the effects of"), credited to a page titled "5,000 HZ Frequencies To Boost Growth Of Plants"

Where a study does exist, it does not say what the line says. The list's delta entry, credited to a preset in brainwave software, says delta lowers cortisol and raises DHEA and melatonin. One small study measured exactly those. Eight healthy adults listened to delta binaural beats every day for 60 days, with cortisol, DHEA and melatonin among eleven body measures taken before and after. The abstract reports a change in only two of the eleven, insulin-like growth factor 1 and dopamine, and both went down. Scores on a questionnaire about how anxious people tend to feel did fall. It was an uncontrolled pilot, and the authors called for "a larger, randomized and controlled trial" (Wahbeh and colleagues).

Some lines borrow a real finding and move it. Tibetan Buddhist practice does appear in EEG research: eight long-term practitioners trained in the Tibetan Nyingmapa and Kagyupa traditions, with 10,000 to 50,000 hours of practice behind them, showed sustained, high-amplitude gamma activity during compassion meditation (Lutz and colleagues). That is gamma, measured at 25 to 42 Hz in the meditators' own brains while they meditated. The list puts Tibetan chants at 4.5 Hz, in theta.

The list is sometimes candid about itself. Beside 9.41 Hz, "Pyramid frequency (outside)", the compiler admits to finding no good definition of the term and ends the note with one word: "Anybody?"

People do reach for binaural beats hoping for something like a drug. In the Global Drug Survey 2021, 5.3 per cent of 30,896 respondents had used binaural beats to experience altered states. Most used them "to relax or fall asleep" (72.2 per cent) or to change their mood (34.7 per cent), and 11.7 per cent "to get a similar effect to that of other drugs" (Barratt and colleagues). The survey records why people listen. It does not test whether any beat does what they hoped.

Can a sound change your brain waves?

A sound can make the hearing brain follow its rhythm; whether the rest of the brain follows is far less clear. Sounds repeated about 40 times a second produce a steady 40 Hz response that can be recorded from the scalp (Galambos and colleagues). A binaural beat can do the same, more faintly. With 108 Hz in one ear and 118 Hz in the other, you hear a beat at the difference, 10 Hz, which is made inside the head, since neither ear receives it. The pitch you hear comes from the two tones. The 10 Hz is not a pitch but a rate, the speed of the beat: played as a tone on its own, 10 Hz would sit well below the 20 Hz at which a pure tone stops sounding like one. A slow beat is heard as one sound moving across the head, and as the rate rises it becomes a flutter in loudness, then roughness. The cortex's responses followed beats from 3 to 60 Hz and were largest at 40 Hz (Ross and colleagues). The response to a 40 Hz binaural beat was clear with tones around 400 Hz, undetectable above 3 kHz, and smaller than when both tones went to the same ear (Schwarz and Taylor).

The step that matters most to a list is the fourth: that the whole brain falls into step with the beat and takes on its band. A review of 14 EEG studies of binaural beats found five in line with that idea, eight against it and one mixed, and none of the studies of beats in the beta range found it (Ingendoh and colleagues). Even when an EEG does show more activity at a beat's rate, that is a change in a recording. It does not show a change in how anyone feels, let alone the exact effect a list attaches to that number.

What do binaural beat studies show?

Small, mixed effects on some measures, from studies of uneven quality. A meta-analysis of 22 studies on memory, attention, how anxious people felt and how strongly they felt painful stimuli found a medium overall effect, g = 0.45 (Garcia-Argibay and colleagues). Another, of 15 studies on memory and attention, found a similar effect, g = 0.40, with conflicting results for theta and beta beats (Basu and Banerjee).

Reviews that grade the quality of the evidence are more cautious. One, of 16 trials on how painful people found what they were going through, judged most of them at high risk of bias and the evidence low to very low in quality (Shamsi and colleagues). Another gathered 13 randomised trials of theta beats, with 630 people; only four could be combined, and the evidence was of low certainty (Fatima and colleagues).

None of these studies tests a line from a frequency list. They compare a beat with silence, noise or another sound on broad measures such as memory or attention. Whether 4.5 Hz brings a shamanic state, or 1.05 Hz grows hair, is not something they set out to show, and nothing in them suggests it.

Is 10 Hz the safest frequency?

No study in these sources tests whether one beat is safer than another. The phrase comes from the list's 10 Hz line, which credits it to a 1988 item called "Octaves and windows", with no author named. Ten hertz is a natural rate for a beat because it sits inside alpha, the band of the relaxed, eyes-closed waking rhythm, in every one of the seven sets of edges in the chart. That makes it typical, not safe or special.

Where 10 Hz has been compared with another rate, it did not come out as neutral. Thirty healthy adults heard a 10 Hz and a 40 Hz binaural beat in separate sessions. Afterwards, a brain-computer interface classified their responses to target sounds less accurately after the 10 Hz beat and more accurately after the 40 Hz beat (Li and colleagues). One laboratory task says little about everyday listening, but it is a measured difference, which is more than the "safest frequency" line rests on.

What plays on this page?

The player plays the site's 10 Hz alpha binaural beat: a steady 108 Hz tone in your left ear and 118 Hz in your right, made live in your browser. Harmonance measured the master file of the nine-hour version of the same beat, window by window, 6,481 windows in all. Through the middle of the recording the two tones sit at 107.99999 and 117.99998 Hz, a difference of 9.99999 Hz, and no window strays more than 0.00065 Hz from its target. Each ear carries only its own tone, with the other ear's tone more than 80 dB below it. The 10 Hz is not a sound in the file: it exists only where the signals from your two ears meet.

Use headphones, so each ear hears its own tone, and keep the volume low. Listen if you are curious, notice what you notice, and stop whenever you like. The 10 Hz alpha binaural beat page plays it on its own, and the 10 Hz guide looks at the number itself. The binaural beats collection holds a session for each band, from the falling delta session to the gamma session, and Studio has a binaural beat builder for setting a pair of tones yourself. For how a beat is made, read the guides to binaural beats and to binaural beats and isochronic tones. For names sometimes added to either end of the map, the epsilon and lambda guides check them against the glossary.

Sources and limits

This guide sets the bands as the clinical glossary defines them beside the frequency lists that circulate online, traced to the list printed in US patent 8,298,078. The chart draws each source's band edges as that source writes them; where a source gives no gamma band, none is drawn, and bands that run past 40 Hz are cut at the chart's edge and labelled with their end. The list's sources are counted from the codes in its bibliography: Barbara Hero's four tables count as one source, as do the two issues of the Megabrain Report given on one line, and sorting the entries into kinds follows what each entry says, so a few could sit in more than one kind. The Czech copy was compared with the patent's list line by line; it is not linked, because it names no author, date or sources. The PubMed counts come from searches of titles and abstracts for binaural beats with each named word, so a study could exist outside PubMed, or use a word only in its full text. The 10 Hz measurement is Harmonance's own check of its master, not a figure from its recipe.

1. Kane and colleagues, A revised glossary of terms most commonly used by clinical electroencephalographers and updated proposal for the report format of the EEG findings, Clinical Neurophysiology Practice, 2017, and its correction to the beta band, 2019.
2. Chang, Nelson, Pant and Mostow, Toward exploiting EEG input in a reading tutor, International Journal of Artificial Intelligence in Education.
3. US patent 8,298,078 B2, Wagering game machine with biofeedback-aware game presentation, WMS Gaming, granted 2012.
4. Clarke, Barry, McCarthy and Selikowitz, Age and sex effects in the EEG: development of the normal child, Clinical Neurophysiology, 2001.
5. Simões, Pfaff and Freudenreich, Observation of Schumann resonances in the Earth's ionosphere, NASA Goddard Space Flight Center.
6. Møller and Pedersen, Hearing at low and infrasonic frequencies, Noise and Health, 2004.
7. Wahbeh, Calabrese and Zwickey, Binaural beat technology in humans: a pilot study to assess psychologic and physiologic effects, Journal of Alternative and Complementary Medicine, 2007.
8. Barratt and colleagues, Who uses digital drugs? An international survey of 'binaural beat' consumers, Drug and Alcohol Review, 2022.
9. Lutz, Greischar, Rawlings, Ricard and Davidson, Long-term meditators self-induce high-amplitude gamma synchrony during mental practice, PNAS, 2004.
10. Ross and colleagues, Human cortical responses to slow and fast binaural beats reveal multiple mechanisms of binaural hearing, Journal of Neurophysiology, 2014.
11. Galambos, Makeig and Talmachoff, A 40-Hz auditory potential recorded from the human scalp, PNAS, 1981.
12. Schwarz and Taylor, Human auditory steady state responses to binaural and monaural beats, Clinical Neurophysiology, 2005.
13. Ingendoh, Posny and Heine, 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.
14. Garcia-Argibay, Santed and Reales, a meta-analysis of binaural beats and cognition, Psychological Research, 2019.
15. Basu and Banerjee, Potential of binaural beats intervention for improving memory and attention: insights from meta-analysis and systematic review, Psychological Research, 2023.
16. Shamsi, Azadinia and Shaygan, a systematic review of binaural beats and how painful people find their experience, BMC Complementary Medicine and Therapies, 2024.
17. Fatima and colleagues, a systematic review and meta-analysis of randomised trials of theta binaural beats, Explore, 2026.
18. Li and colleagues, Gamma-band binaural beats neuromodulation enhances P300 classification in an auditory brain-computer interface paradigm, IEEE Transactions on Neural Systems and Rehabilitation Engineering, 2025.
19. The 10 Hz beat's measurement: Harmonance's check of its own master, window by window over nine hours.