- Gamma waves are brain rhythms faster than 30 Hz; the international clinical EEG glossary puts the top of the band at 80 Hz, while research papers often use 30–100 Hz.
- In 2009 two studies in Nature showed in mice that driving fast-spiking parvalbumin interneurons creates gamma rhythms and silencing them removes gamma.
- Most of the scalp EEG signal above 20 Hz comes from muscles: in a study of people recorded with and without paralysis, power above 20 Hz differed 10- to 200-fold.
- The human hearing brain follows sounds repeated about 40 times a second with an especially clear 40 Hz response, first described by Galambos and colleagues in 1981.
- 40 Hz flickering light reduced amyloid in Alzheimer's model mice at MIT in 2016, but a 2023 study at NYU found no effect on amyloid and no real gamma entrainment in two mouse models.
- No 40 Hz device is approved for Alzheimer's disease; a 6-month trial of 76 patients found no difference from sham on its main measure, and the 670-patient HOPE trial had not published results as of 1 October 2026.
What are gamma waves?
Gamma waves are the fastest of the five named brain-wave bands: electrical rhythms above 30 cycles per second. An EEG records tiny voltage changes from scalp electrodes, produced mainly by the synaptic currents of large groups of cortical neurons working in step. The signal always contains many frequencies at once; delta, theta, alpha, beta and gamma are agreed slices of one spectrum, not separate signals or states of mind.
Gamma is also the smallest slice. Brain recordings follow a 1/f-like pattern: power falls steadily as frequency rises, so slow rhythms are large and fast ones faint. The clinical EEG glossary notes that gamma is most commonly recorded with electrodes inside the skull rather than on the scalp. Gamma rhythms appear in many brain regions, awake and asleep, usually in short episodes rather than as a steady hum.
The interactive map above shows the five bands as moving traces slowed down four times, with relative heights; click gamma to see when it is strongest and where it shows. The six schematic EEG states below it show each band's share from relaxed wakefulness to REM sleep. All of it is schematic, not a real recording.
Gamma wave frequency: where does the band start and end?
Gamma starts above 30 Hz in almost every definition, but its upper edge varies. The International Federation of Clinical Neurophysiology (IFCN) glossary, revised in 2017, defines gamma as above 30 Hz up to 80 Hz, and the map on this page follows it. Research papers use 30–100 Hz, 30–80 Hz or even 20–80 Hz.
This is not unique to gamma. A 2018 review of 184 resting-EEG studies found that the most common definition of each band was used by only 30–50 % of the studies, so what one paper calls gamma is not necessarily what another means.
| Source | Gamma range |
|---|---|
| IFCN clinical glossary (Kane and colleagues, 2017) | Above 30 to 80 Hz |
| Most common definition in 184 studies (Newson and Thiagarajan, 2018) | 30–40 Hz |
| All definitions found in those 184 studies | Anywhere from 20 to 100 Hz |
| Optogenetic study of parvalbumin neurons (Sohal and colleagues, 2009) | 30–80 Hz |
| Optogenetic study of fast-spiking cells (Cardin and colleagues, 2009) | 20–80 Hz |
The frequency you will meet most often is 40 Hz, near the bottom of the band, because it is the rate at which the human brain follows a rhythmic sound most readily. For comparison, the glossary places delta below 4 Hz, theta at 4–8 Hz, alpha at 8–13 Hz and beta at 14–30 Hz; gamma begins where beta ends.
What causes gamma waves in the brain?
Gamma rhythms come from a loop between excitatory neurons and fast inhibitory neurons. A 2012 review by Buzsáki and Wang ties gamma generation to inhibition delivered onto the cell bodies of neurons, and states plainly that its functions and mechanisms remain a matter of debate.
The most direct evidence came in 2009, when two papers in the same issue of Nature used optogenetics, switching chosen neurons on and off with light, in mice. Cardin, Moore and colleagues drove fast-spiking interneurons in the whisker area of the cortex at 8 to 200 Hz, which selectively amplified gamma; driving excitatory pyramidal neurons amplified only lower frequencies. The authors called it the first causal evidence that fast-spiking cells generate gamma.
Sohal, Deisseroth and colleagues showed the reverse side. Silencing parvalbumin interneurons, the main fast-spiking type, suppressed gamma; driving them generated a gamma-frequency rhythm. Input arriving in a gamma pattern was passed on through the circuit with less noise.
Cardin's study added one detail: the timing of a sensory input within the gamma cycle set how large and precise the cortical response was.
What do gamma waves do? The binding hypothesis
The best-known idea about gamma is that it helps the brain bind separate features into one percept. In 1989 Wolf Singer's group in Frankfurt recorded neurons in the visual cortex of cats and found that they fired in 40–60 Hz rhythms. Neurons in separate columns of cortex fell into step with each other when they responded to the same object. The authors proposed that synchronised oscillations could be how the brain links the edges, colours and movement of one object across different parts of the visual field.
Fries developed a broader version in 2009: gamma synchronisation is found in many cortical areas and tasks, and he argued it is a basic operation of cortical computation, a way of selecting which inputs get through.
A related framework links gamma to alpha. Jensen and Mazaheri proposed in 2010 that alpha works as pulsed inhibition of brain areas not needed for a task, while in the areas doing the work gamma rises and alpha falls. On this view gamma marks the active, engaged patch of cortex and alpha the parts held back.
All three are hypotheses with real data behind them, mostly from electrodes in or on the brain. They do not support the popular idea that a person can be 'in gamma' as a mental mode, or that more gamma means a smarter mind.
Gamma waves and meditation: what did the monks study show?
The most quoted human gamma study comes from the University of Wisconsin. In 2004 Lutz, Davidson and colleagues recorded the EEG of eight long-term Buddhist practitioners from Tibetan traditions, with 10,000 to 50,000 hours of practice over 15 to 40 years, and ten students with no meditation experience. During a 'non-referential compassion' meditation, the practitioners produced sustained, high-amplitude gamma oscillations that were synchronised across distant parts of the scalp.
The ratio of gamma (25–42 Hz) to slow activity (4–13 Hz) was already higher in the practitioners at rest, before they started meditating. It rose sharply during meditation over most of the scalp and stayed higher afterwards. The beginners showed nothing comparable.
Three limits matter: only eight practitioners; they were much older than the controls (mean age 49 against 21); and the study shows a difference between groups, not that meditation caused it. Scalp gamma is also vulnerable to muscle signals, as the next section explains.
The wider meditation literature does not show a gamma signature. A 2015 systematic review of 56 EEG studies of mindfulness with 1,715 participants found that mindfulness most often went with more alpha and theta power, and found no consistent pattern for gamma. The theta and alpha pages on this site cover those findings.
Why are gamma waves so hard to measure on an EEG?
Gamma is the hardest rhythm to measure from the scalp because muscles and eyes produce electrical signals in the same frequency range, much larger than the brain's own gamma. Even slight tension in the forehead, jaw or neck adds fast activity to the recording.
A 2007 study measured the problem. Whitham and colleagues recorded the EEG of two people with and without complete paralysis of the muscles. Power above 20 Hz differed between the two states by a factor of 10 to 200. Their conclusion: most scalp EEG above 20 Hz is of muscle origin, and earlier scalp gamma studies need re-evaluation.
The eyes add a second problem. In 2008 Yuval-Greenberg, Deouell and colleagues combined EEG with eye tracking and showed that a widely reported gamma burst 200–300 ms after a picture appears was time-locked to tiny involuntary eye movements (microsaccades) and came from the eye muscles. They stressed that real neuronal gamma has been shown conclusively with other methods; the problem is this particular scalp signal.
Consumer EEG headbands sit on the forehead and behind the ears, close to those muscles, and a 2017 comparison found them more prone to blink and muscle artefacts than medical systems. Treat any headband or app that reports your 'gamma' with caution: a clenched jaw can look like a burst of insight.
Why 40 Hz? The brain's response to rhythmic sound and light
When a sound is repeated about 40 times a second, the hearing brain answers with a single, stable 40 Hz wave. Galambos, Makeig and Talmachoff described this 40 Hz auditory potential from the human scalp in 1981. It appears even at sound levels close to the threshold of hearing, which made it attractive for hearing tests.
Later work showed that such steady-state responses can be recorded to sounds repeated anywhere from 1 to 200 times a second, but a 2003 review by Picton and colleagues notes that in awake people they are particularly prominent near 40 Hz. Audiologists use them to estimate hearing thresholds objectively, and anaesthetists to monitor anaesthesia.
Vision behaves in a similar way. In a 2001 study by Herrmann, ten people watched light flickering at every rate from 1 to 100 Hz. The visual cortex followed the flicker up to at least 90 Hz, with the strongest responses near 10, 20, 40 and 80 Hz.
A steady-state response follows the stimulus while it lasts. Whether it changes anything beyond that, such as memory or disease, is a separate question that needs its own evidence.
Can sound boost gamma waves? Binaural beats and entrainment
A sound that pulses 40 times a second produces a 40 Hz response in the auditory cortex; that much is established. A binaural beat, where each ear hears a slightly different tone, gives a weaker response: a 2005 study found that a 40 Hz binaural beat evoked a 40 Hz brain response only with low tones, around 400 Hz, and that it was smaller than the response to the same beat physically present in the sound.
Across all frequencies, evidence that binaural beats change the EEG is inconsistent: a 2023 systematic review found five EEG studies supporting entrainment, eight contradicting it and one mixed. A 2019 meta-analysis of 22 studies found a medium effect on memory, attention, anxiety and pain, which shows that people respond to the beats, not that their brain waves changed.
The site's page on 40 Hz binaural beats covers the sound itself: which tones to use, the attention studies and how to listen. This page stays with the brain.
40 Hz and Alzheimer's: the MIT mouse studies and the failed replication
The interest in 40 Hz comes from Li-Huei Tsai's laboratory at the Massachusetts Institute of Technology. In 2016 her group reported in Nature that in a mouse model of Alzheimer's disease, gamma was reduced before amyloid plaques appeared. Driving parvalbumin interneurons at 40 Hz, and not at other frequencies, lowered amyloid-β and recruited microglia, the brain's immune cells. A non-invasive 40 Hz flickering light lowered amyloid-β in the visual cortex of young mice and reduced plaque load in older ones. The paper has received author corrections in 2018 and 2024.
In 2019 the group added sound: seven days of 40 Hz tones improved memory and reduced amyloid in the auditory cortex and hippocampus of 5XFAD mice, and sound plus light, but neither alone, reduced plaques across wide areas of the neocortex. A 2024 paper linked the effect to increased flow of cerebrospinal fluid through the glymphatic system; blocking that system abolished the amyloid removal.
In 2023 a team led by György Buzsáki at New York University repeated the light experiment in two mouse models with electrodes in the visual cortex, entorhinal cortex and hippocampus. The flicker did not engage the brain's native gamma oscillations, and there was no reliable change in plaques, microglia or amyloid-β. The mice also avoided the flickering light. The mouse evidence is split between two respected laboratories, and the question is open.
40 Hz light and sound in people: what have the trials found?
The first human results, published in 2022 by Chan, Li-Huei Tsai and colleagues, came in two stages. Single sessions in 43 people, including 2 epilepsy patients with brain electrodes, showed that 40 Hz light and sound were safe and that the rhythm reached the hippocampus and other deep structures. In the pilot, 15 people with mild probable Alzheimer's disease used 40 Hz light and sound daily for three months in a single-blind, randomised, placebo-controlled design. The 40 Hz group showed less hippocampal shrinkage and ventricular enlargement, better delayed face-name recall and more regular daily activity rhythms. With 15 people and exploratory outcomes, it is a pilot.
The OVERTURE trial, published in 2024, tested Cognito Therapeutics' device in 76 people with mild to moderate Alzheimer's disease: double-blind, two to one active or sham, one hour a day at home for six months. It was safe, with no signs of brain swelling on scans, and adherence was 85–90 %. There was no difference from sham on the primary outcome or on two standard cognitive and clinical scales. Some secondary measures, including daily-living activities and whole-brain volume on MRI, showed less decline with nominal significance. The trial was not designed to prove efficacy, so the company's larger percentage figures from those secondary measures are not a result.
The pivotal HOPE trial of the same company's device finished enrolling 670 participants at about 70 US sites in 2025, with daily home use for 12 months against a sham device. The company describes the device as investigational and not for sale. As of 1 October 2026 no topline results had been published. No 40 Hz device is approved for Alzheimer's disease.
One safety point applies to any flickering light, including 40 Hz. Flicker can trigger seizures in people with photosensitive epilepsy. Rates of 15–25 Hz are the most provocative, but anything from 1 to 65 Hz can do it, and about 1 in 4,000 young people aged 5–24 is at risk. Do not use flickering light if you or a family member has epilepsy or has had a seizure.
Gamma waves benefits: what is shown and what is not
Lists of gamma-wave benefits mix animal studies, single small studies and marketing. Sorted by what was actually tested, they look like this.
| Claim | What the evidence says |
|---|---|
| Fast inhibitory neurons generate gamma | Shown causally in mice with optogenetics (two Nature studies, 2009) |
| Gamma binds features into one percept | A hypothesis since 1989, supported by animal recordings, still debated |
| Long-term meditators show strong gamma | One study of 8 practitioners (2004); not consistent across meditation research |
| The brain follows a 40 Hz sound or flicker | Measured since 1981 for sound and in flicker studies; used in hearing tests |
| Binaural beats raise gamma | Weaker response than a pulsed sound; entrainment evidence inconsistent |
| 40 Hz clears amyloid | Shown in MIT mouse studies, not reproduced by an NYU group in 2023 |
| 40 Hz treats Alzheimer's disease in people | Not shown; a 76-patient trial missed its main measure, the pivotal trial has not reported |
| A headband can measure your gamma | Scalp gamma is dominated by muscle and eye signals; treat headband readings with caution |
| More gamma means higher intelligence or insight | No evidence in the sources checked for this page |
The calm summary: gamma rhythms are real, they are made by identifiable cells, and they are one of the most active topics in neuroscience. What they are for is still debated, and the jump from a mouse cortex to a listening track or a headband reading is much longer than marketing suggests.
Questions people ask
What are gamma waves?
Gamma waves are the fastest rhythms in a brain recording, above 30 Hz. The clinical EEG glossary puts the top of the band at 80 Hz; research papers often use 30–100 Hz. They are produced by a loop between excitatory neurons and fast inhibitory neurons.
What frequency are gamma waves?
Above 30 Hz, up to 80 Hz by the international clinical definition and up to 100 Hz in many research papers. The most studied single frequency is 40 Hz.
What do gamma waves do?
Nobody knows for sure. Leading hypotheses say gamma helps bind features of an object into one percept and helps select which signals get through a network. Experts on gamma mechanisms describe its function as still a matter of debate.
Why is 40 Hz special?
It is the rate at which the hearing brain follows a rhythmic sound most clearly, a response described in 1981 and used in hearing tests. It became famous after MIT mouse studies of 40 Hz light and sound.
Do 40 Hz light and sound help with Alzheimer's disease?
It has not been shown in people. MIT studies in mice found less amyloid, but an NYU team could not reproduce the light result in 2023. A 6-month trial of 76 patients found no difference from sham on its main measure, and the 670-patient HOPE trial had not published results as of 1 October 2026.
Do monks have more gamma waves?
In one 2004 study, eight long-term Buddhist practitioners showed strong, synchronised gamma during compassion meditation and higher gamma at rest than ten beginners. The groups were small and differed in age, and a review of 56 mindfulness studies found no consistent gamma pattern.
Can I measure my gamma waves with a headband?
Not reliably. Most of the scalp signal above 20 Hz comes from muscles, and tiny eye movements create gamma-like bursts. Consumer headbands sit near the jaw and forehead muscles and pick up these artefacts easily.
Can binaural beats increase gamma waves?
A 40 Hz binaural beat produces a 40 Hz response with low tones, but weaker than a sound that physically pulses at 40 Hz, and EEG entrainment by binaural beats is inconsistent across studies. The 40 Hz binaural beats page covers the sound.
Is 40 Hz flickering light safe?
It was well tolerated in the small human trials. But flicker from 1 to 65 Hz can trigger seizures in photosensitive epilepsy; avoid it if you or a family member has epilepsy or has had a seizure.
Are gamma waves present during sleep?
Yes. Gamma rhythms occur in many brain regions both in waking and in sleep, usually in short episodes. On a scalp EEG they are small compared with the slow waves of sleep.
- 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.
- He B. J. Scale-free brain activity: past, present, and future. Trends in Cognitive Sciences, 2014, 18 (9), 480–487.
- Buzsáki G., Wang X.-J. Mechanisms of gamma oscillations. Annual Review of Neuroscience, 2012, 35, 203–225.
- Cardin J. A., Carlén M., Meletis K., Knoblich U., Zhang F., Deisseroth K., Tsai L.-H., Moore C. I. Driving fast-spiking cells induces gamma rhythm and controls sensory responses. Nature, 2009, 459 (7247), 663–667.
- Sohal V. S., Zhang F., Yizhar O., Deisseroth K. Parvalbumin neurons and gamma rhythms enhance cortical circuit performance. Nature, 2009, 459 (7247), 698–702.
- Gray C. M., König P., Engel A. K., Singer W. Oscillatory responses in cat visual cortex exhibit inter-columnar synchronization which reflects global stimulus properties. Nature, 1989, 338 (6213), 334–337.
- Fries P. Neuronal gamma-band synchronization as a fundamental process in cortical computation. Annual Review of Neuroscience, 2009, 32, 209–224.
- Jensen O., Mazaheri A. Shaping functional architecture by oscillatory alpha activity: gating by inhibition. Frontiers in Human Neuroscience, 2010, 4, 186.
- Lutz A., Greischar L. L., Rawlings N. B., Ricard M., Davidson R. J. Long-term meditators self-induce high-amplitude gamma synchrony during mental practice. Proceedings of the National Academy of Sciences USA, 2004, 101 (46), 16369–16373.
- Lomas T., Ivtzan I., Fu C. H. A systematic review of the neurophysiology of mindfulness on EEG oscillations. Neuroscience and Biobehavioral Reviews, 2015, 57, 401–410.
- Whitham E. M., Pope K. J., Fitzgibbon S. P., et al. Scalp electrical recording during paralysis: quantitative evidence that EEG frequencies above 20 Hz are contaminated by EMG. Clinical Neurophysiology, 2007, 118 (8), 1877–1888.
- Yuval-Greenberg S., Tomer O., Keren A. S., Nelken I., Deouell L. Y. Transient induced gamma-band response in EEG as a manifestation of miniature saccades. Neuron, 2008, 58 (3), 429–441.
- Ratti E., Waninger S., Berka C., Ruffini G., Verma A. Comparison of medical and consumer wireless EEG systems for use in clinical trials. Frontiers in Human Neuroscience, 2017, 11, 398.
- Galambos R., Makeig S., Talmachoff P. J. A 40-Hz auditory potential recorded from the human scalp. Proceedings of the National Academy of Sciences USA, 1981, 78 (4), 2643–2647.
- Picton T. W., John M. S., Dimitrijevic A., Purcell D. Human auditory steady-state responses. International Journal of Audiology, 2003, 42 (4), 177–219.
- Herrmann C. S. Human EEG responses to 1–100 Hz flicker: resonance phenomena in visual cortex and their potential correlation to cognitive phenomena. Experimental Brain Research, 2001, 137 (3–4), 346–353.
- Schwarz D. W. F., Taylor P. Human auditory steady state responses to binaural and monaural beats. Clinical Neurophysiology, 2005, 116 (3), 658–668.
- 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.
- Iaccarino H. F., Singer A. C., Martorell A. J., et al., Tsai L.-H. Gamma frequency entrainment attenuates amyloid load and modifies microglia. Nature, 2016, 540 (7632), 230–235.
- Martorell A. J., Paulson A. L., Suk H.-J., et al., Tsai L.-H. Multi-sensory gamma stimulation ameliorates Alzheimer's-associated pathology and improves cognition. Cell, 2019, 177 (2), 256–271.e22.
- Murdock M. H., Yang C.-Y., Sun N., et al., Tsai L.-H. Multisensory gamma stimulation promotes glymphatic clearance of amyloid. Nature, 2024, 627 (8002), 149–156.
- Soula M., Martín-Ávila A., Zhang Y., Dhingra A., Nitzan N., Sadowski M. J., Gan W.-B., Buzsáki G. Forty-hertz light stimulation does not entrain native gamma oscillations in Alzheimer's disease model mice. Nature Neuroscience, 2023, 26 (4), 570–578.
- Chan D., Suk H.-J., Jackson B. L., et al., Tsai L.-H. Gamma frequency sensory stimulation in mild probable Alzheimer's dementia patients: results of feasibility and pilot studies. PLoS ONE, 2022, 17 (12), e0278412.
- Hajós M., Boasso A., Hempel E., et al., Malchano Z. Safety, tolerability, and efficacy estimate of evoked gamma oscillation in mild to moderate Alzheimer's disease. Frontiers in Neurology, 2024, 15, 1343588.
- Cognito Therapeutics. Press release on completed enrolment of the HOPE pivotal trial (NCT05637801). Business Wire, 1 July 2025.
- Fisher R. S., Harding G., Erba G., Barkley G. L., Wilkins A., Epilepsy Foundation of America Working Group. Photic- and pattern-induced seizures: a review for the Epilepsy Foundation of America Working Group. Epilepsia, 2005, 46 (9), 1426–1441.
This page has no sound. The chart draws a schematic trace for each of the five bands (delta 0.5–4 Hz, theta 4–8 Hz, alpha 8–13 Hz, beta 14–30 Hz, gamma 30–80 Hz, following the IFCN glossary), slowed down four times, with relative heights, and a schematic EEG of six states with the relative share of each band; none of it is a real recording.