There is a tab open in your browser right now, or there was last week. Ten hours of something called Deep Focus Gamma Waves, forty-something million views, comments below it from people saying it saved their dissertation. You put it on at two in the afternoon when the document needed real thinking and your attention had the structural integrity of wet paper.
The question worth asking isn’t whether binaural beats are real. They are — your brain genuinely manufactures a rhythm that doesn’t exist in the air, and EEG picks it up. The question is whether the thing in your headphones has any meaningful relationship to the thing that worked in a laboratory.
It mostly doesn’t. And when researchers finally tested the version people actually use, at the scale people actually use it, the results went the wrong way.
Educational, not medical advice. This is low-risk, but if you have a seizure disorder, a hearing condition, or a psychiatric diagnosis, talk to a clinician before using audio brain stimulation.
The phantom tone
Play a 340 Hz tone into your left ear and a 380 Hz tone into your right. There is no 40 Hz sound anywhere in the room. Yet you perceive a 40 Hz pulse, because your auditory system constructs it centrally.
The pathway is well mapped. Each tone produces a phase-preserved signal converging on the medial superior olivary nucleus in the brainstem — the first relay receiving input from both ears, and where beat perception is born. From there the combined signal travels to the inferior colliculus and spreads cortically. The theorised payoff is entrainment: cortical oscillations falling into step with the perceived beat, the way metronomes on a shared table gradually sync.
That much is not in dispute. It shows up on EEG.
Here’s the catch that matters for focus specifically. Gamma beats around 40 Hz are intrinsically faint. Beat salience is weaker at 40 Hz than at low frequencies, and higher EEG bands carry less power to begin with, so the gamma signal is more easily drowned by noise. The band most associated with sharp cognition is the hardest one to drive with sound.
What actually worked, and under what conditions
The most rigorous test of this question is a 2025 preregistered, randomised, within-subjects crossover trial with an active pure-tone control and EEG validation — 80 participants randomised, 64 analysed, across sixteen conditions, running a 33-minute continuous performance task.
The EEG confirmed the mechanism cleanly. Power rose at 16 Hz for beta beats (p = 0.001) and at 40 Hz for gamma beats (p < 0.001), more strongly for gamma. The entrainment is physically real.
Then the behaviour.
Binaural beats did not reduce the vigilance decrement — the drift in attention over a long task, which is the thing you were hoping to fix (Treatment × Block p = 0.751 and 0.540). What did improve was overall target detection, and only under specific conditions:
- 40 Hz gamma on a 340 Hz carrier tone: p = 0.002. On a 400 Hz carrier: nothing.
- With white noise present: p = 0.033. Without it: nothing.
Read those two lines again, because they are the most useful thing in this article. Change the carrier tone by sixty hertz and the effect disappears. Remove the background masking and the effect disappears.
One more detail from that trial, and it cuts against the easy dismissal: self-reported perceived performance did not differ from control on any measure (all p > 0.08). Participants couldn’t tell. Whatever was happening, it wasn’t expectancy.
So there is a real effect. It is narrow, it is about general target detection rather than sustained vigilance, and it is fragile to parameters that no streaming track controls.
Which brings us to the streaming tracks.
A thousand people, at home
In 2023 a team at Adam Mickiewicz University in Poznań did something the field had not done: they tested binaural beats the way people use them.
Their reasoning is stated plainly in the paper. Laboratory exposure to binaural beats is typically very short — seconds to a few minutes — and incidental, usually a single visit. The recordings people stream run from several dozen minutes to a dozen hours, daily, at home, on whatever headphones they own. These are, the authors argue, completely different conditions. Nobody had checked whether the second one works.
They tested 1,000 people at home on a two-part fluid intelligence task built from a matrix-reasoning bank designed to imitate Raven’s Progressive Matrices, comparing beats against silence and against other sounds.
Listening to binaural beats was not neutral. It dramatically deteriorated performance.
And the part that makes this hard to wave away: the deterioration was not belief-mediated. The researchers measured participants’ Theories of Intelligence and Need for Cognition and found no relationship with the effect (r = 0.032, p = 0.536; r = 0.050, p = 0.334). This wasn’t a placebo running backwards. Something about the actual stimulus, under actual home conditions, made people worse at a reasoning task.
The funding is clean, for what it’s worth — European COST and Horizon 2020 grants, an academic scholarship, no industry money anywhere near it.
One honest limitation, which I’d rather state than bury. The frequency used in that experiment is recorded in my notes as 15 Hz — beta, not gamma. I could not verify this against the full text. If it’s correct, then strictly this is evidence about home-use binaural beats in general and not a direct test of 40 Hz. It does not prove that a gamma track makes you worse. What it does establish is that the leap from entrainment works in principle to therefore the ten-hour track helps you is unwarranted, and can run in the opposite direction from what everyone assumes.
Why entrainment isn’t enough
The seductive myth here is that measurable entrainment guarantees a cognitive benefit. It does not, and there are two separate reasons.
The evidence on entrainment itself is split. A 2023 systematic review of fourteen EEG studies found five supporting the entrainment hypothesis, eight contradicting it, and one mixed — with enough methodological heterogeneity that the authors declined to reach a verdict. A 2017 study looking for EEG changes across theta, alpha, beta and gamma found no significant change in any band, and no change in heart rate or skin conductance either.
And even where entrainment happens, it can diverge from benefit. In the 2025 trial, adding white noise improved attention while weakening the 40 Hz entrainment signal. The measurable brainwave effect and the useful cognitive effect moved in opposite directions. Entrainment is necessary for the theory. It is nowhere near sufficient for the result.
There’s a plausible anatomical reason. Sustained attention is governed largely by tonic arousal from the locus coeruleus, the brainstem’s norepinephrine hub. The leading interpretation of the 2025 data is that binaural beats may entrain cortical oscillations while leaving deep subcortical arousal structures untouched. If that’s right, it explains the whole pattern: you can measure the beat in the cortex and still not move the system that actually holds your attention on a document for two hours.
Worth noting: this directly contradicts the most widely circulated popular account. Huberman Lab describes 40 Hz beats as working by “integrating within deep brain centers” and increasing striatal dopamine — hedged elsewhere as “possibly by impacting dopamine and acetylcholine levels,” which is the more defensible framing. Direct receptor-level evidence for binaural beats specifically: none that I could find. The cortical-only interpretation and the deep-brain interpretation cannot both be right, and only one of them comes from the trial with the EEG validation and the active control.
The confusion that does the most damage
If you have read that 40 Hz “clears amyloid,” “boosts BDNF,” or “modifies microglia,” those findings are real. They do not belong to binaural beats.
They come from GENUS — Gamma ENtrainment Using Sensory stimulation — the MIT line of work that drives gamma using flickering light and amplitude-modulated clicking sound, not a phantom tone. In that separate literature, 40 Hz sensory stimulation reduced amyloid and tau, decreased synapse loss, and recruited microglia and astrocytes in Alzheimer’s mouse models, with the amyloid effect abolished by a GABA-A antagonist.
Mouse models. Flickering lights. A different delivery mechanism entirely.
For 40 Hz binaural beats and BDNF, NGF, myelination, astrocyte mitochondria or blood-brain-barrier function, the honest entry is: no data. A strobing LED box and a YouTube tone are not interchangeable, and anyone blurring them is selling you a mouse study about Alzheimer’s pathology as a study hack.
This is the same move we keep running into — with krill oil, where a brain-uptake finding for one molecular form gets marketed for a different one. The pattern is always the same: real research, wrong product.
What survives
Strip out the overreach and something small is left standing.
A 2019 meta-analysis of 22 studies and 35 effect sizes found an overall medium, significant, reasonably consistent effect (g = 0.45) across cognition, anxiety and pain — with longer and pre-task exposure more effective than short during-task exposure. A 2023 meta-analysis of fifteen studies found a medium pooled effect for memory and attention, while its own systematic-review layer noted conflicting results, particularly for theta and beta.
A 2017 study found that 40 Hz beats on a 340 Hz carrier narrowed the attentional spotlight — participants showed a considerably smaller global-precedence effect, meaning they attended more to local detail. That’s a real, specific change, and notice that it is not “better focus.” It’s a different distribution of attention, useful for proofreading and possibly counterproductive for spotting the shape of an argument.
And a 2020 MEG-validated study found something odd: no gain within a session, but improved learning between sessions, with the 40 Hz group consolidating more. An effect on learning rather than on acute focus.
Against those sit the nulls: a 2023 controlled study finding no significant effect of gamma beats on working memory, directly checking a 2022 study that found one. And the vigilance-decrement null from the best-designed trial in the set.
What I’d actually do
If you want to use this, use the version that was tested, not the version that autoplays.
40 Hz beats on a low carrier tone, around 340 Hz, with white noise present, started before the task rather than during it, for fifteen to thirty minutes. That’s the combination with actual p-values behind it. Most streaming tracks use very low carriers, skip the masking entirely, and run for ten hours — which may be precisely why casual use underwhelms.
Headphones are not optional. The effect requires different frequencies reaching each ear. Speakers deliver no binaural beat at all.
Treat it as a primer, not wallpaper. Habituation is the predictable failure mode of any novelty-dependent stimulus, and Huberman Lab explicitly cautions against running beats through every work bout for that reason — advice I’d follow regardless of the mechanism debate, because it costs nothing.
Don’t run gamma near bedtime. Focus-frequency tracks late in the day are a plausible route to fragmented sleep.
And a caution about the sources you’ll find if you go looking: much of the safety and dosing advice online comes from companies selling focus audio. Their guidance on ramping duration and keeping volume modest is sensible, and it is also marketing from an interested party. I’m repeating it because it’s harmless and probably right, not because it’s evidence.
The thing I keep coming back to
Somebody in a lab in 2025 found that changing a carrier tone from 340 Hz to 400 Hz was the difference between a statistically significant improvement and nothing at all. Sixty hertz on a parameter no listener has ever thought about, let alone controlled.
Now consider what that implies about the ten-hour track. It wasn’t designed against any of these findings. Its carrier was chosen for pleasantness. It has no masking noise. It runs for a work-day rather than a pre-task primer. It is, in every parameter the research says matters, a different intervention wearing the same name.
I don’t think binaural beats are fake. The entrainment is real, the narrow attentional effects are probably real, and I’d expect the next decade to find genuine uses in learning consolidation rather than in raw focus. What I think is that the gap between the laboratory version and the streaming version is wide enough that most people’s verdict — “I tried it, it did nothing” — is a review of a product that was never tested.
That’s worth generalising. When something works in research and disappoints in practice, the first question isn’t whether the research was wrong. It’s whether you were given the same thing.
About this article
Written by Drew Anton. Covers nootropics, stimulants, sleep and focus protocols, and wearables. Not a physician or research scientist — reads the primary literature closely and refuses to round up.
Medical review: None. NeuriFuel does not currently have a licensed clinician on the editorial team, and this article has not been medically reviewed. We state this rather than implying an authority we do not have. See our About page for our full methodology.
Sources: Built from fourteen primary studies including two meta-analyses, a systematic review of EEG findings, and the largest home-use trial conducted. Null results are reported alongside positive ones. Commercial sources are identified as commercial in the text. Where a value could not be verified — including the beat frequency used in the home-use trial — it is stated as unverified rather than assumed.
Corrections: Found an error? Write to hello@neurifuel.com with a source and we will fix it and log the correction.
Last updated: 29 July 2026
References
- Melnichuk A, Cooper RK Jr, Hawk LW Jr (2025). A parametric investigation of binaural beats for brain entrainment and enhancing sustained attention. Sci Rep 15:4308. DOI 10.1038/s41598-025-88517-z; PMID 39910150
- Klichowski M, Wicher A, Kruszwicka A, Golebiewski R (2023). Reverse effect of home-use binaural beats brain stimulation. Sci Rep 13:11079. DOI 10.1038/s41598-023-38313-4; PMC10329717 — funded by EU COST Action CA16118 / Horizon 2020, no industry funding
- Ingendoh RM, Posny ES, Heine A (2023). Binaural beats to entrain the brain? A systematic review. PLOS ONE 18(5):e0286023. DOI 10.1371/journal.pone.0286023
- Garcia-Argibay M, Santed MA, Reales JM (2019). Efficacy of binaural auditory beats in cognition, anxiety, and pain perception: a meta-analysis. Psychol Res 83:357–372. DOI 10.1007/s00426-018-1066-8; PMID 30073406
- Basu S, Banerjee B (2023). Potential of binaural beats intervention for improving memory and attention. Psychol Res. DOI: data not specified
- Ross B, Lopez MD (2020). 40-Hz binaural beats enhance training to mitigate the attentional blink. Sci Rep 10:7002. DOI 10.1038/s41598-020-63980-y
- Colzato LS, Barone H, Sellaro R, Hommel B (2017). More attentional focusing through binaural beats. Psychol Res 81(1):271–277. DOI 10.1007/s00426-015-0727-0
- Fujikawa I, et al. (2025). Effect of inaudible 40 Hz binaural beats on attention. Exp Brain Res 243:158. DOI 10.1007/s00221-025-07097-6
- Borges LR, Arantes APBB, Naves ELM (2023). Influence of binaural beats stimulation of gamma frequency over memory performance. Healthcare 11(6):801. DOI 10.3390/healthcare11060801 — null result
- Wang L, Zhang W, Li X, Yang S (2022). The effect of 40 Hz binaural beats on working memory. IEEE Access 10:81556–81567. DOI 10.1109/ACCESS.2022.3185257
- López-Caballero F, Escera C (2017). Binaural beat: a failure to enhance EEG power and emotional arousal. Front Hum Neurosci 11:557. DOI 10.3389/fnhum.2017.00557 — null result
- Leistiko NM, Madanat L, Yeung WKA, Stone JM (2024). Effects of gamma frequency binaural beats on attention and anxiety. Curr Psychol 43(6):5032–5039. DOI 10.1007/s12144-023-04681-3
- Chockboondee M, et al. (2024). Effects of daily listening to 6 Hz binaural beats over one month: an ERP study. Sci Rep 14:18059. DOI 10.1038/s41598-024-68628-9
- Effects of binaural and monaural beat stimulation on attention and EEG (Flanker task). PMC8448709
Distinct modality — NOT binaural beats:
- Iaccarino HF, et al. (2016). Gamma frequency entrainment attenuates amyloid load and modifies microglia. Nature 540(7632):230–235. PMC5656389 — 40 Hz GENUS light/click sensory stimulation
Authority sources: Huberman Lab, Focus Toolkit (Podcast #88) — protocol and mechanism commentary; Examine.com — binaural beats entry.