The binaural beats literature is small, and most of what circulates about it is years out of date or was never accurate to begin with. Two studies published in Scientific Reports in 2024 are worth knowing about — not because they settle anything, but because one of them quietly complicates the mechanism everyone assumes.
I have read both papers rather than their press coverage. Below is what they did, what they found, and where each one is weak. At the end there is a checklist you can use on the next study someone waves at you, which is probably the more useful part.
Study one: a month of daily listening
Chockboondee, Jatupornpoonsub, Lertsukprasert & Wongsawat (2024). “Effects of daily listening to 6 Hz binaural beats over one month: an event-related potentials study.” Scientific Reports, 5 August 2024.
Sixty healthy adults, mean age 26, split evenly into an experimental and a control group. The experimental group listened to a 6 Hz binaural beat — carriers at 250 Hz in the left ear and 256 Hz in the right — for at least ten minutes a day for a month. Everyone was assessed at baseline, two weeks and four weeks using auditory and visual oddball tasks with event-related potentials recorded.
The finding: P300 amplitude increased in the listening group at each visit. By two weeks the auditory P300 amplitude was higher than controls; by four weeks amplitude was up and latency was down, and response times to auditory stimuli had dropped significantly.
P300 is a reasonable thing to measure here. It is an ERP component associated with attention allocation and stimulus evaluation, it is well characterised, and a larger, earlier P300 is generally read as more efficient attentional processing. This is the longest daily-use protocol in the binaural beats literature that I am aware of, which alone makes it worth attention.
Where it is weak
Two problems, and the authors name both.
The control group received nothing. Not a sham tone, not a placebo audio track — no intervention at all. That means the comparison is not “binaural beats versus a similar sound” but “doing a daily ten-minute audio routine versus doing nothing”. Sitting still with headphones on every day for a month could plausibly change how you perform an attention task, entirely independently of what was playing. The study cannot separate those.
It was not blinded, and compliance was not monitored. Participants knew which group they were in, which is exactly the condition under which expectation effects appear. And in the authors' own words, they “lacked a direct method to monitor or enforce compliance outside the laboratory” — so the actual dose is unknown. They recommend future work use digital monitoring and include placebo controls, which is the right call.
None of this makes the result worthless. It makes it a promising signal that has not yet been separated from the obvious alternative explanation.
Study two: the one that complicates things
Fan, Zhu, Suzuki, Suzuki, Watanabe, Watanabe & Abe (2024). “Binaural beats at 0.25 Hz shorten the latency to slow-wave sleep during daytime naps.” Scientific Reports, 30 October 2024.
This one is much smaller — twelve participants — but considerably better designed. A randomised crossover, so every participant did every condition: a sham, a 0 Hz condition (250 Hz to both ears, so no beat at all), a 0.25 Hz beat, and a 1 Hz beat. Four ninety-minute afternoon naps, full polysomnography, scored to AASM criteria.
The 0.25 Hz condition shortened the time to reach N2 (p = 0.023) and N3 (p = 0.045) compared with sham, with effect sizes of 0.773 and 0.682. The 1 Hz beat did nothing measurable.
Note the frequency. Quarter of a hertz is far below the delta band that every sleep protocol on the internet — including, until recently, most of ours — tells you to use. If this holds up, the conventional advice to reach for 2 Hz is pointing at roughly the right region for the wrong reason.
The part nobody quotes
Here is what makes this study genuinely interesting, and it is the line that gets left out of every summary I have seen:
They found no evidence of neural entrainment. No significant differences in delta or sigma power. No sign that the brain's rhythm was following the stimulus. The sleep latency shortened, and the mechanism everyone assumes was responsible was not detectable.
That is a problem for the standard story. The entire folk explanation of binaural beats — the one on every product page, ours included — is that the difference frequency drags cortical rhythm toward itself, and the state follows. Here the outcome moved and the rhythm did not. Whatever helped these people fall asleep faster, it does not look like it was entrainment.
There are gentler readings. Twelve people is not many, and power spectral analysis may simply have been underpowered to detect a small shift. But it is at minimum a data point against assuming the mechanism is settled.
Where it is weak
The authors are unusually candid, and list seven limitations. The ones that matter most: it is exploratory with no correction for multiple comparisons, which with p-values of 0.023 and 0.045 is not a small caveat. Participants were young healthy adults, so it says nothing about older people or anyone with a sleep disorder. Several participants could not reliably perceive the beats and only high-discrimination participants were included — so this is a best case, not an average one. Left and right ear were not counterbalanced. And it was daytime napping; the authors explicitly say nighttime sleep needs validating separately.
What this changes in practice
Honestly? Not much, and I would rather say that than manufacture an implication.
The wind-down case gets slightly stronger. Both studies concern getting into a state rather than maintaining one overnight. That is consistent with what the rest of the literature supports, and it is why running beats all night remains a weaker proposition than using them for twenty minutes before bed.
Very low frequencies may be worth trying for sleep. If 2 Hz has not worked for you, 0.25 Hz is now a defensible thing to experiment with rather than a random number. You can set it in the free generator. Be aware that a beat that slow is barely perceptible as a pulse — you are not meant to notice much.
Daily use over weeks has a little more support than single sessions. Consistent with the general advice on this site, and now with a month-long protocol behind it, caveats and all.
Nobody has shown a neurotransmitter change. Neither study measured one. Neither claimed to. If you have seen these papers cited as evidence that binaural beats raise dopamine, you have seen them misrepresented — that claim has its own page.
How to read the next study yourself
This is the evergreen part. Most misinformation in this field is not fabricated from nothing — it is a real study with its qualifiers stripped off. Five questions catch nearly all of it.
1. What was the control condition? The single most important question. “No intervention” tells you far less than a sham tone, because it cannot separate the audio from the ritual around it. A crossover design where everyone does every condition is stronger still.
2. How many people? Much of this literature runs on ten to thirty participants. That is fine for exploratory work and not fine for confident claims. Effect sizes from small samples are systematically inflated.
3. What did they actually measure? EEG measures electrical rhythm. Questionnaires measure what people report. Neither measures neurochemistry, and no binaural beats study has measured a neurotransmitter, because doing so requires PET imaging or an implanted probe.
4. Who was excluded? The nap study only kept participants who could reliably perceive the beats. Entirely reasonable, and it means the result describes responders rather than people in general. Exclusion criteria are usually where the generalisability goes.
5. Did the authors correct for multiple comparisons? If a study tests many outcomes and reports the two that reached p < 0.05 without correction, treat those as leads rather than findings. Good papers say so; the nap study does.
Apply those five to any “clinically proven frequency” article and most of them dissolve. Apply them to these two studies and you get what I have described: one promising but confounded, one small but well designed and awkward for the standard mechanism.
The bottom line
The field is still small, still underpowered, and still much less settled than the marketing around it. Two decent 2024 studies moved it slightly: daily listening over a month is associated with attentional changes, and a very slow beat shortened sleep latency without any detectable entrainment.
That last part is the one I would remember. It is possible for something to work and for the reason everyone gives to be wrong. If binaural beats help you sleep, that help is real regardless of which mechanism turns out to be responsible — and the honest position is that we do not yet know.
If you want the broader picture, the science page covers what the individual studies measured, and this guide is for when nothing seems to happen at all.
Common questions
What did the 2024 binaural beats studies find?
Two studies in Scientific Reports. Chockboondee et al. found that a month of daily 10-minute listening to a 6 Hz beat increased P300 amplitude and reduced response times versus a no-intervention control. Fan et al. found that a 0.25 Hz beat shortened the time to reach N2 and N3 sleep during daytime naps versus sham — but found no evidence of neural entrainment.
Is 0.25 Hz better than 2 Hz for sleep?
One small crossover study (n=12) found 0.25 Hz shortened sleep latency where 1 Hz did not, in daytime naps, in participants selected for being able to perceive the beats. That is worth experimenting with, not a reason to abandon 2 Hz. It has not been replicated and has not been tested for nighttime sleep.
Do these studies prove binaural beats work?
No. One had no sham control and no blinding, so it cannot separate the audio from the daily routine around it. The other was exploratory with twelve participants and no correction for multiple comparisons. Both are reasonable signals; neither is proof.
Why does it matter that no entrainment was found?
Because entrainment is the mechanism every product page cites — the idea that the difference frequency pulls cortical rhythm toward it. In the nap study the sleep outcome improved while delta and sigma power showed no significant change, meaning something else may be doing the work. It does not mean the effect is not real; it means the explanation is less settled than usually presented.
How can I tell if a binaural beats claim is credible?
Ask five things: what the control condition was, how many participants there were, what was actually measured, who was excluded, and whether the authors corrected for multiple comparisons. Claims about dopamine or serotonin are the fastest tell — no binaural beats study has measured a neurotransmitter.
Try it yourself, free
Set any frequency — including 0.25 Hz — in the browser generator. No signup, no download, twenty-minute sessions.
Open the free generator