Binaural beat frequencies: what the evidence says

Compiled November 2026 for Deep Binaural. Not medical advice.

Every preset in the app plays one pure tone in the left ear and a slightly different one in the right ear. This document records where each pair of numbers comes from, what the studies behind it actually found (including null results), and how confident anyone should be. Where no study supports a number, it says so.

At a glance

Group Preset Left (Hz) Right (Hz) Beat (Hz) Minutes Evidence Numbers from
Sleep Deep sleep 250 253 3 (delta) 90 Some evidence Jirakittayakorn & Wongsawat 2018
Sleep Fall asleep 248 252 4 (delta/theta) 60 No clear benefit Shumov et al. 2023
Focus Focus 320 360 40 (gamma) 33 Some evidence Reedijk et al. 2013; Melnichuk et al. 2025
Focus Creativity 335 345 10 (alpha) 17 No clear benefit Reedijk et al. 2013
Focus ADHD (unproven) 400 415 15 (beta) 20 No clear benefit Malandrone et al. 2022
Calm Calm 200 209.3 9.3 (alpha) 10 Some evidence Isik et al. 2017
Calm Relax 196 203 7 (theta) 20 Some evidence McConnell et al. 2014; Atwater (Monroe Institute)
Calm Meditation 250 256 6 (theta) 10 Some evidence Jirakittayakorn & Wongsawat 2017
Calm Acute pain 450 440 10 (alpha) 10 Some evidence Aly et al. 2023
Explore Lucid dreams 200 204 4 (delta/theta) 30 Reported experiences Monroe, US 5,356,368
Explore Out-of-body 100 104 4 (delta/theta) 30 Reported experiences Monroe, US 5,213,562

Every preset starts with both ears on the midpoint of the two tones and glides to the target pair over 10 seconds. That start is a comfort feature: no study compares gradual and fixed starts (see How binaural beats work).

How to read the grades

A comparison against silence cannot separate the beat from “any quiet steady sound”, so silence-controlled evidence alone never earns more than some evidence.

The short version

How binaural beats work (and their limits)

Headphones are required. A binaural beat exists only when each ear hears only its own tone; the brain produces the beat. If both tones reach the same ear, whether mixed electronically, played through speakers, or summed by a phone’s Mono Audio setting, they interfere physically and make an ordinary monaural beat instead (Oster 1973; Ingendoh et al. 2023).

Tone pitch (the “carrier”). Binaural beats are heard only with low tones. Classic psychophysics found them most detectable, with the widest range of beat rates, around 400 to 500 Hz (Oster: best near 440 Hz), fading toward 1,000 Hz; reported upper limits range from about 900 to 1,500 Hz (Licklider et al. 1950; Perrott & Nelson 1969; Oster 1973). Slow beats (2 to 6 Hz) were about equally strong with 150 and 300 Hz tones and weaker with 600 Hz tones (Rutschmann & Rubinstein 1965), and Grose, Buss & Hall expect beats around 400 to 500 Hz to be less salient than at 150 to 300 Hz (Grose, Buss & Hall 2012); cortical responses to 3 to 6 Hz beats were larger with a 250 Hz than a 1,000 Hz tone (Pratt et al. 2009). Below about 90 Hz, listeners may confuse the beat with the tones themselves (Oster), which puts Monroe’s 100 Hz example near the bottom edge. Modern studies mostly use 200 to 500 Hz.

Beat rate. As the difference between the ears grows, the sensation moves from a slow sweep inside the head, to a loudness wobble, to roughness, and finally to two separate tones. That happens at around 26 to 35 Hz for tones near 400 to 440 Hz (Licklider; Oster). A 40 Hz beat is therefore at or past the limit of what is heard as a beat, yet it still produces a measurable 40 Hz brain response with tones around 400 to 500 Hz. That response is smaller than for the same beat played acoustically, disappears with tones above about 800 Hz, and appears less often in older adults (Schwarz & Taylor 2005; Draganova et al. 2008; Grose & Mamo 2012).

The sensation is faint, and not everyone hears it. Older listeners showed the 40 Hz response less often (9 of 14 versus all 10 young listeners), low-tone hearing loss in one ear widens the range where no beat is heard, and Oster found that some neurological patients could not hear binaural beats at all (Grose & Mamo 2012; Karino et al. 2005).

Does the brain follow the beat? There is a small brain response locked to the beat frequency while it plays (Ross & Lopez 2020 found strong 40 Hz entrainment in MEG). But the claim behind most apps, that a beat shifts your whole brain into a band and therefore a state, is inconsistently supported (5 consistent, 8 contradictory, 1 mixed of 14 controlled studies; Ingendoh et al. 2023).

Background noise is optional. The 2019 meta-analysis found similar effects with or without white or pink noise masking, and larger effects with longer listening and listening before a task (Garcia-Argibay et al. 2019). Several of the source studies did use a noise bed; the app plays the tones alone, and each preset notes where that departs from its source.

Gradual starts are not evidence-based. No study compares a ramped or step-down beat schedule with a fixed beat. Laboratory onsets use 20 to 25 ms ramps only to avoid clicks. The closest published idea is a “dynamic” beat that swept one ear from 100 to 103 Hz (Lee HA et al. 2024). The app’s 10-second convergent start is a comfort choice.

“Inaudible” beats don’t fit the mechanism. Products using ultrasonic carriers (for example 18,000 / 18,010 Hz) cannot work as binaural beats: pure-tone beats are not reliably heard above about 1,500 Hz, and no 40 Hz binaural response appears above 3 kHz (Kim et al. 2023 compared such stimuli only with silence).

The Monroe Institute: what is actually published

Robert Monroe’s Hemi-Sync process is the best-known binaural-beat system, so we searched his patents and the Monroe Institute’s (TMI) own documents, including the research archive TMI uploaded to archive.org.

Patents (all expired)

Patent What it covers Filed Granted Expired
US 3,884,218 Inducing and maintaining stages of sleep 30 Sep 1970 20 May 1975 20 May 1992
US 5,213,562 Inducing mental, emotional and physical states 25 Apr 1990 25 May 1993 25 May 2010
US 5,356,368 Method and apparatus for inducing desired states of consciousness 1 Mar 1991 18 Oct 1994 18 Oct 2011

US patents filed before June 1995 lasted 17 years from grant; Google Patents lists all three as “Expired – Lifetime”. The name Hemi-Sync is a registered trademark, which is a separate matter from the patents.

Focus levels are not published

No TMI primary source we found gives frequencies for Focus 10, 12, 15, 21 or higher. TMI describes its Hemi-Sync patterns as proprietary (TMI 1992), and its current Research Overview and “Monroe Sound Science” pages describe its “proprietary sound technology” without listing frequencies (TMI research overview). The only published description of Focus 10 (“mind awake, body asleep”) is qualitative: a delta beat introduced first, with a low-beta beat mixed in later (Atwater 1986).

The “Focus level frequency” tables that circulate online come from an anonymous USENET measurement of the tapes, reproduced in the SBaGen software’s documentation (focus.txt). It lists Focus 10 as 100[1.5], 200[4], 250[4] and 300[4] Hz (centre[beat]), but its own per-tape lines differ (for example “l/r 100/104, 414/410, 504/500”), and a second amateur analysis found only a ~1.6 Hz beat near 108 Hz during the count into Focus 10. These are not TMI publications.

The declassified 1983 US Army assessment of the Gateway process (by LTC Wayne M. McDonnell) gives no Focus-level frequencies; its only Hemi-Sync number tied to out-of-body experiences is a second-hand magazine quote of “around 2877.3 CPS”, far above any brainwave range (CIA-RDP96-00788R001700210016-5).

What studies of TMI’s recordings show

The presets

Each section gives the numbers, why they were chosen, and the caveats shown in the app. The collapsible notes keep the original researcher’s rationale and caveats (written before fact-checking) and list every source they found, with the protocol, the finding and a verbatim quote as corrected by the fact-check.

Deep sleep

250 Hz left · 253 Hz right · 3 Hz beat (delta) · 90 min · Some evidence

In one small randomized sleep-lab study, this exact 3 Hz beat increased deep (N3) sleep compared with silence. In that study EEG sleep staging switched the tones on during light sleep and off during deep sleep; playing them from bedtime, as here, has not been tested against a control. Follow-up studies are mixed.

Why these numbers. The tones are exactly those of Jirakittayakorn & Wongsawat (2018): 250 Hz left, 253 Hz right, at a quiet 60 dB. The 90-minute timer (about one sleep cycle) is borrowed from Dabiri et al. (2022), an uncontrolled study that played a 3 Hz beat at bedtime; it is a reasonable guess, not a tested dose.

Caveats.

Full research notes for this preset

Researcher’s rationale (written before fact-checking; where it differs, the text above and the checked sources below take precedence): The left/right tones come straight from Jirakittayakorn & Wongsawat 2018, which states that 250 Hz went to the left ear and 253 Hz to the right, at 60 dB SPL, through in-ear headphones taped in place. That study used closed-loop delivery: the tone started at each first epoch of N2, stopped at the first epoch of N3, and was used only within the first 5 h of sleep, adding up to a mean of 193.21 +/- 35.27 min per night. An app without real-time sleep staging cannot do this. The 90-minute session length is borrowed from a different, open-loop protocol, Dabiri et al. 2022. They played a 3 Hz beat (174 Hz left / 177 Hz right) for 90 min at bedtime, chose 90 min to match one sleep cycle, and had no control group. Playing continuously from bedtime has not been tested with this 250/253 Hz pair. The same 250/253 Hz pair was also used by Hestermann 2024 (mixed with ASMR, during N2) and Hou 2026 (30 min before bed, mixed with music).

All caveats the researcher found (before fact-checking):

  • The control in the key study was silence, not the same tone without a beat. So the study cannot show that the beat itself, rather than any steady quiet sound, caused the change.
  • Very small, unbalanced groups: 15 analysed versus 8 controls, healthy adults averaging 24 years, one night in a lab. Experimental-group N3 went from 90.6 +/- 16.3 min (baseline night) to 125.9 +/- 30.8 min (stimulation night).
  • The study switched the tone on at N2 and off at N3. A simple timer cannot reproduce this, and continuous playback (including during N3) was not tested.
  • Replication is mixed. Hestermann 2024 (3 Hz with ASMR, 8 vs 7 people) found no statistically significant N3 change. Dabiri 2026 (174 Hz carrier, n=20, no sham sound) and a 9-person conference abstract (Anuekjit 2025) reported more N3. Fan 2024 found 1 Hz beats had no effect on nap sleep.
  • Sound during sleep can cause awakenings. In Huwiler 2022, 1 Hz binaural beats played during NREM sleep raised slow-wave activity but also significantly raised the chance of waking.
  • Keep the volume low. The key study used 60 dB SPL, which it describes as the threshold of auditory stimulation during sleep.
  • EEG entrainment by binaural beats is inconsistent across studies (Ingendoh 2023 systematic review: 5 of 14 studies supportive, 8 contradictory, 1 mixed).
  • Monroe’s sleep patent (US3884218) describes amplitude-modulated sounds, not binaural beats, so it gives no support to this preset.

Sources (fact-checked; corrections applied):

  • Jirakittayakorn N, Wongsawat Y (2018). A Novel Insight of Effects of a 3-Hz Binaural Beat on Sleep Stages During Sleep. Front Hum Neurosci 12:387. — rct, verified.
    • Protocol: 250 Hz left ear / 253 Hz right ear (3 Hz beat), pure tones, no noise bed, 60 dB SPL, Beats X wireless in-ear headphones taped in place. Three consecutive lab nights (adaptation, baseline, experimental). On night 3 the tone started at each first N2 epoch and stopped at the first N3 epoch, repeated until the 5th hour of sleep, with a mean of 193.21 +/- 35.27 min of stimulation. 24 healthy adults (mean age 24.1) randomized: 16 experimental (15 analysed) vs 8 controls. The sham was silence. Sleep was staged from EEG/EOG/EMG (AASM).
    • Found: Compared with controls, the experimental group showed a larger night-3 minus night-2 increase in N3 duration (p=0.0011 for minutes, p=0.0017 for %), a smaller N2, and a shorter N3 latency (p=0.0226). No change in arousal index or WASO. Delta power rose only in the experimental group.
    • “A 250-Hz carrier tone was presented to the left ear and a 253-Hz tone was presented to the right ear. This binaural beat file was set to 60 dB SPL as the threshold of auditory stimulation during sleep […] The sham stimulus used in this study was silence for comparing sleep stages between a binaural beat and a non-binaural beat. […] The results showed that the N3 duration of the experimental group was longer than that of the control group, and the N2 duration of the experimental group was shorter than that of the control group.”

  • Dabiri R, Monazzam Esmaielpour MR, Salmani Nodoushan M, Khaneshenas F, Zakerian SA (2022). The effect of auditory stimulation using delta binaural beat for a better sleep and post-sleep mood: A pilot study. Digit Health 8:20552076221102243. — uncontrolled study, verified.
    • Protocol: 174 Hz left / 177 Hz right (3 Hz beat), 90 min while lying down to sleep, through stereo headphones, 60-75 dB SPL (most chose 60 dB). 20 healthy students. One baseline week with no intervention, then one stimulation week. Outcomes: sleep diary and POMS.
    • Found: Diary measures improved in the stimulation week versus baseline (sleep failure, awakenings, sleep duration, sleep quality), as did anxiety and anger. There was no control condition, so expectancy and time effects cannot be excluded.
    • “In the first week, there was no intervention, but in the second week, all subjects were exposed to a 90 min binaural beat in the delta frequency range. […] The stimulus in this investigation was a 3-Hz binaural beat on a 174-Hz carrier tone that was created specifically for the study by Audacity software. A tone of 174 and 177 Hz were played in the left and right ear, respectively.”

  • Dabiri R, Monazzam Esmaeelpour MR, Nodoushan MS, Khanehshenas F, Zakerian SA (2026). The effect of delta binaural beat on sleep onset and maintenance based on EEG signal analysis. Acta Psychol (Amst) 269:107506. — controlled study, verified.
    • Protocol: Abstract only (full text blocked). ‘Delta’ binaural beat on a 174 Hz carrier. The per-ear beat frequency, timing and duration are not in the abstract. 20 healthy students, 3 nights: one stimulation night and two nights with no intervention. No sham sound.
    • Found: Did not reduce sleep onset time. The authors report longer N3 and REM sleep and fewer awakenings on the stimulation night.
    • “In this study, we investigated the effect of delta binaural beats on sleep onset and sleep maintenance using a 174 Hz carrier tone. […] Delta binaural beat did not reduce sleep onset time, but improved sleep maintenance parameters, including increasing the duration of N3 and REM sleep and reducing the amount and number of awakenings.”

  • Hestermann E, Schreve K, Vandenheever D (2024). Enhancing Deep Sleep Induction Through a Wireless In-Ear EEG Device Delivering Binaural Beats and ASMR: A Proof-of-Concept Study. Sensors 24(23):7471. — controlled study, verified.
    • Protocol: 250 Hz left / 253 Hz right (3 Hz), mixed with ASMR at a 60:30 dB ratio (ASMR:beat), through Sony WI-C200 earphones. The sound started on entering N2 and switched to ASMR only at N3. 15 participants randomly assigned: 8 ASMR+beat, 7 ASMR only. Baseline night then test night, recorded with an in-ear EEG device.
    • Found: N3 percentage was numerically higher with ASMR+beat (24.50% vs 17.80% baseline and 19.28% control), but this was not statistically significant. The only significant result was a theta-power change in the control group.
    • “This beat featured a 250 Hz carrier tone for the left ear and a 253 Hz tone for the right ear, creating a 3 Hz binaural beat within the delta range, ideal for NREM stage 3 sleep. […] The only statistically significant finding, after normalization, was the difference in theta band power between the control group and its baseline”

  • Anuekjit T, Jansri U, Chintanadilok J (2025). 0467 The Effects of 1-4 Hz Binaural Beats on Delta Brain Wave During Sleep in University Students: A Pilot Study. Sleep 48(Suppl 1):A204 (conference abstract), doi:10.1093/sleep/zsaf090.0467. — uncontrolled study, corrected after re-check.
    • Protocol: Conference abstract (verbatim text via Crossref). ‘1-4 Hz’ beats built from musical chords, played for 30 min during sleep. 9 students. A no-beat night was followed one week later by a beat night (fixed order, not randomized). Polysomnography. Carrier not reported.
    • Found: Per the abstract: N3 increased (p=0.033) and N2 decreased (p=0.049) on the beat night; other measures were not significant.
    • “1-4 Hz binaural beats were created by musical chords to prevent monotony from using only one value of frequency […] On the first night, participants slept without listening to binaural beats, while on the second night (7 days interval), they listened to 1-4 Hz binaural beats for 30 minutes during sleep. […] The results of this pilot study shows that the specific binaural beats increase total sleep time (TST), sleep efficiency, sleep onset, N1, N3 and REM stages but only N3 stage is statistically significant (p= 0.033). For the N2 stage, it decreases with statistical significance (p = 0.049), as calculated by the Wilcoxon-Signed Ranks test by SPSS software.”

  • Huwiler S, Carro Dominguez M, Huwyler S, Kiener L, et al. (2022). Effects of auditory sleep modulation approaches on brain oscillatory and cardiovascular dynamics. Sleep 45(9):zsac155. — controlled study, verified.
    • Protocol: A 1 Hz binaural beat on a ‘400 Hz’ carrier. The paper does not say which ear got which tone, so per-ear values cannot be derived. It was one of several conditions played in pseudorandom 10 s ON / 10 s OFF windows during detected stable NREM sleep, at 45 dB, compared within the same night to sham windows. Study 1: 23 healthy men (mean age 40).
    • Found: Binaural-beat windows increased slow-wave activity about as much as closed-loop stimulation did. They also significantly increased the probability of waking (about 9% in the early night) and raised low alpha activity, which the authors read as a possible sign of arousal.
    • “BINAURAL BEATS with a difference of 1 Hz between the two ears and carrier frequency of 400 Hz […] Post hoc comparisons of the conditions revealed a significantly increased wake probability for BINAURAL BEATS condition […] we only found binaural beats to increase the alpha activity of the stimulation ON and OFF window”

Fall asleep

248 Hz left · 252 Hz right · 4 Hz beat (delta/theta border) · 60 min · No clear benefit

The largest controlled test (202 students, these exact tones) found people fell asleep no faster than with the same tone without a beat. A double-blind insomnia trial found beats added nothing to music. Only very small studies were positive.

Why these numbers. The tones and the 60-minute length come from Shumov et al. (2023), chosen because it is the largest and best-controlled test of falling asleep, not because it was positive. That study added pink noise; this app plays the tones alone.

Caveats.

Full research notes for this preset

Researcher’s rationale (written before fact-checking; where it differs, the text above and the checked sources below take precedence): Frequencies are exactly those of Shumov et al. 2023, reused in Shumov et al. 2024: binaural condition left channel 248 Hz, right channel 252 Hz, with pink noise on both channels. Participants heard it through stereo headphones for 1 h while falling asleep at home between 21:00 and 24:00. That study was chosen because it is the largest and best-controlled test of sleep onset (same-pitch tone-only and monaural-beat controls, randomized counterbalanced order, 202 analysed), not because it was positive. It was null. The 60-minute session is that study’s playback length. Playing pure tones without the pink-noise bed is an adaptation.

All caveats the researcher found (before fact-checking):

  • Null result: stimulus type (binaural vs monaural vs no-beat tone) had no significant effect on sleep latency (rANOVA p=0.21; 202 of 221 participants analysed).
  • Dabiri 2026 (abstract only, 174 Hz carrier delta beat, n=20) also reports that the beat did not reduce sleep onset time.
  • In the Bang 2019 double-blind RCT (n=43, mild insomnia), theta beats embedded in music for 30 min before bed for 2 weeks did not improve Insomnia Severity Index more than the same music without beats. The 6 Hz value comes from the companion EEG paper (Lee E 2022), not the trial abstract.
  • Two other trials found no advantage for beats mixed into music: Lee M 2022 (n=76) versus ASMR, and Hou 2026 (n=66) versus pink noise plus music. In Hou 2026, beats plus music beat a no-audio control on PSQI, but participants were not blinded.
  • In Shumov 2024 (n=63), beats alone gave the longest sleep latency of the three conditions tested. The authors concluded that suggestion (relaxation instructions), not the beats, was decisive.
  • Positive sleep-onset results are small or silence-controlled. Lee HA 2024 (n=10) used a dynamic 100 Hz / 100-103 Hz beat versus no sound. Shumov 2017 (n=14) used a multi-tone chord; for the whole group it was p=0.043 versus the no-beat tone and not significant versus monaural beats.
  • Shumov 2023 measured sleep onset with a smartphone test (lift a finger in response to a signal every 7-13 s), not with polysomnography.
  • 4 Hz sits on the delta/theta boundary. The ‘theta’ label is convention, not evidence.
  • Shumov 2024 reports stimulus level only as ‘Мощность всех аудиостимулов составляла 6 Дб’ (‘the power of all audio stimuli was 6 dB’), which is unclear. Use a quiet, comfortable level.

Sources (fact-checked; corrections applied):

  • Shumov DE, Sveshnikov DS, Bakaeva ZV, Dorokhov VB (2023). [Monotonous sound stimuli embedded with either binaural or monaural beats have the same effect on sleep latency: study on a large subject group]. Zh Nevrol Psikhiatr Im S S Korsakova 123(4):109-113. (In Russian) — controlled study, verified.
    • Protocol: Binaural: left 248 Hz / right 252 Hz (4 Hz). Monaural: 248 + 252 Hz in both channels. Control: 250 Hz in both channels. All mixed with pink noise. Each was a 1 h two-channel recording heard through stereo headphones at home, 21:00-24:00, while falling asleep. 221 medical students, each tested with all 3 sounds in random counterbalanced order (202 analysed). Sleep latency came from a smartphone tap-response test. Quote sources: Russian text from the mediasphera full text; English sentence from the PubMed abstract (https://eutils.ncbi.nlm.nih.gov/entrez/eutils/efetch.fcgi?db=pubmed&id=37084374&rettype=abstract&retmode=text).
    • Found: Null: sleep latency did not differ between binaural, monaural and no-beat sounds (p=0.21; pairwise p=1.0 after correction).
    • “Спектральные характеристики стимулов: Контроль — частота 250 Гц, симметрично как по левому каналу, так и по правому; МБ — частоты 248 и 252 Гц, симметрично как по левому каналу, так и по правому; ББ — левый канал 248 Гц; правый канал 252 Гц. […] Repeated measures analysis of variance (rANOVA) revealed no significant statistical effect of stimulus type on SL (p=0.21).”

  • Shumov DE, Sveshnikov DS, Bakaeva ZV, Yakunina EB, Dorokhov VB (2024). [The effectiveness of suggestive techniques to improve sleep is enhanced by combining with sound stimulus based on binaural beats]. Zh Nevrol Psikhiatr Im S S Korsakova 124(5 Vyp 2):20-25. (In Russian) — controlled study, corrected after re-check.
    • Protocol: Binaural: left 248 Hz / right 252 Hz plus pink noise, heard for the whole 1 h session. Compared with 10 min of recorded body-scan relaxation instructions alone, and with instructions plus the binaural sound. A no-beat 250 Hz tone was used only in the introductory session and excluded from analysis. Smartphone sleep-latency test at home. 63 participants analysed, randomized counterbalanced order. Quote sources: Russian text from the mediasphera full text; English sentences from the PubMed abstract (https://eutils.ncbi.nlm.nih.gov/entrez/eutils/efetch.fcgi?db=pubmed&id=38934662&rettype=abstract&retmode=text).
    • Found: Sleep latency (mean +/- SE, 63 analysed): instructions plus binaural 1149 +/- 113 s, instructions alone 1269 +/- 112 s, binaural alone 1469 +/- 89 s (longest). Only instructions+binaural vs binaural alone was significant (p=0.0055). Instructions alone vs binaural alone was a trend (p=0.077). Instructions+binaural vs instructions alone was not significant (p=0.30), so adding the binaural sound did not significantly shorten latency beyond the instructions. There was no binaural-versus-no-beat comparison. The authors conclude that suggestion is the determining factor.
    • “Спектральные характеристики стимулов: ББ — левый канал 248 Гц; правый канал 252 Гц, с наложением на оба канала розового шума […] Pairwise SL comparison showed reliable difference between stimuli (3) - 1149±113 s, and (1) - 1469±89 s (p<0.01). […] But it is the suggestion as a psychotherapeutic technique that is determinant.”

  • Bang YR, Choi HY, Yoon IY (2019). Minimal Effects of Binaural Auditory Beats for Subclinical Insomnia: A Randomized Double-Blind Controlled Study. J Clin Psychopharmacol 39(5):499-503. — rct, verified.
    • Protocol: Abstract only; carrier not reported. The abstract says ‘θ BB’ embedded in music; the companion paper (Lee E 2022) specifies 6 Hz played with music as the carrier on a Dlogixs device. 30 min before sleep for 2 weeks. Control: the same music without beats. 43 completers with subclinical insomnia, double-blind.
    • Found: Null for sleep: Insomnia Severity Index fell in both groups with no significant difference between them (P=0.656). Awake EEG relative beta power was higher after beats.
    • “Subjects with subclinical insomnia were randomly assigned and listened to music for 2 weeks with or without θ BB for 30 minutes before going to sleep. […] The insomnia severity decreased at 2 weeks in both groups without significant differences in Insomnia Severity Index score (4.41 ± 4.32 vs 2.71 ± 4.66, P = 0.656)”

  • Lee E, Bang Y, Yoon IY, Choi HY (2022). Entrapment of Binaural Auditory Beats in Subjects with Symptoms of Insomnia. Brain Sci 12(3):339. — rct, corrected after re-check.
    • Protocol: EEG secondary analysis of the Bang 2019 trial. A 6 Hz beat played together with music (no pure-tone per-ear frequencies published) versus a sham device playing music only, 30 min before bed for 2 weeks.
    • Found: Awake in-lab EEG only; no sleep outcomes are reported here. Before the intervention, music without beats raised relative delta and theta power in several regions and lowered alpha, while music with 6 Hz beats raised theta only in occipital regions (p=0.009). After 2 weeks, music with beats raised parietal theta in participants who had used the music-only device (p=0.009). The only between-group difference was a larger occipital beta decrease in the beats group (p=0.035). The theta rise was therefore not shown to be specific to the beats. Sleep outcomes from this trial (Bang 2019) were null.
    • “Participants used either a test device in which music and BB of 6 Hz played simultaneously or a sham device (control group) in which only music played for 30 min before going to bed every day for two weeks.”

  • Lee M, Lee HJ, Ahn J, Hong JK, Yoon IY (2022). Comparison of autonomous sensory meridian response and binaural auditory beats effects on stress reduction: a pilot study. Sci Rep 12:19521. — rct, verified.
    • Protocol: Beats with music via a Dlogixs app: 5 Hz for 30 min before sleep and 8 Hz for 15 min in the daytime, for 3 weeks. Compared with ASMR. Double-blind randomized, 76 completers with stress. Carrier tones not reported.
    • Found: ISI and PSQI improved in both groups, with no significant group-by-time interaction (no advantage for beats).
    • “Subjects listened to ASMR or BB with music (8 Hz for daytime, 5 Hz for nighttime) for 15 min in daytime and 30 min before going to sleep for 3 weeks. […] No significant group-by-time interaction was found in PSS, BDI-II, ISI, STAI-S, PSQI, and QOL-BREF scores at 3 weeks between the two groups.”

  • Hou H, Wang J, Yue X (2026). The impact of sound intervention on sleep quality and stress levels in college students: a randomized controlled trial. Front Psychol 17:1859138. — rct, verified.
    • Protocol: 250 Hz left / 253 Hz right (3 Hz) mixed with instrumental music, 30 min before bed through headphones, 5 nights a week for 4 weeks. Arms: pink noise + music, beats + music, or a no-audio control. 66 poor sleepers (PSQI >7), 22 per arm. Assessors blinded, participants not.
    • Found: PSQI, beats+music vs no-audio control: MD -1.23 (95% CI -1.98 to -0.48) at week 4. Beats+music vs pink noise+music: not significant (adjusted p=0.319-0.659), so the beat-specific effect was not shown.
    • “binaural beats were generated using 250 HZ as the carrier frequency with a 3 Hz differential frequency superimposed, i.e., the left and right ears were presented with a sine wave of 250 Hz and 253 Hz, respectively […] Control group: no audio interventions, maintain routine. […] the statistical test showed that the difference in PSQI between the two groups did not reach the level of significance”

  • Lee HA, Lee WJ, Kim SU, et al. (2024). Effect of dynamic binaural beats on sleep quality: a proof-of-concept study with questionnaire and biosignals. Sleep 47(10):zsae097. — controlled study, corrected after re-check.
    • Protocol: Left ear fixed at 100 Hz; right ear varied between 100 and 103 Hz (a 0-3 Hz ‘dynamic’ beat). Played before sleep, at sleep onset, and at REM-to-NREM transitions. Crossover against SHAM (no sound), 10 healthy adults, polysomnography. Per the publisher full text (read as a summary, not verbatim): sweep period about 1 min, about 50 dB through earphones, 20-min bouts.
    • Found: Sleep latency was shorter with the dynamic beat than with silence (6.1 +/- 4.3 vs 12.5 +/- 7.9 min, p=.027). In the pre-sleep period, the proportion of wake was lower (53.1% vs 70.2%, p=.006) and N1 higher (40.6% vs 23.0%, p=.013). Whole-night measures did not differ significantly: sleep efficiency 89.6% vs 85.8% (p=.064), WASO 9.1% vs 12.3% of sleep period (p=.10), total sleep time 438 vs 420 min (p=.13), with no other sleep-stage differences. Subjective drowsiness after sleep fell more with the beat (p=.004). Heart rate variability was lower during pre-sleep and sleep onset. The control was silence, so the effect cannot be credited to the beat rather than to sound in general. The full-text numbers were read as a summary, not verbatim.
    • “Specifically, the carrier frequency of the right ear varied between 100 and 103 Hz over a period, while the left ear remained fixed at 100 Hz, yielding a frequency difference range of 0 to 3 Hz. […] DBB significantly reduced sleep latency compared to the SHAM condition.”

  • Shumov DE, Arsen’ev GN, Sveshnikov DS, Dorokhov VB (2017). Comparative analysis of the effect of stimulation with a binaural beat and similar kinds of sounds on the falling asleep process: a brief note. Moscow Univ Biol Sci Bull 72:33-36 (Russian version: Vestnik Moskovskogo Universiteta Ser 16 Biologiya 72(1):39-43). — controlled study, verified.
    • Protocol: A four-tone chord over pink noise. Left 47.89 / 95.74 / 191.48 / 239.87 Hz; right 48.39 / 97.74 / 195.48 / 243.87 Hz, giving beats of 0.5, 2, 4 and 4 Hz. Controls: monaural (both channels mixed) and a no-beat chord. 15.5 min of stimulation through Sennheiser CX-200 in-ear earphones at 50-53 dB SPL during a daytime nap (15:00-18:00). 14 healthy adults, random order, EEG (latency to first spindle / N2).
    • Found: Whole group: binaural vs no-beat p=0.043, binaural vs monaural p=0.13 (not significant). Means were reported only for a post hoc subgroup of 10 ‘successful’ subjects (binaural 429 s vs no-beat 809 s).
    • “левое ухо (L) — 1) 47,89 Гц; 2) 95,74 Гц; 3) 191,48 Гц; 4) 239,87 Гц; правое ухо (R) — 1) 48,39 Гц; 2) 97,74 Гц; 3) 195,48 Гц; 4) 243,87 Гц, что давало в итоге спектр звукового аккорда из 4 ББ — 0,5 Гц; 2 Гц; 4 Гц; 4 Гц […] при стимуляции ББ и ИМ для всей группы из 14 человек составила p = 0,043 по критерию знаковых рангов Уилкоксона, при стимуляции ББ и МБ – p = 0,13 (достоверных различий нет)”

Focus

320 Hz left · 360 Hz right · 40 Hz beat (gamma) · 33 min · Some evidence

Some small controlled studies found better attention-task scores with 40 Hz beats. A 58-person study found nothing. In the best-designed study, beats did no better than a plain tone overall; 40 Hz beats helped with tones centred on 340 Hz but not 400 Hz, and, in a separate analysis pooling 16 and 40 Hz beats, beats helped only when quiet white noise played underneath.

Why these numbers. Reedijk et al. (2013) print this exact pair for a 40 Hz beat: 320 Hz left, 360 Hz right, centred on 340 Hz. Melnichuk et al. (2025) found the attention benefit with 40 Hz beats on that 340 Hz centre during a 33-minute task, hence the session length. Several of the positive studies added white noise; this app plays the tones alone.

Caveats.

Full research notes for this preset

Researcher’s rationale (written before fact-checking; where it differs, the text above and the checked sources below take precedence): Per-ear values are derived from Melnichuk et al. 2025, the gamma-beat x 340 Hz-carrier cell that raised average CPT hits. Tones were ‘equidistant’ from the carrier, with the lower tone always in the left ear; the paper prints only the beta example (332/348 Hz), so 320 Hz left / 360 Hz right for 40 Hz is derived, not printed. The 33-minute session matches the ~33-min task during which the tones played without a break; in half the participants they started 10 min before the task. Melnichuk used 70 dB(A) tones and, for half the sample, 60 dB(A) white noise. Bare sines are an adaptation.

All caveats the researcher found (before fact-checking):

  • In Melnichuk 2025 the overall beats-vs-pure-tone effect was not significant (p = 0.385). The benefit appeared only for gamma at the 340 Hz carrier (p = 0.002), did not appear at 400 Hz (non-significant, opposite direction), and did not reduce the vigilance decrement. 64 were analysed (80 randomized), about 4 per design cell. The authors call for replication.
  • In Melnichuk 2025, beats helped only with a white-noise background (p = 0.033), not without (p = 0.434, opposite direction). This app plays bare sine tones.
  • Leistiko 2023 (n = 58; 340 Hz left / 380 Hz right in pink noise; 5 min before and during the task) found no effect on any Attention Network Test measure.
  • Rakhshan 2022 (240 Hz base, 40 Hz arm) found no effect of any beat on ranked working memory or response time, and slightly more verbal false alarms with 40 Hz. Borges 2023 (400 Hz right/440 Hz left under music) found no memory or EEG change. In Wang 2022, accuracy in the 40 Hz group dropped mid-task.
  • Engelbregt 2019’s faster Flanker times were matched by monaural beats, and its white-noise condition always ran first, so practice is a confound.
  • The Leiden studies (Reedijk 2015, Colzato 2017, Hommel 2016) describe the 40 Hz beat only as ‘based on a 340 Hz carrier tone’, not per ear. Reedijk’s effect was in a low-blink-rate subgroup only. Hommel’s effect (more dual-task cross-talk) was read as more flexibility, not focus. A 2017 40 Hz paper from the same group was retracted for manipulated data.
  • Positive results with inaudible-range carriers (Fujikawa 2025) and uncontrolled pre/post studies (Shakya 2026; Alhasani 2026) are not evidence for this audible preset.
  • Not the same as 40 Hz tACS or 40 Hz audiovisual flicker. Evidence for those does not transfer.

Sources (fact-checked; corrections applied):

Creativity

335 Hz left · 345 Hz right · 10 Hz beat (alpha) · 17 min · No clear benefit

In the controlled study these tones come from (24 people), beats had no average effect on brainstorming. People with low eye-blink rates did better; others did the same or worse. Later controlled studies found no creativity effect; the one positive report had no control group and used a self-rated personality scale.

Why these numbers. The tones are exactly those of Reedijk et al. (2013): 335 Hz left, 345 Hz right, played for 3 minutes before and during 14 minutes of creativity tasks. That study added quiet white noise; this app plays the tones alone.

Caveats.

Full research notes for this preset

Researcher’s rationale (written before fact-checking; where it differs, the text above and the checked sources below take precedence): The left/right tones are copied from Reedijk et al. 2013 Methods: 335 Hz to the left ear, 345 Hz to the right (340 Hz centre, 10 Hz beat); the paper’s Introduction gives the reverse assignment as a generic example, so Methods is used. The 17-minute session is 3 min of listening before the tasks + 10 min Alternate Uses Task + 4 min Remote Associates Task, with sound running throughout. Adaptation: the study added band-filtered white noise (20 Hz-10 kHz) in both ears, so pure sines depart from the tested protocol. The same study’s 40 Hz condition (320 Hz left / 360 Hz right) worked the same as alpha; alpha is chosen only because the authors’ abstract singles it out for low-blink-rate listeners.

All caveats the researcher found (before fact-checking):

  • No average effect: divergent-thinking flexibility did not differ between beats and the plain tone (F(2,46) < 1); an effect appeared only after adding baseline eye-blink rate as a covariate.
  • Listeners with higher blink rates (about 20 per minute or more) were unaffected or impaired, so the same track may help some people and hinder others; users cannot tell which group they are in.
  • Alpha (10 Hz) and gamma (40 Hz) gave the same result, so there is no evidence that 10 Hz specifically matters.
  • No effect on convergent creativity (Remote Associates Test) in Reedijk 2013 or in the Petrovich 2018 thesis (delta/theta beats in rainforest sounds); Bayrami 2026 also found no creativity gain.
  • One small study (n=24, 22 women), never replicated; no meta-analysis covers creativity.
  • Tested with white noise in both ears, in-ear headphones, and 3 min of listening before plus during the tasks.
  • Theta (6 Hz) creativity claims rest on an uncontrolled study using a self-report creative-personality scale (Mendes 2025); a preprint titled ‘Binaural beats theta does not increase creativity’ could not be retrieved. Popular charts linking theta to creativity lack controlled binaural-beat support.
  • No mood change accompanied the effect (PANAS and affect grid unchanged).
  • Indirect only: 40 Hz beats increased dual-task crosstalk, read as cognitive flexibility (Hommel 2016, n=40); that is not a creativity test.
  • Ear assignment is arbitrary in the literature: Chen 2025 built its 10 Hz beat the other way round (345 Hz left / 335 Hz right); no study has tested whether it matters.

Sources (fact-checked; corrections applied):

ADHD (unproven)

400 Hz left · 415 Hz right · 15 Hz beat (beta) · 20 min · No clear benefit

Small controlled studies in people with ADHD or high inattention found no objective attention benefit. In one, beats slowed reactions and did worse than medication and silence; in another, participants found them disturbing and stressful. The only positive results were subjective. This is not a treatment.

Why these numbers. The tones are from Malandrone et al. (2022), an add-on trial in adults with ADHD: 400 Hz left, 415 Hz right (placebo: 400 Hz in both ears). That report gives no session length, so 20 minutes follows Kennel et al. (2010), a double-blind pilot in children and teenagers that also found no objective benefit.

Caveats.

Full research notes for this preset

Researcher’s rationale (written before fact-checking; where it differs, the text above and the checked sources below take precedence): Per-ear values exactly as in Malandrone et al. 2022 (Eur Psychiatry conference abstract): 415 Hz right, 400 Hz left (15 Hz); placebo was 400 Hz in both ears. That abstract gives no session length, so 20 min comes from Kennel et al. 2010 (20 min, 3x/week for 3 weeks; its beat frequency is not in the retrieved abstract). This matches the registered NCT07793045 protocol (20-min sessions).

All caveats the researcher found (before fact-checking):

  • Do not present as ADHD treatment or symptom relief, and do not suggest replacing or changing medication. No controlled study showed objective improvement in ADHD symptoms or attention.
  • Malandrone 2022: ADHD-RS-5, mind-wandering and SART did not change. The ‘symptom severity’ claim in its conclusion contradicts its own results. It is a conference abstract with no sample size, and participants were already on medication.
  • Kennel 2010 (n = 20 children/adolescents, beats vs sham CD, 3 weeks): no group differences on TOVA or Color Trails.
  • Vorster 2020 (MSc thesis, n = 11 adults with ADHD; 200 Hz right with a 216-224 Hz sweep left, 30 min): beats worsened sustained-attention RT relative to medication, and the author reports silence did better than beats.
  • Alves-Castro 2025 (7 high-inattention vs 7 control men; 500/515 Hz in pink noise): beats not beneficial and rated ‘disturbing and stressful’. Participants were self-screened on the ASRS and not clinically diagnosed.
  • Mandapati 2026 combined 10 Hz beats with 10 Hz light flicker in VR, with no control group, so it is not binaural evidence. NCT07793045 (music with embedded beats) is still recruiting.
  • Benefits of white/pink noise for inattentive children (stochastic resonance; Soderlund and colleagues) come from noise, not binaural beats, and do not count as binaural evidence.
  • Consider whether to ship this preset at all. If shipped, keep the ‘experimental’ framing.

Sources (fact-checked; corrections applied):

Calm

200 Hz left · 209.3 Hz right · 9.3 Hz beat (alpha) · 10 min · Some evidence

Before medical and dental procedures, about 10 minutes of 9 to 10 Hz beats lowered anxiety compared with silence in several small, mostly unblinded trials. Compared with the same music without beats, or in healthy people, results are mixed.

Why these numbers. The tones are exactly those of Isik et al. (2017) and Salehabadi et al. (2024): 200 Hz left and 209.3 Hz right, played for 10 minutes before wisdom-tooth surgery.

Caveats.

Full research notes for this preset

Researcher’s rationale (written before fact-checking; where it differs, the text above and the checked sources below take precedence): The left/right tones are copied from Isik et al. 2017 (‘200Hz for the left ear and 209.3Hz for the right ear’) and Salehabadi et al. 2024 (200 Hz left, 209.3 Hz right). Both played it for 10 min after local anaesthetic, before wisdom-tooth surgery, against a no-intervention control, hence the 10-minute session (Menziletoglu 2021 also used 10 min, at 210/220 Hz). The 9.3 Hz beat is kept unrounded. Isik’s abstract does not mention any background sound; Menziletoglu explicitly used pure tones with no background music, which is what the app plays. Salehabadi’s text is ambiguous (‘instrumental music’ but also ‘frequency-only’).

All caveats the researcher found (before fact-checking):

  • Almost all supporting trials compared beats with silence or no intervention, usually unblinded. Expectation, rest and distraction can explain much of the effect.
  • Same audio with vs without beats: anxiety did not differ significantly in Wiwatwongwana 2016 (P=0.085) or Opartpunyasarn 2022 (P=.282), and Weiland 2011’s beat and no-beat soundtracks reduced anxiety by similar amounts. Padmanabhan 2005, Ratanalerdnawee 2025 and Parodi 2021 (a proprietary ‘DMSPS’ variant) found an added benefit.
  • Against different calming audio, beats were no better than classical music (Olcucu 2021) or 432 Hz-tuned music (Menziletoglu 2021).
  • Not every silence-controlled procedural trial was positive: Kim 2025 (1 Hz) and Akkurt 2026 found no anxiety benefit.
  • Healthy volunteers outside clinics: Chen 2025 (double-blind, 10 Hz at 345/335 Hz in pink noise, 15 min) found no effect on anxiety. Yang 2025 found 6, 10 and 25 Hz indistinguishable, and Hautus 2021 found no frequency ordering.
  • The pooled perioperative effect (Xiong 2025) mixes frequencies and music beds, with I2 over 90%. Its per-frequency results were not retrievable, so it does not show that 9-10 Hz specifically works. Shukla 2025 rated the dental evidence low certainty (GRADE).
  • Isik’s abstract reports within-group change only (beats p=0.001; control unchanged p=0.625). Neither Isik nor Salehabadi reports randomisation in the retrieved text.
  • A 6 Hz protocol (200/206 Hz, Madhumitha 2025) also reduced anxiety against no audio. Nothing shows that 9.3 Hz beats other bands for anxiety.
  • The Pain preset uses almost the same beat rate (10 Hz) on a different carrier. No retrieved study shows that carrier choice changes outcomes.
  • Not a treatment for anxiety disorders. The evidence covers acute, situational anxiety.

Sources (fact-checked; corrections applied):

Relax

196 Hz left · 203 Hz right · 7 Hz beat (theta) · 20 min · Some evidence

In a double-blind trial (21 people) of theta beats made by the Monroe Institute, people felt more relaxed after exercise than with placebo tones that had no beat. Heart-rate-variability measures changed differently during listening but no longer differed by the end. It has not been independently replicated.

Why these numbers. The 7 Hz beat and 20 minutes come from McConnell et al. (2014). Their Monroe-made stimulus wobbled around 7 Hz over chords and pink noise, and the paper gives no per-ear tones. 196 Hz left and 203 Hz right is a 7 Hz pair the Monroe Institute published for its own EEG research (Atwater). One steady pair is a simplification.

Caveats.

Full research notes for this preset

Researcher’s rationale (written before fact-checking; where it differs, the text above and the checked sources below take precedence): The 7 Hz beat and the 20 minutes come from McConnell et al. 2014. Their TMI-made stimulus swept 5.5-8.5 Hz (7 Hz plus or minus 1.5 Hz on a 4-s loop) over chord carriers in pink noise, and the paper prints no per-ear frequencies. The pair Left 196 Hz / Right 203 Hz is printed in Atwater’s TMI frequency-following-response study (2000, updated 2004). There it was one of two 7 Hz pairs played together (with 296/303 Hz) as 2-second bursts. Adaptation: one steady pair, with no 1.5 Hz wobble, no chord and no pink noise.

All caveats the researcher found (before fact-checking):

  • This is one pair taken from multi-tone mixes. McConnell used a 5.5-8.5 Hz wobble on several chord carriers in pink noise; Atwater played two 7 Hz pairs at once.
  • The per-ear numbers come from 2-second bursts in a 7-person EEG study of the frequency-following response, not from a relaxation study.
  • Only self-rated relaxation differed (p=0.036). The HRV difference had disappeared by the end of the 20 minutes, and baseline LF/HF already differed between sessions.
  • The sample was healthy college students resting after exercise. The Monroe Institute supplied the stimulus, so this is not independent of the source.
  • The idea that ‘theta = relaxation’ is a general claim. No Monroe source tested 196/203 Hz on its own for relaxation.
  • Monroe’s own patent does not tie 7 Hz to relaxation. Its ‘Relaxation’ Mood Minder spectrum is 1.5 + 4 + 6 Hz, and it lists 7 Hz under ‘Attention’ (4 + 7 + 16 Hz).

Sources (fact-checked; corrections applied):

  • McConnell PA, Froeliger B, Garland EL, Ives JC, Sforzo GA (2014). Auditory driving of the autonomic nervous system: listening to theta-frequency binaural beats post-exercise increases parasympathetic activation and sympathetic withdrawal. Frontiers in Psychology 5:1248. — rct, corrected after re-check.
    • Protocol: Double-blind, placebo-controlled, randomized and counterbalanced within-subjects crossover; n=21 healthy college students aged 18-29 (22 recruited, 1 did not finish). After 20 min of moderate exercise, 20 min of listening on Koss HQ1 headphones at 61-63 dB. The Monroe Institute supplied both tracks. Several carrier tones formed chords (C-major-7, C-major, G-major, D-minor, C-major, with 10-s cross-fades) over pink noise. The interaural difference was 7 Hz, varied by plus or minus 1.5 Hz over a 4-s loop (5.5-8.5 Hz). The placebo had the same carrier tones and pink noise but ‘no varying interaural phase difference’. The paper gives no per-ear carrier frequencies for its stimulus (only chord names); the 500 Hz left / 510 Hz right in its Introduction is a generic example, not the study stimulus.
    • Found: More self-reported relaxation with the beats than placebo (t=2.248, p=0.036, d=0.49), rated only after the session with no pre-session rating. HRV showed significant condition-by-time interactions for LF/HF (p=0.012), LF (p=0.003) and HF (p=0.034), meaning the trajectories differed: LF/HF rose under placebo but fell under beats from the start to the middle of listening (p=0.044), and the HF follow-up contrast was only a trend (p=0.089). The HRV levels themselves did not differ significantly between conditions at any time point per the Results text; Table 3 flags only lower LF with beats at the start of listening (p=0.013), and by the end nothing differed. Baseline LF/HF was already lower in the beats session (p=0.035), and HRV changes did not correlate with relaxation changes. One small study (n=21); the stimulus came from TMI.
    • “multiple carrier tones were presented over a background of pink noise (20–20,000 Hz with power attenuated in non-audible frequency ranges) with a 7 Hz interaural frequency difference that was continuously varied by plus or minus 1.5 Hz over a 4-s period”

  • Atwater FH (2000; updated 2004). Binaural beats and the frequency following response. The Monroe Institute, unpublished report (TMI research files on archive.org). — controlled study, corrected after re-check.
    • Protocol: N=7. 2-second bursts in a Latin square with placebo tones. 7 Hz mix: Left 296 / Right 303 Hz played together with Left 196 / Right 203 Hz. 16 Hz mix: 292/308 plus 192/208 Hz. Placebo: 300/300 and 200/200 Hz.
    • Found: Each condition was compared only with silence (Dunnett test after a one-way ANOVA, with the placebo tones entered as a separate group). The 7 Hz EEG amplitude during the 7 Hz beat was significantly above silence (N=7), but the overall ANOVA was not significant (F=2.46, p=0.087), and the 7 Hz beat was never tested directly against the placebo tones, which also raised 7 Hz EEG (means 0.171 vs 0.125 vs 0.069 microvolts for 7 Hz beat, placebo, silence). The 16 Hz rise during the 16 Hz beat was not significant in the Dunnett test (ANOVA p=0.126); placebo tones raised 16 Hz EEG almost as much (0.065 vs 0.055 vs 0.033). Simple t-tests against silence (2004 version only) were significant for both 7 Hz (p<.001) and 16 Hz (p=.007). The 2000 version concludes ‘no evidence of a 16 Hz FFR’. A tiny time-averaged auditory response, not a change of state. TMI in-house report with no peer-reviewed venue.
    • “Significant (p =< 0.05) increases in 7 Hz EEG amplitudes were, however, demonstrated during the 7 Hz stimulus condition, even when the increases in EEG during the alternative stimuli were considered.”

Meditation

250 Hz left · 256 Hz right · 6 Hz beat (theta) · 10 min · Some evidence

In a small EEG study (28 people, compared with silence), these exact tones raised theta activity typical of meditation within 10 minutes, but mood effects were mixed. An unblinded online trial found 5 minutes of 6 Hz beats improved overall self-rated mood more than pink noise or silence, with calmness changing most.

Why these numbers. The tones are exactly those of Jirakittayakorn & Wongsawat (2017): 250 Hz left, 256 Hz right. Theta activity peaked at 10 minutes, which the authors called the suitable listening time.

Caveats.

Full research notes for this preset

Researcher’s rationale (written before fact-checking; where it differs, the text above and the checked sources below take precedence): Per-ear values from Jirakittayakorn & Wongsawat 2017: 250 Hz carrier in the left ear, 256 Hz in the right, 65 dB SPL, over-ear headphones. They played it for 30 min, found theta power peaked at 10 min and wrote that ‘the suitable exposure duration was 10 min’, hence the 10-minute session. Amd 2026 used the same 250 Hz left carrier for 6 Hz but put the lower tone (244 Hz) in the right ear, for 5 min.

All caveats the researcher found (before fact-checking):

  • In the source EEG study only ‘worried’ improved. ‘Bad tempered’, ‘depressed’, ‘confused’ and ‘mixed up’ rose from before to after listening in the beat group only (within-group tests at a lenient p<0.1, never compared directly with the silence group), which the authors attribute to possible over-stimulation.
  • Low-frequency beats have worsened mood elsewhere. In Lane 1998, theta/delta beats (1.5 and 4 Hz) produced more negative mood than beta beats. In Chen 2025, 6 Hz beats trended toward more stress (p=0.06).
  • ‘Meditative state’ in Jirakittayakorn 2017 is inferred from an EEG pattern. Participants did not meditate, the control heard silence rather than a sham tone, and allocation is not described as randomised. The control group is reported as 11 people but listed as 4 females and 6 males.
  • Amd 2026 was self-administered online with no blinding and a single author. Its between-group result is on overall mood change. Study 1’s frequency effect was not significant on the full dataset (p=.091), and the ‘12 Hz’ group got 250/239 Hz (an 11 Hz difference).
  • Mechanism doubtful: in Ingendoh 2023’s review, 5 of 14 EEG studies supported entrainment and 8 contradicted it. Lopez-Caballero 2017 found no EEG or arousal change with 4.53 Hz beats.
  • In Lavallee 2011, 7 Hz beats during meditation changed EEG in experienced meditators but not novices. No trial shows that beats improve meditation practice outcomes.
  • 6 Hz is also the beat rate of the Zampi 2016 chronic-pain trial, whose carriers were not retrievable.

Sources (fact-checked; corrections applied):

  • Jirakittayakorn N, Wongsawat Y (2017). Brain Responses to a 6-Hz Binaural Beat: Effects on General Theta Rhythm and Frontal Midline Theta Activity. Front Neurosci 11:365. — controlled study, corrected after re-check.
    • Protocol: 250 Hz left / 256 Hz right (6 Hz), 65 dB SPL, over-ear headphones, 30 min seated; control heard silence (told a stimulus was playing); n=28 (17 beat, 11 control; allocation not described as randomised); QEEG, Brunel Mood Scale.
    • Found: Within the beat group, theta power rose above baseline at nearly every site except T4 at the 10-min mark. That was the time point with the most sites and the largest change. Compared with the silence group, the rise was significant only at F3, C3, F7, T3, T5, Fz, F8 and Cz at 10 min, and at frontal sites at 20 and 25 min; there was no difference at 5, 15 or 30 min. The authors liken the 10-min frontal/parietal-central plus Fz pattern to meditation. They say listening past 10 min did not raise theta further and may cause fatigue; by 30 min only Cz still differed from baseline. Mood (BRUMS): the authors judged within-group pre/post paired t-tests at a lenient p<0.1. On that basis, ‘worried’ went down and ‘bad tempered’, ‘depressed’, ‘confused’ and ‘mixed up’ went up in the beat group only. These items were never tested directly against the control group. Items that changed in both groups (annoyed, worn out, exhausted, sleepy, tired, lower vigor) did not differ between groups. The authors put the negative changes down to possible over-stimulation.
    • “The carrier tone of 250 Hz was presented to the left ear, and the offset tone of 256 Hz was presented to the right ear.”

  • Amd M (2026). Effects of self-administered binaural beats on meditative and introspective states. PLoS One 21(4):e0335580. — rct, corrected after re-check.
    • Protocol: Self-administered, unsupervised online web app built on the Web Audio API. Sessions were 5 min, through headphones worn by the participant. Volume was self-set after a 5-s ‘Test Audio’ of 250 Hz tones in both channels. The 250 Hz carrier was in the LEFT channel; the right channel played 247, 244, 241 or 239 Hz for the stated 3, 6, 9 or 12 Hz beats (250-239 = 11 Hz, though the paper calls it 12 Hz). Study 1 (n=101): 3/6/9/12 Hz, assigned at random by the tutorial program. Study 2 (n=118): 6 Hz beats only (n=34), 6 Hz beats + pink noise (n=34), pink noise only (n=28), and silence (n=22, headphones on with app volume at 0%). Participants were not blind to their condition: setup instructions differed by group and over 99% correctly confirmed their assigned condition. The researcher was blind to allocation until data analysis (stated for Study 1). Single author; not preregistered; mood rated on 9-point visual analogue scales before and after listening.
    • Found: Study 2: beat conditions gave larger overall mood change scores than pink noise or silence (post hoc p<.01), with calmness changing most; equivalence tests showed meaningful calm/focus gains within beat conditions. Study 1: frequency main effect not significant on the full data (p=.091); 6 Hz > 3 Hz (Tukey p=.015). Expectancy not controlled.
    • “The 250 Hz carrier frequency was delivered to the left stereo channel, with group-specific frequencies (247, 244, 241, or 239 Hz) delivered to the right channel to generate 3, 6, 9, or 12 Hz binaural beats, respectively.”

  • Ingendoh RM, Posny ES, Heine A (2023). 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 18(5):e0286023. — systematic review, verified.
    • Protocol: n/a (14 EEG studies, heterogeneous frequencies and methods).
    • Found: 5 studies supported brainwave entrainment, 8 contradicted it, 1 mixed.
    • “with five studies reporting results in line with the brainwave entrainment hypothesis, eight studies reporting contradictory, and one mixed results.”

  • Lopez-Caballero F, Escera C (2017). Binaural Beat: A Failure to Enhance EEG Power and Emotional Arousal. Front Hum Neurosci 11:557. — controlled study, verified.
    • Protocol: Binaural and acoustic beats at 4.53, 8.97, 17.93, 34.49 and 57.3 Hz in 3-min epochs between pink-noise epochs; EEG, heart rate, skin conductance. Abstract only.
    • Found: No change in the matching EEG bands and no heart-rate or skin-conductance change at any beat frequency.
    • “For all the beat frequencies used for stimulation, no significant changes between Baseline and Beat epochs were observed within the corresponding EEG bands, neither with binaural or with acoustic beats.”

  • Lane JD, Kasian SJ, Owens JE, Marsh GR (1998). Binaural auditory beats affect vigilance performance and mood. Physiol Behav 63(2):249-252. — controlled study, verified.
    • Protocol: Pink noise with simple tones, or beta (16 and 24 Hz) or theta/delta (1.5 and 4 Hz) binaural beats, during a 30-min vigilance task on 3 days; n=29, blind to beats. Abstract only.
    • Found: Beta beats gave better vigilance and less negative mood than theta/delta beats.
    • “In addition, the beta-frequency beats were associated with less negative mood.”

  • Chen B, Wong C, Dzebley R, Stone JM (2025). Effect of Binaural Beats on Affective Symptoms and Performance on the Digit Span Test. Cureus 17(10):e94182. — rct, verified.
    • Protocol: Same trial; theta arm 343 Hz left / 337 Hz right in pink noise, 15 min, vs 340 Hz in both ears.
    • Found: Theta showed a trend toward increased stress (p=0.06); no anxiety effect.
    • “The theta BB audio file consisted of a 343 Hz wave on the left side and a 337 Hz wave on the right side, generating a 6 Hz theta BB.”

  • Lavallee CF, Koren SA, Persinger MA (2011). A quantitative electroencephalographic study of meditation and binaural beat entrainment. J Altern Complement Med 17(4):351-355. — controlled study, corrected after re-check.
    • Protocol: Novice and experienced meditators meditated alone, with 7 Hz (‘facilitative’) or 15 Hz (‘hindering’) beats; QEEG. Carriers and n not in abstract. Abstract only.
    • Found: From the abstract only. With 7 Hz (‘facilitative’) beats, experienced meditators showed increased left temporal delta power (the abstract’s Conclusions call it theta); novices did not show this effect. With 15 Hz (‘hindering’) beats, novices showed more gamma power than experienced meditators relative to baseline, and could not maintain occipital theta power. The study measured EEG power only, not how deep meditation was, so it does not test whether beats deepen meditation for beginners.
    • “Experienced meditators displayed increased left temporal lobe δ power when the facilitative binaural beats were applied, whereas the effect was not observed for the novice subjects in this condition.”

Acute pain

450 Hz left · 440 Hz right · 10 Hz beat (alpha) · 10 min · Some evidence

Small trials found 10 Hz beats lowered short-term pain in a lab heat-pain test and during some procedures, but reviews rate the evidence low to very low quality. In the trial these exact tones come from, beats in music did no better than the same music.

Why these numbers. Reviews single out alpha (about 10 Hz) for acute pain. The tones are from Aly et al. (2023), whose figure puts 450 Hz in the left ear and 440 Hz in the right; that trial embedded them in music for 5 minutes. The 10 minutes follows the lab and surgery protocols, so the combination matches no single tested protocol.

Caveats.

Full research notes for this preset

Researcher’s rationale (written before fact-checking; where it differs, the text above and the checked sources below take precedence): Band: the Maddison 2023 meta-analysis and the Shamsi 2024 review single out alpha entrainment for acute pain, matching the 10 Hz protocols of Ecsy 2017, Olcucu 2021 and Loong 2022, none of which gave carriers in retrievable text. Per-ear tones are from Aly 2023, the only pain trial whose retrieved text states both tones: ‘440 Hz in one ear and 450 Hz in the other’. The text does not say which ear; fact-checking found that the paper’s Figure 2 puts 450 Hz in the left ear and 440 Hz in the right, which the app uses. Aly played 5 min daily inside music. The 10-minute session instead follows Ecsy (10 min per frequency before pain testing) and Loong (listening started 10 min before surgery).

All caveats the researcher found (before fact-checking):

  • In Aly 2023, the source of these exact tones, beat music did no better than identical music without the beat on any pain score. Both did better than no music only late in the week.
  • Shamsi 2024: risk of bias mostly high, evidence low to very low, no meta-analysis possible. Maddison 2023 pooled visual and auditory entrainment (I2=73%), and its ‘alpha for acute pain’ conclusion rests on 2 auditory studies (Ecsy, Olcucu).
  • In Olcucu 2021, patients tolerated the pure-beat arm less well than classical music or silence.
  • Sham-controlled 4 Hz trial (Tani 2021: 256/260 Hz vs 256/256 Hz): morphine use halved but pain ratings did not differ.
  • Ecsy 2017 tested healthy volunteers with brief laser heat pain. The auditory reduction was about 0.5 NRS points (per Maddison 2023).
  • Chronic-pain trials used theta (5-6 Hz), not alpha: Zampi 2016 and Gkolias 2020 positive, Thanyawinichkul 2022 null. A 2026 theta meta-analysis gives SMD -0.53 (low certainty, 4 trials, n=173). Shamsi’s recalculation found no chronic-pain effect.
  • Not a substitute for medical pain care. New, severe or unexplained pain needs clinical assessment.

Sources (fact-checked; corrections applied):

Lucid dreams

200 Hz left · 204 Hz right · 4 Hz beat (delta/theta border) · 30 min · Reported experiences

No study has tested binaural beats on lucid dreaming, but a listener testimonial published by the Monroe Institute describes drifting into one. The popular ‘40 Hz’ claim comes from electrical brain stimulation, not sound, and a replication found sham worked as well. The methods with evidence are mental techniques such as MILD and wake-back-to-bed.

Why these numbers. No source gives a binaural frequency for lucid dreaming. 200 Hz left and 204 Hz right is the first pair of an example signal in Monroe’s 1994 sleep-device patent; using one pair from that layered signal is our adaptation. 30 minutes covers falling back asleep after a wake-back-to-bed alarm.

Caveats.

Full research notes for this preset

Researcher’s rationale (written before fact-checking; where it differs, the text above and the checked sources below take precedence): No study, patent or Monroe Institute document we retrieved gives a binaural-beat frequency for lucid dreaming. TMI’s Lucid Dreaming program cites only ‘specific sound frequencies’. The closest documented binaural protocol aimed at dreaming sleep is Monroe’s US patent 5,356,368 ‘sleep processor’. It repeats a 90-minute cycle ‘through alpha, theta, delta, and back to dreaming sleep’, but describes the cycle stages only by EEG band and gives no Hz for the dreaming phase. Its only explicit numbers are a generic ‘septon’ example (left 200/204/208 Hz, right 204/208/212 Hz: three 4 Hz binaural plus four monaural 4 Hz beats). We took the first binaural pair, 200 Hz left / 204 Hz right. Reducing the septon to one pair is our adaptation, not Monroe’s. The preset is intended as a quiet backdrop while practising wake-back-to-bed + MILD, the combination with controlled support (ILDIS: alarm 5 h after bedtime, then the technique). The 30-minute length has no lucid-dream source. We chose it to cover the return-to-sleep window (in ILDIS, falling asleep within 10 min of the technique predicted success), and it matches the low end of TMI’s 30-45 min exercises.

All caveats the researcher found (before fact-checking):

  • No primary source publishes a binaural frequency for lucid dreaming. 200/204 Hz is one pair taken from a generic example in a Monroe sleep patent. It is an adaptation, not a tested protocol, and 4 Hz is not linked to lucidity anywhere.
  • ‘40 Hz gamma for lucid dreams’ traces to Voss et al. (2014), which used transcranial alternating current (scalp electrodes, 30 s bursts in REM sleep), not sound. A replication (Blanchette-Carrière et al. 2020) found lucid dreams in about the same share of sham naps (17.4% vs 18.5%, per Carr’s 2020 summary).
  • The EEG basis for 40 Hz is also contested. Baird, Tononi & LaBerge (2022) found the reported 40 Hz increase in lucid REM sleep disappears once eye-movement artifacts are removed. A 2025 multi-lab analysis found minimal sensor-level EEG differences between lucid and non-lucid REM.
  • ‘Theta for lucid dreams’ has no lucid-dream study behind it. It comes from linking theta with drowsiness and stage-1 sleep (Monroe’s 1975 patent describes stage-1 sleep EEG as theta, 4-8 cps) and with Focus 10 ‘mind awake, body asleep’.
  • Lucid dreams happen mostly in REM sleep, typically later in the night. A beat played while you fall back asleep is not playing during those REM periods. We found no data on whether beats during sleep help or disturb REM or dream recall.
  • The sound-based methods that have actually been tested use cues during REM (ring tones, music, or sounds paired with training, i.e. targeted lucidity reactivation), not binaural beats. Success has been modest and variable.
  • Keep the volume low and stop if it keeps you awake. In ILDIS, falling asleep quickly after the technique predicted success. Headphones are required for a binaural beat; consider whether you can sleep comfortably in them.
  • 4 Hz is on the delta/theta border; the same Monroe patent defines theta as 6-8 Hz and delta as 1-4 Hz.

Sources (fact-checked; corrections applied):

Out-of-body

100 Hz left · 104 Hz right · 4 Hz beat (delta/theta border) · 30 min · Reported experiences

No controlled study has tested binaural beats on out-of-body experiences, but people describe them first-hand, for example in testimonials published by the Monroe Institute. We found no Monroe Institute publication giving frequencies for Focus 10 or its out-of-body exercises, and its recordings layer several tone pairs with pink noise and spoken guidance.

Why these numbers. 100 Hz left and 104 Hz right is the example in Monroe’s 1993 patent (the left/right split is from a 1990 Monroe Institute newsletter). The same pair appears in an anonymous amateur measurement of the Gateway ‘Focus 10’ recording. 30 minutes is the short end of the Gateway exercises.

Caveats.

Full research notes for this preset

Researcher’s rationale (written before fact-checking; where it differs, the text above and the checked sources below take precedence): We found no Monroe Institute source (2015 e-book, Gateway manual, Wave I booklet, current FAQ/research pages, an undated MIAS bulletin) that gives a Hz value for Focus 10 or for the out-of-body exercise. TMI says only that it stacks, blends and sequences binaural-beat frequencies. The pair 100 Hz / 104 Hz is copied from the generic example in Monroe’s US patent 5,213,562, which names no ears; left = 100, right = 104 comes from TMI’s 1990 newsletter. That patent never mentions Focus 10 or OBEs, but it is the only Monroe-authored document with a concrete pair. The same pair is one of three simultaneous pairs (l/r 100/104, 300/304, 496/500) in an anonymous, unverified USENET measurement of the Gateway ‘Exploration Sleep (Focus 10)’ tape. That is the exercise TMI’s manual links to leaving the body; the measurement is reproduced in SBaGen’s focus.txt. That file contradicts itself: its summary table gives Focus 10 as 100[1.5], 200[4.0], 250[4.0], 300[4.0] in centre[beat] notation, i.e. a 1.5 Hz beat near 100 Hz plus 4 Hz beats centred on 200/250/300 Hz (so ‘200[4.0]’ = 198/202 Hz). A separate amateur analysis found about 1.6 Hz on ~107.7 Hz in ‘Advanced Focus 10’. Using one pure-tone pair is our simplification of a multi-layer recording. 30 min is the low end of the ‘30- to 45-minute exercises’ in TMI’s Wave I booklet and matches the 30-min Hemi-Sync exposure in Johnson & Persinger (1994).

All caveats the researcher found (before fact-checking):

  • No primary Monroe Institute source publishes Focus 10 (or Focus 12) beat frequencies. 100/104 Hz is a generic patent example that agrees with one line of an anonymous amateur measurement, and that measurement contradicts itself. It is not a tested protocol.
  • Gateway/Hemi-Sync recordings layer several beat pairs over pink noise or surf, with spoken relaxation, affirmations and ‘rolling out’ instructions. A single sine pair is not a Hemi-Sync recording, and OBEs reported by Gateway users cannot be credited to the beat itself.
  • Johnson & Persinger (1994) found stronger ‘sensed presence’ with a Hemi-Sync tape than with two other active conditions. But the outcome was sensed presence, not an OBE, and the abstract reports neither frequencies nor any sham-audio or silence control (full text not retrieved). It does not change this preset’s grade.
  • Fabrizi & Ditye (2025; n = 12) found the Gateway Experience improved psychological well-being compared with a relaxation control; participants’ sense of mysticality showed no effect on the result. The abstract reports no out-of-body outcome, and the guided content cannot be separated from the beats.
  • In a 2026 lab study (n = 35, meditation plus light stimulation, no binaural beats), unusual bodily experiences including OBEs came with more beta/gamma and less delta/theta EEG activity than the state each arose from, but (per the authors’ preprint) still more delta and occipital theta than eyes-closed wakefulness. It did not test whether theta beats cause OBEs.
  • Theta binaural beats as an altered-state inducer did not replicate: a double-blind study (Stevens et al. 2003) found no increase in frontal theta or hypnotic susceptibility after 4 h of theta-beat training, contradicting a 6-person pilot (Brady & Stevens 2000).
  • 4 Hz is on the delta/theta border. Monroe’s 1994 patent calls 1-4 Hz delta and 6-8 Hz theta, while his 1975 patent describes stage-1 sleep theta as 4-8 cps.
  • A 1983 US Army assessment quotes Monroe second-hand saying the OBE tape used ‘Beta signals of around 2877.3 CPS’. That is not a brainwave-range beat and is not usable. Its ~7 Hz figure is the author’s speculation, and its claim that REM sleep plus Hemi-Sync ‘could guarantee’ OBEs is an unverified anecdote.
  • Stereo headphones are required for a binaural beat. Use lying down in a quiet room, never while driving.

Sources (fact-checked; corrections applied):

  • Monroe RA (1993). US Patent 5,213,562: Method of inducing mental, emotional and physical states of consciousness, including specific mental activity, in human beings. Interstate Industries Inc. — patent, corrected after re-check.
    • Protocol: Patent, not a study. Generic example: a 100 Hz carrier in one channel and a 104 Hz carrier in the other give a 4 Hz binaural beat; says up to 100 carrier pairs may be used; names 275 Hz as the ‘most effective harmonic carrier’ for a 4 or 5 Hz beat; ‘Phased Pink Sound’ mixed at least 10 dB below the beat signals.
    • Found: No controlled data: the only examples are EEG figures of individuals (FIG. 3), said to show frequency-following at 0.5-7 Hz (FIG. 3B: 1.5, 4 and 6 Hz; 3D: 2, 4 and 7 Hz; 3F: 0.5, 3 and 4 Hz; 3H: 1.5, 2 and 4 Hz). The text never mentions Focus levels, dreams, lucid dreaming or out-of-body experiences (0 occurrences each), so the 100/104 Hz pair is an illustration, not an OBE protocol.
    • “If a carrier frequency of 100 Hz were employed in one channel of the audio signal, and a carrier frequency of 104 Hz were employed in the other channel, a binaural beat of 4 Hz would result.”

  • The Monroe Institute (2015). Gateway to the Human Mind: Going Out of Body and Other Expanded States (e-book). The Monroe Institute. — institutional, verified.
    • Protocol: n/a
    • Found: TMI says it stacks, blends and sequences binaural-beat frequencies and defines Focus 10 as ‘Mind Awake/Body Asleep’ (‘the level at which the physical body is asleep but the mind is awake and alert’), but publishes no Hz value for any Focus level and presents no data.
    • “By using FFR and binaural beat specific frequencies, then stacking, blending, and sequencing them”

  • Monroe Products / Interstate Industries (n.d.). The Gateway Experience (home-training manual). Copy hosted on archive.org. — institutional, verified.
    • Protocol: Spoken-guidance exercises over ‘Hemi-Sync’ sound with a ‘pink sound’ bed; stereo headphones; darkened room; ‘about 45 minutes’. Discovery #5 ‘Exploration, Sleep’ is the exercise the manual’s FAQ links to leaving the body (‘What if I can’t get out of my body with Discovery #5?’).
    • Found: No beat frequencies or carrier values are published; the exercises combine sound with verbal relaxation, affirmations and visualisation, so any reported OBE cannot be attributed to the beats alone.
    • “Complex blends of Hemi-Sync sound frequencies facilitate different states of consciousness as you proceed through the Gateway Experience.”

  • Monroe Products (2004). The Gateway Experience, Wave I - Discovery (booklet; voiced by Robert A. Monroe). Monroe Products. Copy hosted on archive.org. — institutional, corrected after re-check.
    • Protocol: 30- to 45-minute exercises over stereo headphones; volume set so the words are barely audible; a ‘pink noise’ or ‘pink sound’ bed under the Hemi-Sync signals (‘provided as a bed for the Hemi-Sync® signals’).
    • Found: Delivery guidance only; no frequencies and no outcome data.
    • “Locate a distraction-free environment where you can darken the room and remain undisturbed for the 30- to 45-minute exercises.”

  • The Monroe Institute (n.d.). Monroe Institute Technologies Feedback: testimonials about Hemi-Sync from the late 1990s, sorted by topic (‘Meditation & Spiritual Growth’; ‘Expanded Awareness’). Posted on archive.org by the Monroe Institute in 2020. — institutional, verified (added 17 November 2026).
    • Protocol: Testimonial collection, not a study. Each page says the entries were reported ‘by individuals and/or by professional practitioners about the use of Hemi-Sync® in the late 1990s’. It gives no counts, and the wording was edited: one testimonial appears under two topics with the product named differently.
    • Found: Two entries describe out-of-body states first-hand. One listener says they could ‘astral travel without fear’ within a month of daily use of the Gateway Experience CDs (a guided home course); another entry describes altered states ‘separate from my physical body’ while listening to Metamusic (music with Hemi-Sync tones). With no counts, it shows that such reports exist, not how common they are.
    • “I never dreamed that within a month of beginning the series and using the CDs every day, I would not only astral travel without fear, but would increase my clairaudience and clairvoyance triple fold.”

  • Peters J (ed.) (n.d.). SBaGen 1.4.5 documentation, focus.txt: ‘Monroe Institute Focus Levels’ (reproduces an anonymous USENET measurement of Gateway tapes). — other, corrected after re-check.
    • Protocol: SBaGen’s author says the values come from ‘a document that appeared on USENET, from someone who appears to have measured the frequencies used on some of the Monroe Institute tapes’. Anonymous, unverified measurement of commercial Gateway tapes. Per-tape lines give explicit left/right pairs, e.g. Wave I Tape 5 ‘Exploration Sleep (Focus 10)’: l/r 100/104, 300/304, 496/500 (three 4 Hz beats at once); Tape 4: l/r 100/104, 414/410, 504/500. The summary table uses centre[beat] notation (beat split around the centre): F10 = 100[1.5], 200[4.0], 250[4.0], 300[4.0]. Tones said to sit ~20 dB above pink noise and/or surf; amplitudes not measured.
    • Found: A widely reproduced source of online ‘Focus 10’ numbers (the source itself does not show how widely it is used). It is internally inconsistent: the Focus 10 per-tape lines list 4 Hz beats on ~100/300/500 Hz carriers, while the summary table gives F10 = 100[1.5], 200[4.0], 250[4.0], 300[4.0]. It gives no values for Wave I Tape 3 (Advanced Focus 10), says amplitudes were not measured, and is not a Monroe Institute publication.
    • “Tape 5 - Exploration Sleep (Focus 10) Remembering who and what you are, Rolling out, Floating upwards l/r 100/104, 300/304, 496/500”

  • SBaGen 1.4.5 documentation, focus2.txt: ‘BAVSA analysis of some Monroe recordings’ (e-mail from an anonymous user to J. Peters). — other, corrected after re-check.
    • Protocol: Amateur spectral analysis (BAVSA software) of Hemi-Sync/Gateway audio files; e.g. ‘Wave 1 Discovery #3 Advanced Focus 10’ at 16:18: 107.70+1.60 (about a 1.6 Hz beat on a ~107.7 Hz carrier).
    • Found: For ‘Wave 1 Discovery #3 Advanced Focus 10’, during the count from 1 to 10 into Focus 10 (16:18) and again at 20:08, the writer reports only a ~1.55-1.6 Hz beat near 107.5-107.7 Hz. Reading the ‘a+b’ notation as carrier + beat is an interpretation; the file never defines it. The writer also says binaural beats were hard to find in Hemi-Sync files. focus.txt gives no values for this tape. The result does not show the 4 Hz beats in focus.txt’s other Focus 10 tape lines, and only loosely resembles the summary table’s 100[1.5] component (similar beat, different carrier). The anonymous writer states that the Institute keeps its frequencies secret.
    • “For that reason i am “decompiling” Hemi-Sync audio files, because The Monroe Institute doesn’t show what frequencies uses in each CD and have them as a secret.”

  • McDonnell WM (1983). Analysis and Assessment of Gateway Process. US Army Operational Group, INSCOM, 9 June 1983. CIA-RDP96-00788R001700210016-5 (declassified 2003). — institutional, corrected after re-check.
    • Protocol: n/a (non-experimental assessment by an officer who attended Gateway training)
    • Found: Non-experimental assessment. In the OCR text of this copy it gives no Hz for any Focus level: Focus 12 and 15 are described qualitatively and Focus 10 is named only once, in passing. The only Hemi-Sync tape frequency it ties to OBE is a second-hand magazine quote of Monroe (‘around 2877.3 CPS’, far above any brainwave range). It also lists ‘resonance at approximately 7 Hertz’ as one of three conditions for an out-of-body state. That figure is the author’s speculation built on Itzhak Bentov’s model of body micromotion resonating with the Earth’s ionospheric cavity (6.8-7.5 Hz); it is not a measured or tape frequency. It relays Monroe’s unverified anecdote that an ex-trainer ‘could guarantee out-of-body movements by bringing participants down into a rapid eye movement(REM) state of sleep and then use the Hemi-Sync tape technique’.
    • “To facilitate achieving the out-of-body state, Bob Monroe, the founder of Monroe Institute, is quoted in a recent magazine article as saying that in order to assist the participant the particular Hemi-sync tape concerned with that technique employs Beta signals of “around 2877.3 CPS.”(cycles per second).”

  • Wilson ES, Anderson T (1991). A Psychophysiological Study of the Hemi-Sync Process. Hemi-Sync Journal (The Monroe Institute) 9(4), Fall 1991. Summary of a presentation. — uncontrolled study, verified (added 17 November 2026).
    • Protocol: 9 people (4 ‘adept’ and 5 ‘naive’ subjects) lay on a waterbed in a sound- and light-proof chamber with 24-channel EEG, through silence, pink noise alone, a fixed sequence of four beat combinations in pink noise (Monroe’s Focus 10, 12, 15 and 21), then silence. Experiences were collected at debriefings. Subjects were told the content of the experiment beforehand, and the naive subjects were retested after the Gateway Voyage program. No beat frequencies are given, and no spoken guidance is described.
    • Found: The researchers’ impression was that five subjects ‘seemed to retain rather complete recall of being out-of-body and other psychical content’. Out-of-body reports are not counted separately, and the summary does not compare experiences with the pink-noise stage, so the design cannot show whether the beats caused them. It does not change this preset’s grade.
    • “The five showing bilateral activation seemed to retain rather complete recall of being out-of-body and other psychical content of experience.”

  • Johnson CP, Persinger MA (1994). The sensed presence may be facilitated by interhemispheric intercalation: relative efficacy of the Mind’s Eye, Hemi-Sync Tape, and bilateral temporal magnetic field stimulation. Percept Mot Skills 79(1 Pt 1):351-354. PMID 7991330. — controlled study, verified.
    • Protocol: Experiment 1: 30 university men, tested singly in an acoustic chamber, received 30 min of a Hemi-Sync tape (‘binaural, complex stimulation’), The Mind’s Eye (synchronised binaural + visual patterns) or transtemporal magnetic fields; experiences rated by questionnaire afterwards. Experiment 2: 18 women given the Hemi-Sync tape. Abstract reports no Hz values and no sham-audio or silence condition.
    • Found: Hemi-Sync group reported more intense ‘sensed presence’ and ‘ego-alien intrusions’ than the other two groups. Outcome was sensed presence, not out-of-body experience; ‘binaural’ here means a commercial tape with unknown content.
    • “The men who received the binaural stimulation (Hemi-Sync Tape) reported more intense sensed presences and ego-alien intrusions than did the other two groups.”

  • Fabrizi P, Ditye T (2025). The Gateway Experience Facilitates Psychological Well-Being. Integr Psychol Behav Sci 59(1):34. PMID 40080317. — controlled study, corrected after re-check.
    • Protocol: 12 meditation-naive participants exposed to the Gateway Experience recordings (hypnosis, meditation and binaural beats combined) and to a relaxation control over four weeks; Ryff well-being scale at several time points; mystical experience (MEQ30). Beat frequencies not reported in the abstract.
    • Found: Psychological well-being (Ryff PWS, all six scales) improved significantly more over time in the Gateway condition than in the relaxation control (n = 12). Mysticality (MEQ30), suggested as a possible influence on Gateway, showed no effect. The abstract reports only the Ryff PWS and MEQ30; no OBE outcome is reported. The design cannot attribute the effect to binaural beats rather than the guided hypnosis and meditation content.
    • “Results of a two-way repeated measures multivariate analysis of variance showed that participants’ psychological well-being improved significantly over time in the GE condition compared to the control condition on all measured scales.”

  • Campillo-Ferrer T, Iadarola A, Cordani R, Veneruso M, Demirel C, Nobili L, Oostenveld R (2026). Facilitating unusual bodily experiences and out-of-body experiences across wakefulness and sleep: A high-density EEG and neurophenomenology study. Conscious Cogn 139:104002. PMID 41678848. — uncontrolled study, corrected after re-check.
    • Protocol: N = 35 healthy adults in a sleep lab; meditation and light stimulation (no binaural beats) to facilitate unusual bodily experiences (UBEs) incl. OBEs; high-density EEG; eye-movement signalling.
    • Found: 20 of 35 participants reported 36 UBEs, mostly during meditation (wakefulness) but also in sleep arousals, REM and NREM sleep. Published abstract: exploratory EEG analyses linked UBEs to ‘EEG reactivation’ (more beta/gamma, less delta/theta), suggesting intermediate states that combine features of wakefulness and sleep. Per the PsyArXiv preprint (v2, 2025), these changes are relative to the baseline state each UBE arose from; compared with eyes-closed wakefulness, UBEs still showed more delta (and occipital theta) and less alpha. No binaural beats were used, and the study does not test whether theta beats induce OBEs.
    • “UBEs across sleep and wakefulness were associated with EEG reactivation, marked by increased high-frequency activity (beta and gamma) and decreased low-frequency activity (delta and theta)”

  • Stevens L, Haga Z, Queen B, Brady B, Adams D, Gilbert J, Vaughan E, Leach C, Nockels P, McManus P (2003). Binaural beat induced theta EEG activity and hypnotic susceptibility: contradictory results and technical considerations. Am J Clin Hypn 45(4):295-309. PMID 12722933. — controlled study, verified.
    • Protocol: Double-blind, repeated-measures group design; low and moderate hypnotic-susceptibility participants; 4 hours of theta binaural-beat training (Hz not given in abstract).
    • Found: Null: no specific increase in frontal theta or in hypnotic susceptibility; theta-beat effect on EEG judged very low power.
    • “results were not supportive of the specific efficacy of the theta binaural beat training employed in this study in either increasing frontal theta EEG activity or in increasing hypnotic susceptibility.”

  • Brady B, Stevens L (2000). Binaural-beat induced theta EEG activity and hypnotic susceptibility. Am J Clin Hypn 43(1):53-69. PMID 10911677. — uncontrolled study, verified.
    • Protocol: 6 participants; three 20-min sessions of a theta-enhancing binaural-beat protocol; multiple-baseline time series (Hz not given in abstract).
    • Found: Percent theta rose in 5 of 6; hypnotic susceptibility rose in low/medium-susceptible participants. Not replicated by the larger double-blind study (Stevens et al. 2003).
    • “The protocol designed to increase anterior theta activity resulted in a significant increase in percent theta for 5 of 6 participants.”

  • Monroe RA (1994). US Patent 5,356,368: Method of and apparatus for inducing desired states of consciousness. Interstate Industries Inc. — patent, verified.
    • Protocol: Patent, not a study. ‘Sleep processor’ that ‘repeats a 90 minute cycle of sound that leads the subject through alpha, theta, delta, and back to dreaming sleep’, using signal groups described only by EEG band (replicated EEG waveforms with dominant alpha, theta, delta, lower-delta values), Phased Pink Sound 10-20 dB below and soft voice inserts, then a wake-up group (beta-range EEG replicas or a ‘400 Hz/416 Hz envelope’). Its only explicit carrier example is a ‘septon’: left 200/204/208 Hz, right 204/208/212 Hz (three 4 Hz binaural beats plus four 4 Hz monaural beats). Defines theta as ‘6-8 Hz’ and delta as ‘1-4 Hz’.
    • Found: No controlled data and no measurement of dreams or lucidity; illustrative EEG maps from an individual only. It publishes no beat frequency for the ‘dreaming sleep’ part of the cycle; the septon is a generic example of the device’s carrier spacing.
    • “200 Hz carrier (4 Hz binaural beat) 204 Hz carrier”

  • Monroe RA (1975). US Patent 3,884,218: Method of inducing and maintaining various stages of sleep in the human being. Assignee listed as Monroe Ind Inc. — patent, verified.
    • Protocol: A familiar repetitive sound (e.g. surf) amplitude-modulated by an EEG sleep pattern; volume set to mask ambient noise. Not binaural (the word does not appear).
    • Found: Background text classifies stage-1 sleep EEG as theta 4-8 cps and links stage-1 REM to dreaming; it gives no dream-induction frequency and no data.
    • “The first stage of sleep or state 1 has an EEG pattern, as shown in FIG. 4, which consists of an irregular mixture of theta waves which are-low in amplitude with a frequency of 4 to 8 cps”

Considered and left out

Safety

What we couldn’t verify

How this research was done

In November 2026, we did a deep research about: Monroe Institute primary sources; hearing science and safety; sleep; focus and ADHD; anxiety, meditation and pain; creativity, mood and energy; lucid dreams and out-of-body experiences; and every systematic review and meta-analysis from 2015 to 2026. They read primary texts through PubMed’s E-utilities, Europe PMC full text, Google Patents and TMI’s own archive.org uploads, 1,052 tool calls in all (see research/raw/README.md), and were told to report “no primary source publishes this number” rather than fill gaps.

Separate fact-checking agents then re-fetched the 157 sources behind the presets and the research notes, and checked each citation, quote, frequency and finding against the source text: 99 were confirmed as written, 57 were real but needed a correction (applied here), 1 could not be checked (not used for any preset), and none were refuted. One more source cited only in this document (Klichowski et al. 2023) was checked against its full text by hand. Two sources were added on 17 November 2026, after a further search for first-hand reports of out-of-body experiences and lucid dreams: a 1991 study in the Monroe Institute’s journal and a Monroe Institute testimonial collection. A separate agent re-fetched both and confirmed their quotes, which were then checked again by hand against the downloaded text (see research/raw/verification/reported-experiences.json). An independent reviewer re-graded the presets (all but Relax, which was added afterwards and graded with the same rules), which led to the stricter grading rules above, the “No clear benefit” grade, and several presets being renamed, re-scoped or dropped.

Grades are a judgment about the evidence, not a guarantee. If you find an error, the raw research and verification data are in research/raw/.