3 Hz bedtime beats vs no sound over 4 weeks
Q: Did a 3 Hz bedtime binaural beat improve sleep?
A: A 2026 randomized trial of 66 college students found that 30 minutes of a 3 Hz binaural beat with music, five nights a week for four weeks, lowered sleep-quality scores more than no sound by the second week. By week 4 the same bedtime sound also cut perceived stress more than pink noise, and that gap was still there two weeks after the listening stopped.
Who took part
Sixty-six college students at a boarding school in Luoyang took part. Each already had a sleep-quality score above 7, which on this questionnaire means poor sleep. They lived in dormitories. Most were women. A computer split them into three groups of 22. The people who scored the questionnaires did not know who had heard which sound. Everyone came back for every check. Nobody dropped out.
What they listened to
For four weeks, five nights a week, two groups put on headphones for 30 minutes before bed. One group heard a 3 Hz binaural beat laid over instrumental music. That beat was a 250 Hz tone in one ear and a 253 Hz tone in the other. The other listening group heard pink noise over the same music. The third group heard no added sound and kept their usual night. Two weeks after the headphones stopped, everyone was checked again.
What changed in four weeks
By the second week, the 3 Hz group’s sleep-quality score was already 0.91 points lower than the no-sound group. The range around that gap was 0.12 to 1.70 points. Lower on this score means better sleep. At week 4 the gap was 1.23 points. Two weeks after the listening stopped, it was 1.32 points.
Pink noise did not separate from no sound on that sleep score at week 2 or week 4. A smaller gap showed up only at the week-6 check. The 3 Hz group and the pink-noise group did not separate from each other on the sleep score at any check.
The stress score told a clearer story. At week 4, and again at week 6, the 3 Hz group was 2.55 points lower than pink noise. The range around that gap was 0.69 to 4.40 points. At those same checks the 3 Hz group was also about 2.09 points lower than no sound. Pink noise did not separate from no sound on the stress score at any check.
What stayed flat
Morning saliva cortisol, a stress hormone, moved a little over the six weeks in all three groups. The groups did not separate from one another. The paper’s check for a different pattern by group was p = 0.701. The sleep score and the stress score moved. The morning hormone sample did not show a matching group difference.
What this leaves open
This was one boarding school, with daily dorm checks and a payment at the end. That helps explain why nobody left. It may not travel to a home where nobody is checking. Listening time was what people wrote down the next day, not a record from the headphones. Sleep was a questionnaire, not an overnight recording. People who start with a high sleep score can drift down a little on their own. The 3 Hz group still fell further than the group that heard nothing.
Read the paper, then try a session
If you want to sit with a 3 Hz tone before bed, open the 3 Hz row in the library and listen through headphones. Thirty minutes is the length this trial used. Tesla’s coil and the 3-6-9 line are the old work. The headphones and the app are how you use that work tonight.
The paper is Hou, Wang, and Yue, Frontiers in Psychology, 17 June 2026: The impact of sound intervention on sleep quality and stress levels in college students.
KYMA is working to establish a science-based podcast, because reliable information about this topic is almost impossible to find on ordinary social media.
| Result | What the paper reported |
|---|---|
| Sleep vs no sound | By week 2 the 3 Hz group was 0.91 points lower than no sound (95% CI 0.12 to 1.70 points lower). At week 4 the gap was 1.23 points. At week 6 it was 1.32 points. |
| Stress vs pink noise | At week 4 and week 6 the 3 Hz group was 2.55 points lower than pink noise (95% CI 0.69 to 4.40), and about 2.09 points lower than no sound. |
| Morning cortisol | The groups did not separate. The check for a different pattern by group was p = 0.701. |
Related tones in the KYMA library
Open these library sets if you want a stored program for the same sleep and stress topic. The 3 Hz row already stores the bedtime beat. The other rows match sleep or stress. This trial did not test them. A magnetic-field setting is not stored in the tone.
3 Hz frequency promotes deep relaxation, enhances sleep quality, and supports stress reduction for overall mental well-being. · 5 minutes
Targets the delta frequency range associated with natural growth hormone release during deep sleep. Supports tissue repair and anti-aging. HEADPHONES REQUIRED. · 60 minutes
Creates 2 Hz deep delta with lower carrier for sleep. STEREO HEADPHONES REQUIRED. · 45 min
Low alpha for immediate stress reduction. Calms the nervous system while maintaining alertness. Use during stressful moments. HEADPHONES REQUIRED. · 30 minutes
Frequently asked questions
What frequency was in the bedtime sound?
A 3 Hz beat, made with a 250 Hz tone in one ear and a 253 Hz tone in the other, played for 30 minutes over instrumental music. The library already stores that number on the row titled 3 Hz. For the sleep topic, also open 3 Hz Delta - Growth Hormone Release and Binaural Deep Sleep Induction (150/152 Hz), because those rows match sleep. For the stress score, open 8 Hz Low Alpha - Stress Reduction, because that row matches the stress outcome rather than the 3 Hz beat. This trial did not test those rows. A magnetic-field setting is not stored in the tone.
How did sleep scores compare with no sound?
By week 2 the 3 Hz group was 0.91 points lower than no sound. At week 4 the gap was 1.23 points, and two weeks after the listening stopped it was 1.32 points. Lower on this score means better sleep. Pink noise did not separate from no sound on that score until week 6.
How did the 3 Hz beat compare with pink noise?
On sleep scores, the two listening groups did not separate at any check. On the stress score, the 3 Hz group was 2.55 points lower than pink noise at week 4 and again at week 6. Pink noise did not separate from no sound on stress at any check.
What can I open before bed?
Open 3 Hz, because that row already stores the bedtime beat, and listen through headphones for about 30 minutes. Also open 3 Hz Delta - Growth Hormone Release and Binaural Deep Sleep Induction (150/152 Hz), because those rows match sleep, and 8 Hz Low Alpha - Stress Reduction, because that row matches the stress score. This trial did not test those other rows.