Researchers at MIT and Boston University have demonstrated that precisely timed bursts of pink noise during sleep can increase slow electrical waves in the brain, strengthening cerebrospinal fluid waves that clear away metabolic waste, according to a study published in Science Translational Medicine.
While millions of people rely on ambient sound machines and mobile apps to block out nighttime clamor, a team of neuroscientists has discovered that sound can do far more than just soothe. By targeting the brain’s natural rhythm during deep sleep, researchers found a way to amplify the body’s internal plumbing system and flush out accumulated toxins. The discovery bridges electrical engineering and neuroscience, offering a fresh look at how the brain cleans itself.
How Pink Noise Amplifies Brain Waves and CSF Flow
Throughout waking hours, the brain accumulates metabolic waste products such as lactic acid and worn-out proteins. During deep, non-REM sleep, the glymphatic system deploys rhythmic waves of cerebrospinal fluid, or CSF, to wash through the tissue and clear away that debris. Without this nightly cleansing, toxins can trigger inflammation and contribute to long-term neurological damage.
In a proof-of-concept test involving 14 healthy volunteers who fell asleep inside functional magnetic resonance imaging scanners, researchers set out to see if they could manipulate this clearing mechanism. They used 50-millisecond bursts of pink noise—a static-like sound featuring balanced frequencies where lower pitches are louder and higher pitches are softer, resembling steady rain or a distant waterfall.
“Similar to a child on a swing, if you push them when they’re at the right moment in their movement, you can make that swing go farther. The challenge is: How do you find just the right time?”
Laura Lewis, senior study author, MIT
Because the magnetic fields of an MRI machine interfere with standard monitoring equipment, the team engineered a specialized algorithm capable of processing electroencephalography signals in less than 100 milliseconds. This allowed the system to predict slow-wave peaks and deliver the sound stimulus at the exact moment required to boost electrical amplitude. When timed correctly, the auditory stimulus increased slow-wave amplitudes, which in turn caused blood vessels to constrict and dilate like a pump, driving larger waves of CSF out of the tissue.
Translating Lab Technology Into At-Home Sleep Solutions
The findings mark a significant leap from observational research to active intervention. Back in 2019, the team first measured CSF waves moving through the sleeping brain using fMRI. By proving that those fluid movements could be intentionally strengthened, researchers opened the door to targeted therapies for sleep disturbances and cognitive decline.
However, the researchers caution that casual listeners cannot replicate the effect at home by pulling up a playlist on a smartphone while scrolling in bed. The precise timing required makes off-the-shelf sound machines ineffective for this specific physiological response.
Commercialization efforts are already underway. Study coauthors Laura Lewis and Joshua Levitt have a patent pending on a CSF-flow neurofeedback system and have cofounded Cerebloom Inc. Levitt, who recently earned his PhD from Boston University and was a visiting graduate student in Lewis’ lab, launched the company to develop a headband-like device designed to deliver properly timed auditory stimuli in domestic settings.
Next Steps for Clinical Populations and Brain Health
The accumulation of harmful proteins such as amyloid beta and tau is a hallmark of neurodegenerative conditions like Alzheimer’s and Parkinson’s diseases. By finding a way to enhance waste clearance in healthy adults, the research team aims to test whether the technology can alter disease progression in older adults or clinical populations.
Harvard Medical School assistant professor of sleep medicine Rebecca Robbins notes that the findings do not alter foundational sleep hygiene advice. Whether future wearable devices can successfully translate laboratory-grade neurofeedback into accessible clinical treatments remains the central question for researchers as they plan their next trials.
