Most people have had a stretch where their mind refused to go quiet.
For someone living with schizophrenia, that inner noise can take the shape of voices no one else can hear. And for many, no pill fully makes them stop.
A first-in-human clinical trial at the University of California, San Francisco is now testing whether a focused beam of sound can calm those voices. The approach is noninvasive. There is no surgery and no implant in the skull.
The results were published in the journal Brain Stimulation in 2026.
Also read: Taking vitamin C during pregnancy linked to calmer babies
About 70 percent of people with schizophrenia hear voices others cannot. Roughly 40 percent still hear them even on antipsychotic medication, according to the UCSF researchers behind the trial.
For those readers, options have narrowed to switching drugs, adding talk therapy or trying brain stimulation methods that only reach the outer layers of the brain.
What the UCSF trial is testing
The trial is led by Karuna Subramaniam, a professor of psychiatry at UCSF. It will enroll up to 12 adults with schizophrenia whose voices have not quieted on medication.
Each participant goes through five daily sessions of active ultrasound. Five sham sessions follow, with a washout break in between. The design uses each person as their own control, so the real pulses can be compared with the fake ones in the same brain.
Also read: Taking any of these common drugs? Mixing them with alcohol may be dangerous
The device is a wearable headset built by Attune Neurosciences, a San Francisco company. It aims low intensity ultrasound at the caudate nucleus, a deep structure that helps the brain sort real sounds from internally generated ones.
In schizophrenia, that filter appears to misfire, and internal speech can be experienced as an outside voice.
What a small pilot already showed
Before the current trial, Subramaniam's team ran a single-session pilot in people with schizophrenia who were still hearing voices on their medication.
After one session of focused ultrasound aimed at the caudate, participants reported significant, though temporary, reductions in their hallucinations compared with a sham session.
Also read: He drank 13 cups of water every day for a week: Here's what happened
The pilot also measured changes in how the striatum talked to the auditory cortex, the part of the brain that processes sound.
A single-session pilot is a signal, not a result. It was enough to earn safety clearance for the new, larger trial, which will test whether repeated sessions produce effects that stick.
Enrollment for the UCSF trial has begun. Anyone with schizophrenia who still hears voices on medication can ask their clinician whether research sites in their area are testing focused ultrasound or other brain stimulation approaches.
Also read: How much fiber should you eat per day? The target sorted by age and sex
Also read: Cervical cancer deaths could rise nearly 40 percent by 2050
This article is made and published by Ida-Marie Palm Varbæk, who may have used AI in the preparation.