Summary: In a creative shift for psychiatric research, a study suggests that songwriting and group music-making can help people with psychosis re-connect with reality. The research explores the concept of predictive coding, the brain’s ability to anticipate what happens next.
By engaging in the rhythmic and melodic expectations of music, participants with schizophrenia and auditory hallucinations reported reduced paranoia and a significant shift away from social isolation.
Key Findings:
- Reduced Paranoia: While hallucinations didn’t disappear for everyone, those with less severe symptoms reported a measurable drop in paranoia after the weekly two-hour group songwriting sessions.
- The “We” Shift: Researchers tracked a significant change in language. People with psychosis often use first-person pronouns (I, me, mine) due to distress and isolation. Following the sessions, there was a marked increase in plural pronouns (we, us, ours), signaling a return to social connectivity.
- Beyond Medication: Unlike antipsychotic drugs, which can cause lethargy and “brain fog,” the music therapy had no negative side effects. Participants “came alive,” expressing emotions and creativity that traditional medicine often struggles to reach.
- Permanent Rewiring: Dr. Corlett suspects that consistent music-making may permanently alter brain circuitry, essentially “retraining” the dysfunctional prediction systems of the brain.
Source: Yale
Our brains anticipate sensory signals, such as sight, sound, smell, or touch, by relying on past experiences. When we bite into an apple, for example, we expect a sweet crunch because of all the other times we have eaten one.
Some neuroscientists believe that this neural processing, known as predictive coding, helps ease the brain’s cognitive load and facilitate faster learning. But at times, these expectations or predictions can go wrong, resulting in the hallucinations and delusions that can come with psychosis, a mental state where the mind loses touch with reality.
In a new study published April 9 in the journal Psychosis, Yale scientists demonstrated a way to help those with psychosis re-engage with their surroundings through making music.
“Music is a Golden Road for making predictions,” says Philip Corlett, PhD, an associate professor of psychiatry at Yale School of Medicine and the senior author of the study. When one sings the lyric, “Sweet Caroline …,” for example, the mind conjures the ensuing melody: “Bab, bab, baa …”
Because of this strong link between music and prediction, Corlett’s research group at the Belief, Learning, and Memory Lab set out to assess the impact of song-making on psychotic illnesses, particularly hallucinations.
“People have hallucinations because their predictions are too strong, and that makes them see and hear things that other people don’t see or hear,” Corlett says. Making music might be an avenue to help dysfunctional brains regain their ability to make good predictions, he says.
“Like a roller coaster, music is a safe way of having our expectations violated whilst not having to experience any kind of dangerous and unsafe things.”
For the project, Corlett partnered with Adam Christoferson, a music facilitator, a member of the Citizens Community Collaborative at the Yale Program for Recovery and Community Health, and founder of Musical Intervention, an organization that uses music-making for rehabilitation and community building in New Haven.
Corlett became interested in music therapy for psychosis after observing its effect on patients he met at the Connecticut Mental Health Center.
“I saw some people who I’d seen for the past 10 years or so, coming in and out of the center, not really improving very much,” Corlett says. “And then they sat with Adam for an hour, and they seemed to come alive; they weren’t as negative or finding it hard to express their emotions or connect with others. And I just thought I wanted to be part of this by exploring the science behind it.”
Group music eases paranoia
The researchers recruited 20 people between the ages of 18 and 65 in Connecticut to participate in a six-week longitudinal study. The individuals, who were either referred by their clinicians or self-registered, had schizophrenia or experienced distressing auditory hallucinations at least once per week.
During their initial visit, the participants completed a set of psychometric questionnaires to assess their tendencies toward hallucinations and paranoia. They also participated in an interview with the study facilitators.
For the next four visits, the participants formed groups of five and met weekly for two hours to write songs with guidance from a professional musician. They received recording equipment, including a microphone, guitar, keyboard, and drums, and were encouraged to write their own lyrics.
At the final visit, the participants filled out the same set of questionnaires and sat for a post-session interview.
“We wanted to longitudinally assess people’s changes objectively,” says Deanna Greco, a PhD student in the Corlett lab and the first author of the study.
The researchers did not find a decrease in hallucinations for all participants, but those with less severe hallucinations reported experiencing less paranoia after the sessions.
The researchers also noted a change in the participants’ language. Previous research has shown that people who experience severe psychosis use first-person pronouns (I, me, mine) more often than plural pronouns (we, us, ours), which can indicate social isolation and distress.
After the participants’ final interview, “we were seeing a decrease in their first-person pronoun usage and an increase in plural pronoun usage,” Greco says.
Individuals with psychosis often experience social isolation, paranoid thoughts, and stigma. While the participants came with varying levels of psychosis, they all benefited from the group music activities by gathering with the community and rekindling their creativity, Greco says.
For Christoferson, the result was a validation of what he had seen in the field for the last 25 years of running similar groups.
“The Yale study shows the validity of the Musical Intervention approach,” he says. To him, the song-making activities offer the participants a sense of identity and a way to express their emotions and creativity, which in turn affects their livelihood.
Treating psychosis with music therapy “is a really exciting area of research,” Corlett says.
Individuals with psychosis are typically prescribed antipsychotics to reduce symptoms such as hallucinations and delusions. These medications can cause unpleasant side effects, including difficulty concentrating, lack of motivation, and lethargy.
“Our approach may be outside of the purview of clinical medicine,” Corlett says.
“But the study showed that we can do proper clinical scientific research on music therapy, and that it can do just as well as the more standard and traditional treatments, and perhaps better in some cases, because people come back, they want more, and they don’t experience negative side effects.”
As a follow-up to this study, Corlett and his team are investigating how music as an intervention changes brain circuitry. “I suspect that it does something permanently in the brain,” he says. “And we want to find out what that is.”
Key Questions Answered:
A: It’s about the “Ba, Ba, Baa!” Music forces the brain to make a prediction and then immediately confirms it. For a brain that is struggling to tell the difference between its own internal “noise” and external reality, this rhythmic “call and response” acts like a physical therapy for the mind’s logic center.
A: Not yet, but it’s a powerful “non-clinical” tool. The researchers found that music therapy does what meds can’t: it reduces social isolation and sparks creativity without making the patient feel lethargic or unmotivated. It’s a way to treat the person, not just the symptom.
A: Language is a window into the soul’s isolation. Psychosis often traps people in a terrifyingly solitary world. Moving from “I am hearing voices” to “We are writing a song” is a massive neurological leap from isolation back into the community.
Editorial Notes:
- This article was edited by a Neuroscience News editor.
- Journal paper reviewed in full.
- Additional context added by our staff.
About this music and mental health research news
Author: Colleen Moriarty
Source: Yale
Contact: Colleen Moriarty – Yale
Image: The image is credited to Neuroscience News
Original Research: Closed access.
“Song-making in a group (SING): a longitudinal study for people experiencing psychosis” by Deanna L. Greco, Santiago Castiello de Obeso, Sandy Camilo, Charlotte Freeland, Anthony Pavlo, Claire Bien, Julia Nachemson, Constance Lubinski, Adam Christoferson, Joshua Kenney, and Philip R. Corlett. Psychosis
DOI:10.1080/17522439.2026.2634654
Abstract
Song-making in a group (SING): a longitudinal study for people experiencing psychosis
Background
Creative expression transforms imaginative landscapes into tangible realities. Music-making in particular sets perceptions into motion that both creator and appreciator can experience. This innovative act establishes one’s identity within the community, and engaging in the creative process with others further emphasizes one’s voice within the collective. Group-based music-making’s ability to foster agency and belonging might be especially advantageous for those experiencing isolation and alienation. This investigation examines song-making in a group’s impact on people experiencing psychosis.
Methods
Twenty participants with psychosis engaged in four sessions during which they wrote and recorded songs with four other participants and a music facilitator. This study builds on music therapy research in psychosis by incorporating symptom specific measures and adding linguistic analyses as objective measures of mental states. Symptom changes were assessed by administering paranoia and hallucination questionnaires before and after the music intervention. Fluctuations in the language participants use to describe experiences were quantitatively captured using Linguistic Inquiry and Word Count 2022.
Results
Although a decrease in hallucinations was not observed, paranoia decreased in participants who experience less severe hallucinations as compared to those who experienced them more frequently. Linguistic findings revealed a significant reduction in the usage of the first-person pronoun “I” and significant increases in the plural pronoun “we,” accomplishment, agentic, cognitive processing, and positive emotion language.
Discussion
Suggestions regarding how these language shifts reflect the song-making activity and translate into participants’ lives outside of the song-writing workshop are discussed. Overall, this study highlights the potential of group-based song-making in promoting recovery from psychosis.
