What groups were studied and how many participants were involved?

142 participants were divided into three groups: individuals with schizophrenia who had experienced auditory hallucinations within the past week; people with schizophrenia but no recent or historical hallucinations; and healthy control participants. Participants wore EEG caps and listened to simple syllables through headphones while being instructed to imagine saying the sounds, such as 'bah' or 'bih', with researchers measuring auditory cortex activity during the task.

How did the imagined-speech task work and what was measured?

Participants wore EEG caps and listened through headphones to simple syllables while being instructed to imagine saying the sounds, such as 'bah' or 'bih'. The researchers then measured cortical responses in the auditory processing regions during these imagined-speech trials to assess how internal speech is represented in the brain when not spoken aloud.

What was the main finding about group differences in imagined-speech responses?

The study found that people with schizophrenia but no recent hallucinations produced intermediate auditory-cortex responses between the other two groups and that those with recent hallucinations showed stronger responses to imagined speech. This pattern supports a disruption in the brain's corollary discharge mechanism, making internally generated speech more likely to be misattributed as coming from an external source.

What are the implications and potential applications of this measure?

Investigators say the results provide the most direct experimental evidence to date that misattributed inner speech may contribute to auditory verbal hallucinations. The imagined-speech measure could potentially serve as a biomarker for psychosis risk if longitudinal studies show it predicts transition into psychosis, enabling earlier identification or monitoring of individuals at elevated risk.

EEG study links misattributed inner speech to auditory hallucinations in schizophrenia

Science By James Lawson Nov 15, 2025 12:13

A UNSW Sydney study using EEG on 142 participants found that people experiencing recent auditory hallucinations show increased auditory-cortex responses to imagined speech, supporting the hypothesis that some hallucinated voices are misattributed internal speech.

Study overview

Participants and methods

The study enrolled 142 people divided into three groups: individuals with schizophrenia who had experienced auditory hallucinations within the past week; people with schizophrenia but no recent or historical hallucinations; and healthy control participants without schizophrenia.

Participants wore electroencephalography (EEG) caps and listened through headphones to simple syllables while being instructed to imagine saying the sounds (reported examples include "bah" and "bih"). Researchers measured cortical responses in the auditory processing regions during these imagined-speech tasks.

Findings

People with schizophrenia but no recent hallucinations produced intermediate responses between the other two groups. The pattern supports a disruption in the brain’s prediction mechanism—often described as corollary discharge—so that internally generated speech can be interpreted as coming from an external source.

Implications

Investigators say the results offer the most direct experimental evidence to date that misattributed inner speech may contribute to auditory verbal hallucinations. The measure could potentially serve as a biomarker for psychosis risk if it predicts transition into psychosis in longitudinal studies.

Publication

The full study appears as "Corollary Discharge Dysfunction to Inner Speech and Its Relationship to Auditory Verbal Hallucinations in Patients with Schizophrenia Spectrum Disorders" in Schizophrenia Bulletin (October 21, 2025).

James Lawson

Investigative science and tech reporter focusing on AI, space industry and quantum breakthroughs

University College London (UCL) • United Kingdom