Back

Self-other voice confusion in patients with auditory-verbal hallucinations and nonclinical hallucination proneness

Vukojevic, J.; Jelic, L.; McCormack, K.; Susac, J.; Muselimovic, I.; Bagaric, M.; Brecic, P.; Dellwo, V.; Cifrek, M.; Savic, A.; Orepic, P.

2025-10-07 psychiatry and clinical psychology
10.1101/2025.10.06.25337403 medRxiv
Show abstract

Background and HypothesisAuditory-verbal hallucinations (AVH), hearing voices without external speakers, are a core symptom of schizophrenia. A prominent account proposes that AVH reflect failures to recognize self-generated speech. Similar effects in hallucination-prone individuals suggest that these mechanisms span a continuum from subclinical to clinical manifestations. We used a self-other voice discrimination (SOVD) task, previously identified as a potential biomarker of self-disturbance, to assess self-recognition deficits in patients with (AVH+) and without (AVH-) a history of AVH, as well as in healthy individuals. Study Design41 schizophrenia patients (23 AVH+, 18 AVH-) and 40 healthy controls completed the SOVD task. In a follow-up experiment, 26 additional healthy participants performed a familiar-other voice discrimination task identical to SOVD, but without the self-voice. In patients, performance was tested in relation to symptom severity (PANSS), and in controls to hallucination proneness scores. Study ResultsSOVD performance was selectively impaired in AVH+ patients, while AVH- patients did not substantially differ from controls. Across groups, higher PANSS and hallucination proneness scores were specifically related to reduced self-voice recognition, with no impact on other-voice recognition. This effect did not extend to familiar-other voice discrimination. ConclusionsImpaired self-voice recognition is a promising selective marker of AVH occurrence in the acute phase of schizophrenia and extends to hallucination proneness in the general population. These findings support a dimensional view of hallucinations and point to deficits specific to self-voice processing rather than general voice perception. They also highlight SOVD as a promising cognitive biomarker of AVH, with direct implications for early identification and targeted interventions.

Matching journals

The top 2 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.