Voice patterns in schizophrenia: A systematic review and Bayesian meta-analysis.

Area of Science:

• Psychiatry and Behavioral Sciences
• Computational Linguistics
• Acoustic Phonetics

Background:

• Voice atypicalities are historically recognized features of schizophrenia.
• These vocal changes correlate with negative symptoms (flat affect, alogia) and social impairments.
• Acoustic patterns may serve as potential biomarkers for clinical status and social functioning in schizophrenia.

Purpose of the Study:

• To systematically review and meta-analyze acoustic patterns in schizophrenia.
• To investigate the relationship between voice atypicalities and clinical features, particularly negative symptoms.
• To identify potential acoustic markers for schizophrenia.

Main Methods:

• Systematic literature review and meta-analysis of 46 articles (55 studies).
• Inclusion of 1254 patients with schizophrenia and 699 healthy controls.
• Multilevel Bayesian modeling to estimate effect sizes (Hedges'g, Pearson's r) for acoustic variables.

Main Results:

• Weak atypicalities in pitch variability (g = -0.55) linked to flat affect.
• Stronger atypicalities in speech rate, proportion of spoken time, and pauses (g's -0.75 to -1.89) associated with alogia and flat affect.
• Modest overall effects with significant heterogeneity; pause duration showed a notable effect. Cognitive/social task demands amplified effects.

Conclusions:

• Acoustic pattern analysis is a promising research avenue for schizophrenia markers.
• Current findings indicate modest but significant voice atypicalities related to core symptoms.
• Recommendations for more cumulative, open, and theory-driven research are proposed to advance the field.