medRxivJClub
SPEECH FEATURES IN A DEVELOPMENTAL PERSPECTIVE: A SCREENING TOOL FOR PSYCHOSIS, DEPRESSION AND ANXIETY
medRxiv · · Preprint · Open access
Ribeiro, Amorim + 6 more
Abstract ↗AI summary
The abstract is read at the publisher; the summary is JClub's.
Emotional Narrative Complexity (ENC) and Narrative Complexity (NC) derived from brief picture-elicited narratives increase with schooling, saturating around middle school, with R2=0.31 (ENC) and R2=0.29 (NC).
- Why it matters: Clinicians need scalable, developmentally calibrated speech markers to monitor cognitive and socioemotional growth and to support screening for mental wellness and psychopathology.
- What they did: The platform used word graphs from positively valenced picture narratives in 2149 participants (0–23 years of formal education) to model NC and ENC with asymptotic schooling functions and percentile normative curves.
- The result: In a transdiagnostic sample of 120 participants, ENC alone did not separate groups, but classifiers using ENC components plus education and sex reached AUC=0.726 for symptom risk and AUC=0.800 for diagnosis, improving with additional linguistic features.
The findingWhy it mattersWhat they didThe result
AI summary of the abstract · may contain errors · not medical advice · read the paper.
- Open access