The Dopaminergic Voice: Cross-Modal Benchmarking of Acoustic Prosody and Ventral Striatal BOLD for Anhedonia Classification
Summary
No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.
Summarise with VIRAT Free to try with your Science Catalogs account.Abstract
This pre-registration documents a cross-modal benchmarking study comparing two independent classification pipelines for detecting anhedonia: (Stream A) acoustic prosody features from DAIC-WOZ clinical interviews using Whisper large-v3 and OpenSMILE eGeMAPSv02, and (Stream B) ventral striatal BOLD activation from task-based fMRI (ds000030, UCLA CNP). We evaluate whether a 5-minute voice recording can classify anhedonia with AUC-ROC comparable to fMRI-derived neural markers.
Fields of Study & Subjects
Emotion and Mood Recognition
Treatment of Major Depression
Anxiety
Depression
Psychometrics
Treatment
Cognitive Processes
Publication details
- Publication date
- 1 January 2026
- Language
- EN
- Document type
- Preprint