Detection of acute 3,4-methylenedioxymethamphetamine (MDMA) effects across protocols using automated natural language...

Area of Science:

• Neuroscience
• Computational Linguistics
• Psychopharmacology

Background:

• Clinical assessments of drug-induced mental state changes are subjective.
• Objective biomarkers are needed to accurately characterize altered mental states.

Purpose of the Study:

• To investigate automated speech analysis for detecting objective markers of mental state changes induced by 3,4-methylenedioxymethamphetamine (MDMA) and oxytocin.
• To assess the utility of acoustic, semantic, and psycholinguistic speech features for classifying drug conditions.

Main Methods:

• Utilized computer-extracted speech features from 31 healthy adults across four drug/placebo sessions.
• Employed randomized, double-blind administration of MDMA (0.75, 1.5 mg/kg) and oxytocin (20 IU).
• Analyzed speech tasks during peak drug effects, including group and individual classification on independent datasets.

Main Results:

• Achieved high classification accuracies (up to 87% cross-validated, 92% on independent data) for detecting drug conditions.
• Oxytocin effects were primarily linked to acoustic features (emotion, prosody).
• MDMA effects manifested across multiple speech domains (acoustic, semantic, psycholinguistic); task design influenced speech responses.

Conclusions:

• Automated speech analysis shows potential as an objective tool for measuring drug-induced mental states.
• Speech patterns provide reliable markers for differentiating effects of MDMA and oxytocin.
• Further research can refine speech-based biomarkers for mental state assessment.