387. Task-Specific Sensitivity of Digital Speech Assessment to Symptom Severity in First-Episode Psychosis

Alterations in speech and language are evident in psychotic disorders, reflecting core symptoms of formal thought disorder, negative symptoms, and cognitive deficits. Digital phenotyping approaches, including automated speech analysis, have the potential to enhance clinical care by providing objective and easy-to-deploy symptom monitoring. Digital speech assessment may be particularly useful in first-episode populations, where relapse prediction is crucial for timely intervention. In the current study, we compared the sensitivity of different speech tasks (open-ended vs. structured) to symptom severity in first-episode psychosis.