Study of Speech Patterns to Predict Psychosis Transition in People Aged 15 to 30

What this Trial Is About

Researchers are investigating whether subtle speech disorganization can predict the transition to schizophrenia in patients at ultra-high risk. This longitudinal multicenter cohort study aims to identify specific language markers related to psychotic transition and validate a French predictive model using computerized speech analysis techniques. The focus is on analyzing speech patterns to better understand and predict the development of schizophrenia. Participants undergo an initial assessment including the CAARMS scale and an audio recording of their medical interview. The recorded speech is transcribed and analyzed using computerized methods such as Latent Semantic Analysis LSA, lemmatization, vectorial analysis, and grammatical function evaluation. These analyses seek to identify linguistic markers linked to the transition to psychosis, which will be used to build a predictive model for schizophrenia. Throughout the study, researchers monitor participants for up to two years to observe any transition to schizophrenia. Data from speech analysis and clinical assessments are combined to measure outcomes. The study involves regular evaluations of participants language and psychiatric status, aiming to improve early detection and understanding of psychosis development in at-risk individuals.

Semantic and Syntactic Computerized Analysis of Free Speech