An interview with psychology researcher Neguine Rezaii about her work using machine learning to predict conversion in teenagers from prodromal symptoms to psychotic episodes. The two language patterns found in the subjects' speech were 1) a low semantic density (i.e., little meaning), and 2) speech related to sound or voices. Topics discussed include: how exactly they determined “low meaning"; how the algorithm found, on its own, indicators related to sound-related speech content; the future of using machine learning and automatic diagnosis tools in psychology and therapy; theories that might help explain these findings.
Learn more about the show and get transcripts at behavior-podcast.com.
Learn more about your ad choices. Visit megaphone.fm/adchoices
People Who Read People: A Behavior and Psychology Podcast