Joanna Kuc is a PhD candidate in Experimental Psychology at UCL, specialising in the intersection of natural language processing (NLP), psychedelics and mental health. Committed to fair and explainable AI, she develops innovative tools like AI-driven journaling chatbots to capture real-time data. Joanna has co-authored several peer-reviewed publications and presented at international conferences, including ASSC and APA, where she received the APA Division 21’s Robert W. Proctor Honorable Mention for her work on AI journaling.

Language as a Predictive Marker of 5-MeO-DMT Effects

This session explores how language and voice reflect psychological transformations associated with 5-MeO-DMT. Findings from a study using smartphone-based voice note diaries reveal shifts in vocabulary from external focus to introspection after the experience. Acoustic analysis demonstrates how vocal characteristics evolve across the phases of the experience, from preparation to integration, reflecting dynamic changes in arousal and emotional states. Additionally, we explore how large language models capture emotional cues in text, enhancing insights into psychological preparedness and predicting psychedelic experience trajectories. Overall, we introduce a novel framework that positions language as a predictive biomarker in psychedelic research.