The semantics of dreams

Abstract

Dreams are universal yet deeply personal experiences. While memory and personal concerns influence dream content, the impact of other individual, generalizable traits remains poorly understood. To address this gap, we built a multimodal dataset including dream and wakefulness reports, alongside demographic, psychometric, cognitive, and sleep-related measures in a large adult cohort. Natural language processing characterized the semantic features that quantitatively distinguish dream from wakefulness reports, with this distinction significantly modulated by individual-specific factors. Longitudinal and cross-sample analyses further demonstrated that major external events, such as the COVID-19 pandemic, affect dream content, leaving lasting traces. Overall, the findings highlight a dynamic interplay between stable individual traits and external events in shaping dream experiences, offering novel insights into the cognitive and emotional architecture of dreaming.