Date: 10 November 2026 (Tuesday)
Time: 4:00 pm – 5:00 pm
Venue: Seminar Room 304, 3/F, William M.W. Mong Block, 21 Sassoon Road 
Host: Professor Cora Lai

Jaekyung Kim is an Assistant Professor in the Department of Biological Sciences at KAIST, where he leads the Neural Processing Laboratory. His research explores how sleep shapes memory, neural dynamics, and higher cognitive functions, with a particular focus on uncovering how sleep and dreaming support creativity. His work integrates systems and computational neuroscience with large-scale neural recordings, neuromodulator measurements, and closed-loop circuit manipulation. During his postdoctoral training at the University of California, San Francisco, his work established causal roles for sleep-dependent neural activity in memory consolidation, motor learning, and recovery after brain injury. Building on this foundation, his laboratory at KAIST is developing cross-species approaches spanning rodents, non-human primates, and humans to investigate the biological functions of dreaming and the neural mechanisms underlying creative cognition. His long-term goal is to establish an integrated gene–brain–behavior framework for understanding how sleep and dreams support adaptive behavior, creativity, and lifelong cognitive health.

Sleep supports skill consolidation, but how it differentially regulates accuracy and kinematic variability remains unclear. Using low-dimensional manifold analysis and closed-loop optogenetic disruption of cortico-cortical slow oscillations during NREM sleep in rats, we show that disrupting mPFC–M1 communication selectively impairs motor cortex manifold stabilization and increases kinematic variability, while sparing accuracy gains. Restoring normal sleep after cessation of the intervention rescues cross-area coordination and variability regulation, highlighting a selective role for NREM cortico-cortical communication in refining motor execution. These findings provide causal evidence that mPFC–M1 slow oscillation coupling supports sleep-dependent refinement of motor-related neural manifolds and fine kinematic control.

In parallel, we are developing a cross-species framework to identify conserved dream-related sleep biomarkers. We focus on PGO waves in primates and their rodent homologues, P-waves, which are phasic sleep events associated with REM sleep and dreaming. Across rats, macaque monkeys, and humans, PGO/P-wave–related events showed consistent coupling with cortical theta oscillations, including robust phase locking with frontal theta activity. High-density macaque ECoG further revealed spatially isolated, region-specific cortical profiles, while human scalp EEG analyses showed that PGO-wave–related signatures can be detected non-invasively. Building on these conserved biomarkers, we will ask whether dream-related sleep dynamics mark the moments when latent task representations are reorganized into abrupt behavioral insight. Together, this work links conserved dream-related neural activity with a broader framework for understanding sleep-dependent insight emergence.

All are welcome.