Understanding how consciousness emerges from neural activity remains a central challenge in neuroscience. A major obstacle is the lack of a principled framework to measure, predict, and modulate consciousness. Inspired by statistical physics—where macroscopic variables emerge from microscopic dynamics—we investigate whether consciousness can be understood as an emergent property of large-scale brain networks operating near criticality.

In this talk, I present experimental and computational studies of brain state transitions during anesthesia, a controlled model for the loss and recovery of consciousness. Using human and animal EEG data, we estimate the deviation of brain dynamics from criticality and the proximity to a first-order dynamical regime. These measures allow us to predict the trajectory of recovery from anesthesia-induced unconsciousness and suggest potential strategies to modulate abnormal recovery, such as prolonged emergence and coma.

Finally, I will discuss how this criticality-based framework may extend beyond anesthesia to other complex systems, including chronic pain and artificial intelligence, suggesting
that critical dynamics may provide a universal principle underlying consciousness-like behaviors in complex systems.

Faculty Affiliate: Center for Consciousness Science; Center for the Study of Complex Systems; Michigan Psychedelic Center; Neuroscience Graduate Program, University of Michigan.