Ketamine produces different brain oscillations in distinct regions: gamma oscillations in prefrontal cortex and hippocampus, linked to antidepressant effects, and a 3 Hz oscillation in posteromedial cortex, linked to dissociative effects. By analyzing intracranial recordings from humans and comparing effects with propofol, the authors identified that these frequency-dependent patterns arise from distinct neural circuits, potentially guiding development of biomarkers and treatments for depression.
Learning, semantic processing, and online prediction persist in the human hippocampus during general anesthesia-induced loss of consciousness. Using high-density Neuropixels microelectrodes to record neural activity while playing tones to anesthetized patients, hippocampal neurons reliably detected oddball tones, and this effect grew over about ten minutes, consistent with learning. A recurrent neural network model showed that learning and oddball representation emerge from flexible tone discrimination. When language stimuli were played, single units and ensembles carried information about semantic and grammatical features of natural speech, even predicting semantic information about upcoming words. These results indicate that complex sensory processing occurs in the hippocampus even in the unconscious state.
A subanesthetic dose of ketamine increases gamma oscillations in the prefrontal cortex and hippocampus, brain areas linked to its rapid antidepressant effects, and produces a 3 Hz oscillation in the posteromedial cortex that may underlie its dissociative effects. By adding propofol, which blocks NMDA-mediated disinhibition and shares HCN1 inhibition with ketamine, the study distinguished brain dynamics caused by NMDA-mediated disinhibition from those caused by HCN1 inhibition. The results suggest ketamine engages distinct neural circuits in frequency-dependent patterns to produce antidepressant and dissociative effects, potentially guiding development of new depression therapies with fewer side effects.