For nearly a century, sleep in the brain has been defined by electrical waves that travel slowly across multiple centimeters of the isocortex. In stark contrast, the fast neuronal spike-patterns known to encode sensory information exist in milliseconds of microcircuit activity. My recent pre-print demonstrates that state, too, is reliably encoded in fast, non-oscillatory spike-patterning within diverse individual microcircuits throughout the brain. However, despite revealing a novel, fast and local foundation of sleep and wake, my initial study may underestimate the true minimal scale of the encoding. I hypothesize that the minimal encoding of state in neural activity is spike patterns, not at the level of the microcircuit, but the individual neuron. First, I will verify whether individual neurons encode sleep and wake states in their spike patterns, using previously conducted microelectrode recordings. Statistical comparisons of spike patterns betweens states will lay a reproducible foundation in connection with prior literature. My preliminary results reveal that, while individual neurons encode state in their activity, the general principles require further elucidation. Generally, many encodings are structured by the hierarchical organization of the brain, and this may also be true for sleep/wake encodings. I propose that machine learning can disentangle the general influences of the animal and region from encodings of state that are unique to each individual neuron. Unique neuron- level encodings in spike patterning represent a new minimal scale of state, embedding this information on the scale of neuronal computation. Neuron-level states could provide new insight into brain function and, specifically, animal behavior. My recent pre-print revealed that microcircuit spiking patterns “flicker” between sleep and wake independently of the animal’s overall state. Further, flickers emerge as a function of sleep pressure and drive behavioral discontinuities (twitching, pausing). A new, neuron-level state encoding can reveal whether individual neurons exhibit similar state discontinuities, and whether this explains complex interactions between state and behavior. This work has great potential to disentangle brain states on the scale of brain computation (neurons). Such fundamental knowledge about the basis of sleep and its connection to behavior is highly important to public health. Biomedically, sleep dysfunction is widely implicated in nearly all neurological disorders- from epilepsy to autism to Alzheimer’s. Further, localized forms of sleep amidst wake are increasingly observed in conjunction with attentional disorders and stroke victims. These aims facilitate excellent training, particularly in developing novel machine-learning architectures and performing sophisticated electrophysiological recordings. My highly-supportive co-sponsors are excellent investigators in computational neuroscience and sleep from renowned institutions. Completion ...