During the course of routine clinical study, it has become apparent that the all-or-none concept of state determination (wakefulness, nonrapid eye movement sleep, rapid eye movement sleep) does not always exist, and that ambiguous, multiple, or rapid oscillation of state-determining variables appear in a wide variety of experimental and clinical situations. Six cases of extreme state dissociation are presented, with a review of the human and animal clinical and experimental literature. This multiple component concept of state determination must be kept in mind when pharmacologic or lesion studies are employed to suppress one or another state. Such manipulation may suppress some of the commonly used markers for that state (i.e., polygraphic) without affecting other variables of that state. The existence of mixed states will be a challenge to the development of automated computerized polysomnogram scoring.