Sleep Neuroimaging and Models of Consciousness

Summary

Deep sleep, marked by reduced sensory activity and loss of conscious awareness, offers a natural setting for testing theories of consciousness. This review examines recent fMRI studies of spontaneous brain activity during sleep, linking findings to the global workspace theory, information integration theory, and the dynamical core hypothesis. The authors highlight a shift from studying evoked responses to resting-state activity and emphasize the need for dynamic analysis of functional interactions over time. They also stress the importance of experimentally verifying reduced or absent conscious content during the deepest sleep stages.

Study at a glance

Abstract

Human deep sleep is characterized by reduced sensory activity, responsiveness to stimuli, and conscious awareness. Given its ubiquity and reversible nature, it represents an attractive paradigm to study the neural changes which accompany the loss of consciousness in humans. In particular, the deepest stages of sleep can serve as an empirical test for the predictions of theoretical models relating the phenomenology of consciousness with underlying neural activity. A relatively recent shift of attention from the analysis of evoked responses toward spontaneous (or "resting state") activity has taken place in the neuroimaging community, together with the development of tools suitable to study distributed functional interactions. In this review we focus on recent functional Magnetic Resonance Imaging (fMRI) studies of spontaneous activity during sleep and their relationship with theoretical models for human consciousness generation, considering the global workspace theory, the information integration theory, and the dynamical core hypothesis. We discuss the venues of research opened by these results, emphasizing the need to extend the analytic methodology in order to obtain a dynamical picture of how functional interactions change over time and how their evolution is modulated during different conscious states. Finally, we discuss the need to experimentally establish absent or reduced conscious content, even when studying the deepest sleep stages.