The dynamics of arousal: unifying fluctuations across brain, body, and behavior
Recent studies in awake, behaving animals demonstrate the widespread presence of arousal-related fluctuations across diverse neural, behavioral, and physiological measurements. These observations have led us to view such fluctuations as manifestations of a unified, intrinsic dynamical process that regulates organism-wide physiology. We have recently hypothesized that the spatial structure of brain-wide activity represents one such regulated variable – thus amounting to a generative mechanism underlying the widely studied phenomenon of “resting-state functional connectivity” (Raut et al., 2021 Sci Adv). In this talk, I will elaborate upon this framework by introducing a data-driven approach that enables the reconstruction of observables from a low-rank representation of their shared dynamics. I will demonstrate the success of this procedure across diverse neural and behavioral measurements recorded from awake mice, thus parsimoniously linking high-dimensional observations to a latent dynamical process evolving on a universal, low-dimensional manifold. Taken together, the proposed framework and findings elucidate an arousal-related dynamical mechanism that is both richer and substantially further-reaching than presently recognized. This carries implications for the analysis and interpretation of brain dynamics observed across modalities, tasks, and species.