Resting State Brain Activity Implications For Systems Neuroscience PDF Download

In the study of systems where basic laws have eluded us, as is largely the case in neuroscience, the simplest approach to progress might be to ask: what are the biggest, most noticeable things the system does when left alone? Without any perturbations or fine dissections, can regularities be found in the basic operations of the system as a whole? In the case of the brain, it turns out that there is an amazing amount of activity even in the absence of explicit environmental inputs or outputs. We call this spontaneous, or resting state, brain activity. Prior work has shown that spontaneous brain activity is dominated by very low frequencies: the biggest changes in brain activity happen relatively slowly, over 10's-100's of seconds. Moreover, this very slow activity of the brain is quite metabolically expensive. The brain accounts for 2% of body mass in an adult, but requires 20% of basal metabolic expenditure. Remarkably, the energy required to sustain brain function is nearly constant whether one is engaged in a demanding mental task or simply out to lunch. Furthermore, work over the past three decades has established that the spontaneous activities of the brain are not random, but instead organized into specific patterns, most often characterized by correlations within large brain systems. Yet, how do these correlations arise, and does spontaneous activity support slow signaling within and between neural systems? In this thesis, we approach these questions by providing a comprehensive analysis of the temporal structure of very low frequency spontaneous activity. Specifically, we focus on the direction of travel in low frequency activity, measured using resting state fMRI in humans, but also using electrophysiological techniques in humans and mice, and optical calcium imaging in mice. Our temporal analyses reveal heretofore unknown regularities in the way slow signals move through the brain. We further find that very low frequency activity behaves differently than faster frequencies, that it travels through distinct layers of the cortex, and that its travel patterns give rise to correlations within networks. We also demonstrate that the travel patterns of very low frequency activity are highly dependent on the state of the brain, especially the difference between wake and sleep states. Taken together, the findings in this thesis offer a glimpse into the principles that govern brain activity.