Computational Neuroscience InitiativeBarchi Library, 140 John Morgan Building
Leenoy Meshulam
Computational Neuroscience Center
University of Washington
Emerging simplicity in the nervous systems of mouse and octopus
For an animal to perform any function, millions of neurons in its nervous system furiously interact with each other. Be it a simple behavior or a highly complex computation, all functions involve the concerted activity of many individual units. We seek theoretical approaches that can simplify the rich dynamics of the coordinated activity of thousands of individual cells. In this talk, we shall focus on two seemingly very different nervous systems: mouse brain neural activity patterns, and octopus skin cells activity patterns. We draw on concepts from statistical physics such as the renormalization group, to capture the collective nature of activity. Our approach uncovers hallmarks of striking simplicity despite the complexity of both systems. In the mouse, we uncover scaling behavior and hallmarks of an RG fixed point. In the octopus, camouflage skin pattern activity is reliably confined to a (quasi-) defined dynamical space. Finally, we shall touch upon the benefits of comparing across animals to achieve a multi-scale understanding of the nervous system, i.e., how macroscale properties, such as memory or camouflage, emerge from microscale level activity of individual cells.