The human brain is incredibly complex, raising the question: can we even develop models with substantial predictive power? In this talk, we will see how tools from statistical physics can achieve this goal. As a main result, we show that cellular brain anatomy satisfies universal scaling laws, across various organisms, establishing the notion of "structural brain criticality". We also demonstrate that brain networks share simple organizing principles across the studied organisms, yielding predictive power in terms of additional biological observables. As in the brain the "hardware is the software", our results are expected to have broad implications on understanding brain function and dynamics, as well as on designing the next generation of artificial neural networks.
Istvan Kovacs, Assistant Professor, Northwestern University