It is often claimed that the theory of emergence, should it ever exist, would be a "theory of everything" in the truest sense of the word, not just a "theory of what everything is made of". A general description of the circumstances under which any assemblage of interacting agents would reduce its within system entropy stands as the most obvious candidate for a "missing law of nature". If you buy those assertions, then investigating emergence and the statistical mechanics involved should seem like an enticing way to make a mark on the world. The brain would be the most sensible thing to study as it stands as the quintessential example of a system with many kinds of emergent properties which develop on experimentally accessible timescales and for which a wide diversity of computational models exist. Two mature concepts have already improved our understanding of emergence: evolution and RG. We'll begin with an introduction to a study aiming to learn what information about the network at large is accessible to a neuron. Then we'll see how surprising results lead to parallels with RG both of neural systems and by neural systems. We'll see how evolution is being applied to brain function and how it might lead to a simplified framework for understanding how thoughts are represented and changed, as well as how the brain develops during childhood. Finally we'll end with future steps to enrich the investigation.
Graduate Student Seminars
Emergence And The Brain: Bringing Evolution, Renormalization Group, And Phase Transitions To Bear On The Problem
James Johnson, Department of Physics, Washington University
January 20, 2017 at 4:00 pm