Theoretical and Experimental Studies of Neuronal Network Dynamics: Relating Topology to Function.

Abstract

Within the brain, the interplay between connectivity patterns of neurons and their spatiotemporal dynamics is believed to be intricately linked to the bases of cognition, learning, and memory. In order to understand these processes, which are widely believed to be due to large-scale dynamical interactions, I investigate neuronal systems at the network level through computational simulation and reduced experimental preparations in conjunction with network analysis techniques. In a network of temporally evolving elements, it is possible to define a functional connectivity dependent on the spatiotemporal patterning of activity as well as an underlying anatomical connectivity. How functional integration or segregation of neuronal units arises from underlying anatomical structure could prove key to understanding the neural correlates of cognition and is dependent on various modulatory factors which define different brain states and functional modes. In addition, these dynamics are able to affect anatomical structure through plasticity and learning, completing a feedback loop of information processing and interaction with external environments.

In the first part of my work, I model the rapid formation of novel associative memories in the hippocampus and consolidation to long term storage sites in the neocortex. I examine how global modulation of network excitability can give rise to functional structure reflecting underlying heterogeneous connectivity associated with stored memory. This mechanism, coupled with a neocortical inhibitory feedback and two different timescales of plasticity, can mediate information transfer and memory consolidation. These dynamics are matched with experimental data observed during behavioral learning.

I pair these theoretical studies with an experimental investigation of a reduced hippocampal culture preparation. Within these cultures, cells are able to grow processes and connect together to form networks which can be easily visualized and recorded. I relate the anatomical neuronal network structure of these cells as well as the modulatory effects of a confluent glial network to changes in spiking activity, and find that, as the cultures mature and develop extended processes, bursting dynamics grow more coherent and global. Different glial network conditions modulate functional groupings, with more extensive glial morphology associated with global neuronal signaling and higher synchronization in firing dynamics.