Samora Okujeni: Modularity establishes mesoscale-criticality in neuronal networks
| When |
Nov 16, 2021
from 05:15 PM to 05:45 PM |
|---|---|
| Where | Zoom Lecture. Meeting ID and password will be sent with the invitation. You can also ask Fiona Siegfried for the access data. |
| Contact Name | Fiona Siegfried |
| Contact Phone | 0761 203 9549 |
| Add event to calendar |
 vCal iCal
|
Abstract
Combinatorial richness in the recruitment of neurons is considered beneficial for neuronal representation of information and is maximized in a critical state where neurons, on average, activate only one downstream neuron during activity cascades termed neuronal avalanches. The probability to recruit increasing fractions of a network decays with a power law during such a critical branching process. In contrast, sub- and supercritical propagation entails fading or expansive recruitment, both diminishing the capacity for representation. Numerous publications suggest that neuronal networks self-organize to achieve criticality. Analyses of activity within cortical microcolumns in vivo and within clusters in vitro, however, indicate supercritical recruitment. It thus remains unsolved how networks would fine-tune synaptic connectivity to a singular critical point.
A solution in this context could be a modular network structure to establish a configurational corridor supporting critical dynamics. To address this, we analyzed how mesoscale clustering and modularity in neuronal networks influenced avalanche dynamics and the apparent branching process. Large avalanches were overrepresented in homogeneous networks with explosive recruitment of recording sites, whereas small avalanches dominated strongly clustered networks, corresponding to supercritical respectively subcritical activity propagation. Avalanche size distributions displayed power law scaling, indicating critical activity propagation only in moderately clustered networks.
This could be explained if supercritical branching dominates highly recurrent parts of a network, as in clusters, but that modularity regulates propagation across clusters with a lower mesoscale branching index. In a specific experiment, avalanche statistics would then reflect local or mesoscale propagation, depending on the spatial sampling relative to the network structure. We propose that modularity tunes supercritical neuronal networks towards mesoscale-criticality.
About the speaker and his research
Hosted by
