Neural Diversity and Computation – Towards a Mathematical Account of Tissue Heterogeneity in the Brain

Organizers

Richard Gast | Northwestern University, USA
Jeremie Lefebvre | University of Ottawa, Canada

Abstract

Recent technological developments in cell type identification, synaptic tracing, and multi-neuron recordings have led to massive data sets for cell-type-resolved brain structure and dynamics. These data sets have not only identified new cell types, but also revealed a previously underappreciated level of neural and synaptic heterogeneity within cell types. In contrast, most mathematical models of brain function focus on networks with a few classes of neurons (e.g. one excitatory and one inhibitory), where neurons within a particular class are considered as identical. This workshop will address how different aspects of brain heterogeneity can be expected to affect neural dynamics and computation. Together, we will discuss (i) how the heterogeneity gap between neural network models and biological neural networks may bias our theories of structure-function relationships in the brain, and (ii) how various aspects of brain heterogeneity may interact to achieve stable brain function. In the first session, we will gain insight into the wealth of synchronization patterns that can arise in complex networks and how they are affected by structural heterogeneity. We will then learn about the level of neural heterogeneity expressed in human cortex, and discuss how the results from complex network studies may translate to brain networks. In the second session, we will move on to discuss different approaches that have been employed to explain how neural heterogeneity supports computations in spiking neural networks, supported by evidence from computation with heterogeneous neuromorphic devices. In the final session, we will discuss a range of theoretical results that connect different heterogeneity-inducing aspects of neural networks to emergent network dynamics and computation. We will conclude the last session with an extended discussion of how the various heterogeneities that exist in the brain may serve distinct purposes in the organization of the brain dynamics and function.

Schedule (CEST)

Sunday, Sep 29

14:00

Richard Gast | Northwestern University, USA
Introduction to the topic and goals of the workshop

14:05

Adilson Motter | Northwestern University, USA
Heterogeneity-Promoted Synchronization in Complex Networks

14:40

Anna Zakharova | Technische Universität Berlin, Germany
Partial synchronization patterns in neural networks

15:15

Henrike Planert | Charité – Universitätsmedizin Berlin, Germany
Cellular and synaptic diversity of layer 2-3 pyramidal neurons in human individuals

15:50

Ignite talk by Richard Gast | Northwestern University, USA
Central heterogeneity-related questions for neuroscience

16:00

Coffee break

16:30

Ann Kennedy | Northwestern University, USA
Neural heterogeneity controls computations in spiking neural networks

17:10

Melika Payvand | University of Zurich, Switzerland
Neural computation with heterogenous analog substrates

17:50

Rishikesh Narayanan | Indian Academy of Sciences, India
Diverse roles of heterogeneities in neural circuits

18:30

End of first day

Monday, Sep 30

08:30

Introduction by Jeremie Lefebvre | University of Ottawa, Canada
Recap of session I + II + workshop goals and introduction to session III

08:50

Axel Hutt | National Institute for Research in Digital Science and Technology, France
Neural heterogeneity may induce brain resilience

09:25

Fred Wolf | Max Planck Institute for Dynamics and Self-Organization, Germany
Emergent firing rate heterogeneity in balanced cortical networks – from biophysical control to computational function

10:00

Coffee break

10:30

David Dahmen | Forschungszentrum Jülich, Germany
Effect of synaptic heterogeneity on neuronal coordination

11:05

Luca Mazzucato | University of Oregon, USA
The importance of heterogeneity for flexible cortical computation

11:40

Panel discussion led by Jeremie Lefebvre | University of Ottawa, Canada
Towards mathematical models of brain function that account for the inherent heterogeneity of neural tissue

12:30

End