Moritz Helias: Statistical physics of correlated neuronal variability

Forschungszentrum Jülich

Neuronal networks can be considered as many particle systems with interesting physical properties: They operate far from thermodynamic equilibrium and show correlated states of collective activity [1]. We here discuss recent progress in understanding the structure of these correlated states by methods from statistical physics and disordered systems [2,3,4].

Our analysis shows that the heterogeneity of the network connectivity enables critical dynamics that unfolds in a low-dimensional subspace. The structure of correlations predicted by this theory is found in line with massively parallel recordings from motor cortex [2]. We then demonstrate that networks in such regimes possess optimal capacity to memorize past input sequences [3]. We find that they operate in a hitherto unreported regime that combines instability on short time scales with asymptotically non-chaotic dynamics. As an outlook, we demonstrate how methods from field theory [4] help us understand the interplay of non-linearities and fluctuations that is vital to neuronal network dynamics and function.

1. Dahmen, Bos, Helias (2016)
Correlated fluctuations in strongly coupled binary networks beyond equilibrium.
Phys. Rev. X 6, 031024

2. Dahmen, Grün, Diesmann, Helias (2019)
Second type of criticality in the brain uncovers rich multiple-neuron dynamics.
PNAS https://doi.org/10.1073/pnas.1818972116

3. Schuecker, Goedeke, Helias (2018)
Optimal sequence memory in driven random networks
Phys. Rev. X 8, 041029

4. Kuehn T, Helias M (2018)
Expansion of the effective action around non-Gaussian theories.
J Phys A: Math Theor 51, 375004

 

Guests are welcome!

 

Organized by

Benjamin Lindner / Margret Franke