NeuralFieldTheoriES · Towards a neural field theory for spiking neuron networks with electrical synapses
Horizon 2020 — Marie Skłodowska-Curie Actions
- Duration
- 2022-01-17 → 2024-01-16
- EU contribution
- €160,932
- Participants
- 1
- Scheme
- MSCA-IF
Lines connect the coordinator with its partners.
Results in brief
Towards a neural field theory for spiking neuron networks with electrical synapses
The human brain is probably the most complex system we know. Tremendous efforts and progress in neuroscience research has been made since Hans Berger’s discovery of the alpha rhythm in 1924. But a central question remains unanswered: how do the billions of interacting neurons in our brain generate macroscopic phenomena such as the coherent steering of muscles in locomotion, grasping, pattern recognition, working memory or decision making? Experimental neuroscience has provided ample evidence that these brain functions—and their pathological dysfunction—come along with characteristic neural activity patterns and brain rhythms. Neuronal synchrony is at the heart of perceiving and processing information as well as exchanging it across brain regions, which requires a careful balance of excitation (E) and inhibition (I). Disturbing the E-I balance can critically alter communication pathways and lead to, e.g., excessive beta-band oscillations in Parkinson’s disease or spatially propagating epileptic seizures in form of traveling waves. An increasing number of recent experimental findings point out that electrical synapses are ubiquitous across brain regions and species. Electrical synapses are broadly present between inhibitory neurons and widely believed to promote synchrony. They are therefore immediately implicated in the generation of brain rhythms, they directly affect the E-I balance and are thus critical for the functioning of our brain. But their overall functional role remains elusive, and how exactly they interact with inhibitory synapses to generate oscillations is unknown. In this multidisciplinary project, we strive for providing a comprehensive picture how electrical synapses contribute to and shape large-scale brain dynamics. We will develop the mathematical mean-field theory that allows for a full mechanistic understanding of the neuronal and network mechanisms underlying the emergence of spatio-temporal structures in large networks of neurons with both chemical and electrical synapses. In particular, we will address three objectives: (a) Understand symmetry breaking in two populations of inhibitory neurons with electrical synapses. (b) Derive an exact neural field model for neurons with electrical synapses with realistic spatial connectivity. (c) Develop a novel canonical neural field model with excitation, inhibition and electrical synapses, and subsequently incorporate more degrees of biological realism.
Data: CORDIS, © European Union
Project objective
A major challenge in statistical physics, nonlinear dynamics and theoretical neuroscience over the last half century has been to understand the self-organizing principles governing the dynamics of large networks of neurons. Physicists and applied mathematicians have proposed simple mean-field descriptions of spatially-extended neural networks in terms of a relevant macroscopic observable, the firing rate. This approach has been particularly successful and so-called Neural Field Models (NFM) have become an extremely popular mathematical tool in neuroscience, physics and applied mathematics. Yet, to date, mean-field theories describe networks with chemical synapses, but it remains a major theoretical challenge to incorporate electrical synaptic interactions in such mathematical descriptions. Recently, a mean-field theory for large networks of spiking neurons has been proposed, which exactly links the dynamics of single neurons with that of two mean-field variables: The firing rate and the mean membrane potential. Remarkably, this theory permits to incorporate electrical interactions, but the mathematical derivation and the analysis of the dynamics of the first NFM is lagging. This project proposes the formal mathematical derivation of such NFM, as well as the thorough analysis of its dynamics and bifurcations. Towards this goal, at the host institution UPF in Barcelona, the ER will apply mean-field methods and nonlinear dynamical systems theory to derive the novel NFM (which we conjecture is of reaction-diffusion type). During a secondment at VU Amsterdam, the ER will be trained to become an expert in numerical analysis of partial differential equations, which will further allow him to perform extensive state-of-the-art computer simulations. The expected results will provide completely novel mechanistic insights on the emergence of complex spatio-temporal patterns of neuronal activity due to the intricate interplay between chemical and electrical synapses.
Original text from CORDIS.
Participants
- UNIVERSIDAD POMPEU FABRA · BarcelonaCoordinatorSpain
Links
Data: CORDIS, © European Union