guest ::
| Home > Publications database > Multi-scale Spiking Network Model of Human Cerebral Cortex |
| Home > Publications database > Multi-scale Spiking Network Model of Human Cerebral Cortex |
| Poster (Other) | FZJ-2025-02699 |
; ; ; ; ; ; ;
2025
This record in other databases:
Please use a persistent id in citations: doi:10.34734/FZJ-2025-02699
Abstract: Data-driven models at cellular resolution exist for various brain regions, yet few for human cortex. We present a comprehensive point-neuron network model of a human cortical hemisphere that integrates diverse experimental data into a unified framework bridging cellular and network scales [1]. Like a previous large-scale spiking model of macaque cortex [2,3], our work investigates how resting-state activity emerges in cortical networks.The model represents one hemisphere via the Desikan-Killiany parcellation (34 areas), with each area implemented as a 1 mm² microcircuit that distinguishes cortical layers. It aggregates multimodal data, including electron microscopy for synapse density, cytoarchitecture from the von Economo atlas [4], DTI-based connectivity [5], and local connection probabilities from the Potjans-Diesmann microcircuit [6]. Human neuron morphologies [7] guide layer-specific inter-area connectivity. The full-density model, comprising 3.47 million leaky integrate-and-fire neurons and 42.8 billion synapses, was simulated using NEST on the JURECA-DC supercomputer.Simulations show that equal strength for local and inter-area synapses yields asynchronous irregular activity that deviates from experimental observations. When inter-area connections are strengthened relative to local synapses, both microscopic spiking statistics from human medial frontal cortex and macroscopic resting-state fMRI correlations are reproduced [8]. In the latter scenario, consistent with empirical findings during visual imagery [9], sustained activity flows primarily from parietal through occipital and temporal to frontal areas.This open-source model enables systematic exploration of structure-dynamics relationships. Future work may leverage the Julich-Brain Atlas to refine the parcellation and incorporate detailed cytoarchitectural and receptor data [10]. The model code is available at https://github.com/INM-6/human-multi-area-model.
|
The record appears in these collections: |
PDF