Leijun Ye | Searcharxiv

arXiv2024

Simulation and assimilation of the digital human brain

Here, we present the Digital Brain (DB), a platform for simulating spiking neuronal networks at the large neuron scale of the human brain based on personalized magnetic-resonance-imaging data and biological constraints. The DB aims to reproduce both the resting state and certain aspects of the action of the human brain. An architecture with up to 86 billion neurons and 14,012 GPUs, including a two-level routing scheme between GPUs to accelerate spike transmission up to 47.8 trillion neuronal synapses, was implemented as part of the simulations. We show that the DB can reproduce blood-oxygen-level-dependent signals of the resting-state of the human brain with a high correlation coefficient, as well as interact with its perceptual input, as demonstrated in a visual task. These results indicate the feasibility of implementing a digital representation of the human brain, which can open the door to a broad range of potential applications.

q-bio.NC↗

arXiv2022

Quantifying the attractor landscape and transition path of distributed working memory from large-scale brain network

Many cognitive processes, including working memory, recruit multiple distributed interacting brain regions to encode information. How to understand the underlying cognition function mechanism of working memory is a challenging problem, which involves neural circuit configuration from multiple brain regions as well as stochastic transition dynamics between brain states. The energy landscape idea provides a tool to study the global stability and stochastic transition dynamics in the distributed cognitive function system. However, how to quantify the energy landscape in a realistic large-scale brain network remains unclear. Here, based on an anatomically constrained computational model of large-scale macaque cortex, we quantified the underlying multistable attractor landscape of distributed working memory. In the absence of external stimulation, the landscape exhibits three stable attractors, a spontaneous state, and two memory states. In the attractor landscape framework, the working memory function is governed by the change of landscape topography and the switch of system state according to the task requirement. The barrier height inferred from landscape topography quantifies the global stability of memory state and robustness to non-selective random fluctuations and distractor stimuli. The kinetic transition path identified by the minimum action path approach reveals that the spontaneous state serves as an intermediate state during the switch between the two memory states, the memory stored in the cortical area with higher hierarchy is more stable, and information flow follows the direction of hierarchical structure. These results provide new insights into the underlying mechanism of distributed working memory function, and the landscape and kinetic path approach can be applied to other cognitive function-related problems in brain networks.

q-bio.NC↗