Building on models—a perspective for computational neuroscience

2025/11/01 by Hans Ekkehard Pleßer, Andrew P. Davison, Markus Diesmann +13 · 1 voice
Engineering · Neuroscience · #Advanced Memory and Neural Computing #Neural dynamics and brain function #Ferroelectric and Negative Capacitance Devices

paper · doi:10.1093/cercor/bhaf295

openalex publication_date 2025/11/01 · openalex created_date 2025/11/09 · openalex updated_date 2026/07/31

Abstract

Neural circuit models are essential for integrating observations of the nervous system into a consistent whole. Public sharing of well-documented codes for such models facilitates further development. Nevertheless, scientific practice in computational neuroscience suffers from replication problems and little re-use of circuit models. One exception is a data-driven model of early sensory cortex by Potjans and Diesmann that has advanced computational neuroscience as a building block for more complex models. As a widely accepted benchmark for correctness and performance, the model has driven the development of CPU-based, GPU-based, and neuromorphic simulators. On the 10th anniversary of the publication of this model, experts convened at the Käte Hamburger Kolleg Cultures of Research at RWTH Aachen University to reflect on the reasons for the model's success, its effect on computational neuroscience and technology development, and the perspectives this offers for the future of computational neuroscience. This report summarizes the observations by the workshop participants.

Citations

• NEURONAL CIRCUITS OF THE NEOCORTEX
• Practical Bayesian model evaluation using leave-one-out cross-validation and WAIC
• Sub-realtime simulation of a neuronal network of natural density
• Software in science is ubiquitous yet overlooked
• Foundation model of neural activity predicts response to new stimulus types
• Constructive community race: full-density spiking neural network model drives neuromorphic computing
• Selecting fitted models under epistemic uncertainty using a stochastic process on quantile functions
• Mapping Input Noise to Escape Noise in Integrate-and-fire neurons: A\n Level-Crossing Approach
• Inference of a mesoscopic population model from population spike trains
• Estimating the Dimension of a Model

Discussions

• Driving science, sharing and technology --- a review of the impact of a cortical microcircuit model on the fields of computational neuroscience and neuromorphic computing, a decade after its publicati [bsky, 3 points, 0 comments]

Related

• Constructive community race: full-density spiking neural network model drives neuromorphic computing
• The brain versus AI: World-model-based versatile circuit computation underlying diverse functions in the neocortex and cerebellum
• A Survey of Neuromorphic Computing and Neural Networks in Hardware
• Simple Cortex: A Model of Cells in the Sensory Nervous System
• Modeling the Mind: A brief review