Answers are grounded in this paper and cite their sources.
Spiking neural network models are increasingly incorporating new interaction types and experiment-related features which exceed the expressive scope of existing declarative modeling standards. To address this limitation, we present a provisional set of powerful, multi-purpose extensions to the leading NeuroML 2 specification. These extensions cover arbitrary synaptic and cellular variability, user-specified continuous-time interactions between any combination of mechanisms, event-and time-series streaming for closed-loop experiments, and alternative neuron/mechanism interactions among others–all the while fitting within the existing NeuroML ontology. To demonstrate the expressiveness and effectiveness of these extensions, we implemented them within the NeuroML-based EDEN simulator and used them to capture a diverse collection of highly relevant computational-neuroscience experiments, originally written for different simulators or as custom-made programs–in the process showcasing each of the proposed NeuroML extensions. This work both significantly expands EDEN's modeling capabilities in the present, and offers a reference framework for future NeuroML developments toward broader neuroscientific applicability.
No takes yet. Share an insight, caveat, or question.
Panagiotou et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: