Artificial Deep Neural Networks (DNNs) and, more recently, large-scale foundation models have made breakthrough progress drawing loose structural and functional inspiration from the brain. The pace of advance has reshaped what is possible — and what is distinctive — about artificial intelligence. Yet how seemingly intelligent performance is achieved still differs in fundamental ways from biological cognition, and the dialogue between neuroscience and AI has never been more consequential.

This editorial project focuses on the interface of neuro-inspired computation, launched in conjunction with the fourth Neuro-inspired Computation course held at the University of Tokyo International Research Center for Neurointelligence (WPI-IRCN), 22–25 July 2026.

Neurointelligence spans broad areas of mutual interest to both communities, network architectures, intrinsic and latent dynamics, plasticity and criticality, predictive coding and world models, multi-agent and social learning, neuromodulation, embodied and developmental learning, and the emergence of cognition in biological and artificial agents alike. Recent progress in large-scale recording, AI-driven behavioral quantification, mechanistic interpretability, and energy-efficient spiking and neuromorphic computation has opened concrete opportunities to test theories of intelligence across substrates.

All course participants are strongly encouraged to submit their work to the Research Topic, while the project is also open to non-conference attendees whose research speaks to the same questions.

As biological and artificial intelligence have much to learn from each other, this collection aims to capture the expanding global excitement in this innovative research frontier where neuronal network function in nature and in silico converge.

We welcome all article types, including original research, methods, perspectives, reviews, case reports, and FAIR² Data Articles describing curated, AI-ready datasets that underpin neuro-inspired computation research. Toward this aim, we welcome contributions addressing, but not limited to:

• network architectures (biological and artificial)

• intrinsic and latent neural dynamics

• E–I balance, homeostasis, and criticality

• synaptic plasticity, meta-plasticity, and learning rules

• predictive coding, world models, and self-supervised representation learning

• continual, lifelong, and developmental learning

• reinforcement learning and neuromodulation

• multi-agent, social, and theory-of-mind learning

• cognitive architectures and embodied / developmental robotics

• AI-driven behavioral quantification and closed-loop neuroscience

• mechanistic interpretability of biological and artificial circuits

• spiking, neuromorphic, and energy-efficient computation

• foundation models of brain and behavior, and brain–AI alignment

• human–AI collaboration

Are you interested in contributing a manuscript to this Research Topic?

Neuro-inspired Computation 2026 — Towards FAIR Neuro-AI Data