Neural systems in the brain demonstrate remarkable computational abilities, including reliable perception, robust decision-making, continual learning, and flexible adaptation to changing environments. These capabilities, enabled by sparse, event-driven neural activity, have become an important inspiration for advancing artificial intelligence, especially in contexts requiring real-world robustness and efficiency.

Spiking neural networks (SNNs) provide a powerful framework for modeling the discrete and asynchronous information processing characteristic of biological neurons. Their temporal dynamics and sparse activation patterns have shown promise in reproducing cognitive functions and supporting energy-efficient computation. Recent breakthroughs demonstrate SNNs’ ability to enable low-power perception on edge devices, and increasing interest now lies in connecting SNNs with powerful Large Language Models (LLMs).

Event-based sensing technologies, such as neuromorphic and event cameras, capture rapid changes in a scene with microsecond resolution, generating streams of asynchronous spikes similar to neural signals. The resulting data, with ultra-low latency and high dynamic range, supports robust perception in challenging visual environments and new advances in optical measurement, high-speed analysis, and computational imaging. These innovations highlight the value of biologically inspired sensing and computation for building real-time, energy-efficient intelligent systems for both scientific and industrial applications.

Bridging neuroscience principles with deployable artificial systems represents a promising avenue for uncovering the mechanisms of brain function while also motivating robust, adaptive, and efficient AI. With this Research Topic, we aim to explore how computational models rooted in neuroscience, especially those focusing on spiking dynamics, learning, and perception, can impact artificial systems and reciprocally, how engineering insights can inform computational neuroscience.

We invite contributions that advance the intersection of computational neuroscience, neuromorphic engineering, and artificial intelligence. Areas of interest include, but are not limited to:
- Modeling with Spiking Neural Networks: Robust learning, biologically realistic inference mechanisms, and integration with other AI models (such as LLMs).
- Hardware acceleration for SNNs: GPU-based implementations, event-driven simulation engines, and new neuromorphic platforms.
- Neuromorphic computing: Design of hardware architectures and hardware–algorithm co-optimization inspired by neural computation.
- Bio-inspired learning algorithms: Approaches grounded in neural mechanisms—such as attention and homeostasis—for adaptive control and continuous learning.
- Hybrid neuro-AI architectures: Integrative frameworks that unite neuroscience models and machine learning approaches for perception, memory, and reasoning.
- Event-based sensing and computation: Algorithms and systems leveraging neuromorphic or event-based sensors for vision, measurement, robotics, and autonomous navigation.

We welcome original research articles, computational system demonstrations, theoretical proposals, and comprehensive reviews. Contributions that bridge neuroscience and artificial intelligence, explore new paradigms, or demonstrate translational and real-world impact of brain-inspired approaches are particularly encouraged.

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

Neuroscience-inspired application and systems