From Cortical Synchronous Rhythm to Brain Inspired Learning Mechanism: An Oscillatory Spiking Neural Network with Time-Delayed Coordination

Researchers Tingting Dan and Guorong Wu have introduced S2-Net (spiking-by-synchronization neural network), a novel brain-inspired learning mechanism that uses oscillatory synchronization to coordinate neural spiking. Unlike standard spiking neural networks that rely solely on firing rates, S2-Net incorporates precise spike timing governed by brain-like rhythms. It models each cortical region or image pixel as a spiking neuron embedded in a predefined connectivity scaffold. Through iterative bottom-up and top-down interactions, the network self-organizes spiking activity over a finite memory window, forming oscillatory synchronization without requiring global phase locking. This time-delayed synchronization formulation allows heterogeneous neural spiking across large-scale distributed systems, mimicking the partial and transient synchrony observed in real brains.

The authors report promising results across multiple tasks, including neural activity decoding, energy-efficient signal processing, temporal binding, and semantic reasoning. S2-Net achieves efficient information processing by using rhythmic timing as a control mechanism, reducing energy consumption compared to traditional deep learning models. The paper (19 pages, 6 figures) is available on arXiv (arXiv:2605.01656) and represents a significant step toward biologically plausible AI that can handle temporal dynamics and distributed computation more naturally.