On November 22, a groundbreaking advancement was made by a research team helmed by Professor Yu Qiang from the School of Artificial Intelligence at Tianjin University. Working in tandem with international researchers, the team made significant strides in the study of neural network information processing mechanisms.

Concentrating on synapses—the fundamental building blocks of brain neural networks—this research has, for the very first time, unveiled the core mechanism responsible for processing spatiotemporal information. The related research findings were published in the prestigious international academic journal, Proceedings of the National Academy of Sciences (PNAS).

Within the human brain, billions of neurons transmit and process information in the form of electrical pulses via synapses. Synapses exhibit two crucial regulatory capabilities: 'long-term plasticity' and 'short-term plasticity.' Through the construction of a theoretical model of synaptic computation and learning, the research team uncovered that when 'long-term plasticity' influences 'short-term plasticity,' the brain can transform time-series information into spatial expression patterns. This transformation substantially boosts the memory capacity, anti-interference capability, and complex spatiotemporal information recognition ability of neural networks.

The validity of this model has been confirmed through electrophysiological observations of synapses in the cerebral cortices of both mice and humans, showcasing a high degree of biological plausibility.