Enhancing orderly signal propagation between layers of neuronal networks through spike timing-dependent plasticity

Electrical signal propagation in multi-layer neural networks plays a crucial physiological role. This study constructs a three-layer neural network to simulate the orderly propagation of signals between senses, exploring its neural basis. Comprising input, learning, and output layers, the network employs a voltage-based spike-timing- dependent plasticity (STDP) rule to control synaptic interactions, aiming to learn functional relationships to facilitate signal propagation. Research indicates the network can effectively transmit signals under suitable conditions, whether or not there are synaptic connections between the input and output layers. Statistical analysis shows that learning effects are more pronounced without input-output connection interference. Further scalability tests reveal that learning stability is maintained in large-scale networks without these connections; however, their introduction imposes stricter conditions for successful learning as network size increases. This study offers insights into how neural networks can mimic complex biological signal processing tasks.