Understanding Neural Rhythmic Mechanisms Through Self-oscillations of Complex Neural Networks and Their Adaptation

This study delves into the self-oscillation properties of complex neural networks to elucidate the intrinsic mechanisms driving biological rhythm generation and adaptation to external periodic signals. We scrutinize the influence of electrical coupling and Spike-Timing-Dependent Plasticity (STDP) on network synchronization. Employing a neural network characterized by scale-free topology, we observe that neurons with higher node degrees necessitate activation by an increased number of fellow neurons. Central neurons emerge as pivotal in facilitating swift excitation propagation. In contrast, low-degree circuits sustain activity during burst intervals, with circuit length inherently dictating the rhythmic period. Notably, despite their inherent complexity and diverse rhythm generation, neural networks can adaptively select a low-degree loop congruent with external input rhythms via Hebbian learning principles. These insights offer profound implications for comprehending the variances in human biological rhythms across different environments and hold significant value for planning extended space expeditions.