Dynamic analysis of frequency specificity in multilayer brain networks

Abstract

The brain is a highly complex and delicate system, and its internal neural processes are manifested as the interweaving and superposition of multi-frequency neural signals. However, traditional brain network studies are often limited to the whole frequency band or a specific frequency band, ignoring the potentially profound impact of the diversity of information within the frequency on the dynamics of brain networks. To comprehensively and deeply analyze this phenomenon, the present study is devoted to exploring the specific performance of brain networks at different frequencies. We used the maximum overlap discrete wavelet transform technique to finely divide the time series data into the following frequency bands: scale 1 (0.125–0.25 Hz), scale 2 (0.06–0.125 Hz), scale 3 (0.03–0.06 Hz) and scale 4 (0.015–0.03 Hz). Based on these frequency bands, we constructed multilayer networks from both dynamic and static perspectives, respectively. From the dynamic perspective, we quantitatively evaluated the dynamic differences among different frequency bands using metrics such as flexibility, promiscuity, integration, and recruitment, and found that scale 3 and scale 4 bands performed particularly well. In contrast, from a static perspective, we measured the cross-frequency interaction capability between different frequency bands through metrics such as multilayer clustering coefficient and entropy of multiplexing degree, and the results show that scale 2, scale 3, and scale 4 band networks have enhanced global integration capability and local capability. In addition, we explored the correlation of gender and age with the properties of brain networks in different frequency bands. In the scale 1 frequency band, the organization of brain functions showed significant gender differences, while in the scale 2 frequency band, there was a significant correlation between age and global efficiency. By integrating the dual perspectives of time and frequency domains, this study not only reveals the critical role of frequency specificity in the brain’s information processing and functional organization but also provides new perspectives for understanding the complex working mechanisms of the brain as well as gender- and age-related cognitive differences.