Dynamic Cross-Frequency Coupling Reveals Taskdependent Neural Engagement During Varying Cognitive Demands Publisher



Sajadi SS ; Fazli B ; Karbasi F ; Garosi E ; Jalilian M ; Hosseinzadeh S ; Jafari AH ; Zakerian SA
Source: 32nd National and 10th International Iranian Conference on Biomedical Engineering, ICBME 2025 Published:2025

Abstract

Phase-amplitude coupling (PAC), reflecting crossfrequency interactions between neural oscillations, underlies cognitive processing, attention, and working memory. This study examined workload-dependent PAC across prefrontal, frontal, parietal, temporal, motor regions during low-, medium, and high-load cognitive tasks compared to baseline using Electroencephalography (EEG) of 17 healthy participants. PAC was quantified via comodulograms and regional peak values, focusing on theta-gamma and beta-gamma interactions. Baseline PAC was minimal across regions (mean MI 0.070.09). Low-load tasks elicited significant theta-gamma PAC in prefrontal (0.40) and frontal (0.32) regions, with minor increases in parietal (0.18) and temporal (0.12) cortices. Medium-load tasks recruited parietal cortex more strongly (0.45) alongside prefrontal (0.60) and frontal (0.58) regions, while temporal PAC rose to 0.25. High-load tasks induced robust, multi-frequency PAC: prefrontal (0.85), parietal (0.82), frontal (0.75), temporal (0.50), and motor (0.30) regions remained low, while occipital PAC (0.10). A repeated-measures ANOVA revealed significant effects of workload (F=15.23, p-value <0.001), region (F=18.46, p-value <0.001), and their interaction (F=20.90, p-value <0.001). These region- and frequency-specific patterns reflect executive control, attention, working memory, and motor planning. The findings demonstrate that PAC is significantly sensitive to cognitive workload and highlight it as a potential index of neural engagement during complex tasks. © 2025 IEEE.