Directional brain-heart interaction during cognitive workload: a permutation conditional mutual information approach

The functional interaction between the central nervous system and autonomic cardiovascular regulation, commonly referred to as Brain-Heart Interaction (BHI), has gained increasing attention in neurology and psychology through the joint analysis of Electroencephalography (EEG) and Electrocardiography (ECG) signals. However, the nonlinear and time-varying nature of EEG and ECG signals requires approaches that can capture directional interactions between physiological systems. In this study, we investigated the use of Permutation Conditional Mutual Information (PCMI) for the characterization of directional BHI. First, the method was evaluated on synthetically generated coupled time series to assess its ability to detect the dominant direction of interaction. Next, PCMI was applied to simultaneous EEG and ECG recordings acquired during baseline and mental arithmetic tasks from a cohort of 36 subjects. The results show that BHI complexity increases during mental arithmetic tasks compared with baseline, supporting the role of cognitive workload in modulating brain–heart dynamics. These changes primarily involved anterior and posterior scalp regions, and affected both vagal- and sympathovagal-related HRV components. These findings demonstrate that PCMI provides a framework for studying directional BHI during cognitive tasks, and highlight its potential translational value for investigating broader brain–body interactions.