Navigation is a basic human skill that is closely related to cognitive functioning. Excessive cognitive load may affect mental and behavioral states. This research is dedicated to meticulous deep feature extraction and classification analysis of electroencephalographic (EEG) signals generated during navigation tasks. The aim is to accurately classify and identify cognitive load, thus providing a solid theoretical foundation and practical guidance for improving the safety of navigational behavior in future work. In this paper, EEG signals were collected from 41 subjects at different cognitive load states during navigation tasks, and a navigation dataset was constructed. Power Spectral Density (PSD) and Phase Locking Value (PLV) were used at the neurophysiology level to delineate differences in cognitive load between subjects performing three navigation tasks. Meanwhile, a convolutional neural network (CNN)-based time-space multi-scale composite network (TSMC-Net) was proposed to classify cognitive load. The network provides a direct output of cognitive load. Compared with three typical deep machine learning models such as EEGNet, the experimental results generated by TSMC-Net outperformed the other three methods. Visualization results showed that TSMC-Net classified human cognitive load and extracted signal features efficiently. These findings provide a robust theoretical foundation for understanding cognitive load during navigation tasks and offer practical guidance for enhancing navigational safety and performance in future applications.
Liu et al. (Sat,) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: