Reliability Analysis of Driving Behaviour in Road Traffic System Considering Synchronization of Neural Activity

This paper aims to disclose the reliability of driving behaviour in road traffic system. For this purpose, the drivers' electroencephalography (EEG) signals were collected with Emotiv, a portable device, and used for an experiment in actual driving environment. Through the analysis on the synchronization of 14-channel EGG signals, the author identified a proper threshold, and determined whether the brain network nodes are connected or not. On this basis, a brain network model was created for the drivers. The driving behaviour reliability of the drivers was discussed in detailed considering brain network parameters (clustering coefficient and global efficiency) and the power spectrum features of EEG signals. The research results show that, with the increase in driving time, the intercity drivers became increasingly fatigued and their brain network continued to densify, pushing up the network parameters like clustering coefficient and global efficiency. In this case, the neuronal activities became increasingly synchronized across the brain regions. In addition, the two brain network parameters of the drivers were less discrete and more accurate than the fatigue indicator of EEG power spectrum features. Therefore, the analysis of brain network parameters is a precise and feasible method for discussing driving behaviour reliability.