Impact of Temperament Types and Anger Intensity on Drivers' EEG Power Spectrum and Sample Entropy: An On-road Evaluation Toward Road Rage Warning

Abstract

Road rage", also called driving anger, is becoming an increasingly common phenomenon affecting road safety in auto era as most of previous driving anger detection approaches based on physiological indicators are often unreliable due to the less consideration of drivers' individual differences. This study aims to explore the impact of temperament types and anger intensity on drivers' EEG characteristics. Thirty-two drivers with valid license were enrolled to perform on-road experiments on a particularly busy route on which a variety of provoking events like cutting in line of surrounding vehicle, jaywalking, occupying road of non-motor vehicle and traffic congestion frequently happened. Then, muti-factor analysis of variance (ANOVA) and post hoc analysis were utilized to study the impact of temperament types and anger intensity on drivers' power spectrum and sample entropy of θ and β waves extracted from EEG signals. The study results firstly indicated that right frontal region of the brain has close relationship with driving anger. Secondly, there existed significant main effects of temperament types on power spectrum and sample entropy of β wave while significant main effects of anger intensity on power spectrum and sample entropy of θ and β wave were all observed. Thirdly, significant interactions between temperament types and anger intensity for power spectrum and sample entropy of β wave were both noted. Fourthly, with the increase of anger intensity, the power spectrum and sample entropy both decreased sufficiently for θ wave while increased remarkably for β wave. The study results can provide a theoretical support for designing a personalized and hierarchical warning system for road rage.