| Summary: | Human factors account for the majority of vehicular accidents. Driving is a cognitive activity that requires total concentration, where even a small lapse can lead to fatal consequences. Some of the driver behaviors which have been shown to be factors that contribute to accidents include alcohol and drug consumption, driving in fatigued condition, and the use of cell phone. Emotional state of the people involved is another factor that affects attention and, consequently, driving behavior. Depression, anxiety, and stress are common psychiatric disorders, which can impact one's driving behavior. In this study, we proposed to analyze the driving behavior with respect to emotional state using electroencephalogram (EEG) signals and a driving simulator. These brain signals were correlated with the subjects' mood measured in terms of the Depression Anxiety Stress Scales (DASS-21). EEG results were recorded from four subjects while they performed their driving tasks on the driving simulator. The feature extraction was obtained using a Mel-frequency cepstral coefficients (MFCC) and multilayer perceptron (MLP) as the clasification technique. The results confirmed that the emotional state under the driving task is highly correlated with the DASS-21 psychometric analysis results.
|
|---|