Motion sickness is a common problem when traveling. Research has shown that nearly 1/3 of the population suffers from motion sickness when travelling by sea, land and air. The pathogenesis of motion sickness is complex with no unified conclusions. The sensory conflict hypothesis holds that motion sickness is mainly caused by mismatches of vestibule, vision and proprioception. Since a driver does not need to operate a car when using automatic driving, the feeling conflict is intensified and the automatic driving is less comfortable. This study investigates physiological indexes which can be used to quantitatively evaluate motion sickness. The experiments used a 6-DOF simulator platform combined with a virtual reality (VR) system to simultaneously provide visual and vestibular stimulation to the subjects. The subjective motion scores and EEG (electroencephalogram) signals of 11 healthy subjects were recorded during automatic driving and active driving scenarios to compare the motion responses of the subjects during the two driving scenarios. Analyses of the motion scores and the EEG records show that the subjects' motion scores were 2 points higher during automatic driving than during active driving with increases of motion sickness related to increases in the gravity frequency based on the power spectral density of the θ waves in the motor center (FC2, Cz), sensory center (CP5, P3) and visual center (POz) of the brain during automatic driving. The paired t test showed correlation between the gravity frequency differences based on the power spectral density of the subject during active driving and automatic driving (p < 0.05). The results indicate that subjects are more likely to develop motion sickness during automatic driving and that EEG signals can be used to quantitatively evaluate the degree of motion sickness.