An Electroencephalography (EEG) Dataset for Understanding Driving Expertise from Naturalistic Urban Road Experiments

Modern autonomous driving algorithms in complex urban environments strive for safety, comfort, and intelligence, qualities exemplified by expert drivers. Despite expert drivers serving as ideal models for algorithmic imitation, a critical gap remains in understanding the neural and cognitive mechanisms underlying their decision-making. This paper presents a comprehensive EEG dataset comparing 10 expert and 10 novice drivers in 13 naturalistic urban driving conditions. Our multi-modal dataset synchronizes brain activity with vehicle CAN bus data, traffic participant information, and psychophysiological measures (Electrodermal Activity and Heart Rate) from drivers. Uniquely, we also collected physiological and subjective feedback from two passengers per trip to validate driving performance quality. All participants(Driver and Passengers)completed a series of standardized subjective questionnaires at pre- and post-experiment. Finally, they participated in post-drive semi-structured interviews exploring driver decision-making processes and passenger experience. This novel dataset enables researchers to decode the neural signatures underlying driving expertise, providing valuable insights for developing more human-like, intelligent autonomous driving algorithms that can better navigate the complexities of urban traffic environments.