DriE-Cog dataset: A multimodal physiological–behavioural dataset for intelligent driving emergency response

Current datasets for intelligent emergency response driving and safety are limited in both quantity and quality, hindering the development of reliable and robust driving safety systems. To fill this gap, we introduce DriE-Cog, a multimodal collection of physiological, cognitive, and behavioural data for driving emergency response. DriE-Cog includes data from 51 participants across 4 common driving scenarios, each consisting of 12 emergency driving events. DriE-Cog integrates data from eye tracking (ET), electroencephalography (EEG), photoplethysmography (PPG), galvanic skin response (GSR), and driving behaviour. We validated the dataset by examining its completeness, assessing single-modal feature differences, and evaluating the classification performance of multimodal data. This dataset not only provides a reliable foundation for in-depth studies on intelligent driving emergency response but also supports improvements in driver safety performance and operational stability.