Multimodal Cognitive Distracted Driving Dataset

Multimodal Cognitive Distracted Driving DatasetOverviewThis dataset is designed for research on driver cognitive distraction, one of the four core types of driving distraction defined by the U.S. National Highway Traffic Safety Administration (NHTSA). Unlike most existing driver monitoring systems that focus only on visible visual distraction, this dataset targets the hard-to-detect cognitive distraction in real-world driving, providing synchronized, high-quality multimodal data for distraction detection algorithm development and driving safety research.Key HighlightsSynchronized multimodal data covering neurophysiological, autonomic, and driving behavioral metricsStandardized 3-level graded cognitive distraction paradigm with verified load gradient28 valid licensed driver participants with diverse age and driving experienceStructured following the Brain Imaging Data Structure (BIDS) specification for easy reuseCollected in a high-fidelity simulated driving environment with full experimental controlExperimental DesignParticipants28 healthy licensed drivers (15 male, 13 female; mean age 29.53 years; mean driving experience 3.64 years) were recruited for the experiment.Equipment & ScenarioDriving simulation: Unity-based driving platform with Logitech G29 hardware, 3-screen 120° field of view display, 60 Hz sampling ratePhysiological acquisition: ErgoLAB 32-channel EEG system (256 Hz), wireless PPG and EDA wrist sensors (64 Hz)Simulated scenario: 20 km ring highway modeled after Beijing 3rd Ring Road, with standard lane settings, constant traffic flow, and complete road signageTask ProtocolPrimary driving task: Participants drove at a constant 100 km/h in the right lane, with lane-keeping as the core requirementSecondary cognitive task: Serial oral addition task with 3 difficulty levels (easy/medium/hard) to induce graded cognitive load, with order counterbalanced via Latin square designFull procedure: Resting baseline recording → simulator familiarization → single-task driving baseline → dual-task cognitive distraction blocks → post-block subjective distraction ratingData IncludedNeurophysiological data: 32-channel EEG signalsAutonomic physiological data: PPG (heart rate/HRV) and EDA (skin conductance) signalsDriving behavioral data: Steering wheel angle, lane position, speed, acceleration, and driving event logsAnnotation data: Precise event timestamps for all experimental phases, task difficulty labels, and subjective distraction rating resultsApplication ScenariosDevelopment and validation of cognitive distraction detection algorithmsResearch on neurophysiological and autonomic mechanisms of driving distractionOptimization of driver monitoring systems for intelligent cockpitsHuman-machine interaction safety design for autonomous driving systemsTraffic safety research on driver cognitive state