Predicting driver distraction using a single channel ear EEG

Cognitive distraction poses a major risk to driving safety, yet remains difficult to detect in real time because it often lacks overt behavioral markers. Neurophysiological measures such as electroencephalography (EEG) provide objective indices of cognitive load, but conventional EEG systems are impractical for real-world deployment due to their intrusiveness and computational demands. The present study examined whether cognitive distraction during naturalistic driving can be decoded using a single-channel, non-invasive in-ear EEG, and how its temporal dynamics compare with those obtained from full-cap scalp EEG and behavioral measures. Twenty-seven participants performed a dual-task paradigm in a highly immersive driving simulator, combining continuous vehicle control with low- and high-load arithmetic tasks presented on an in-vehicle display. Single-channel in-ear EEG, 24-channel scalp EEG, eye movements, and head rotation were recorded concurrently. Time-resolved multivariate pattern analysis was applied to decode working-memory load with millisecond precision across modalities. Cognitive distraction was reliably decoded from in-ear EEG, with detection latency and temporal generalization profiles closely matching those of full-cap scalp EEG. Although peak decoding performance was higher for scalp EEG, the timing and temporal stability of distraction-related neural signatures were largely overlapping between the two neural modalities. Eye velocity provided the earliest and most sensitive behavioral marker of distraction, while head rotation contributed complementary but weaker information. Scalp EEG topographies indicated that neural signals underlying decoding were closely linked to oculomotor and visuomotor processes engaged during task performance. These findings demonstrate that single-channel in-ear EEG provides a temporally precise and low-burden neural marker of cognitive distraction during driving. By prioritizing early detection and minimal hardware over maximal classification accuracy, the results identify a practical operating point for wearable EEG-based driver monitoring systems and support the feasibility of fast, off-the-shelf decoding approaches for real-world applications.