Electrophysiology-based detection of emergency braking intention in real-world driving.

OBJECTIVE: The fact that all human action is preceded by brain processes partially observable through neuroimaging devices such as electroencephalography (EEG) is currently being explored in a number of applications. A recent study by Haufe et al (2011 J. Neural Eng. 8 056001) demonstrates the possibility of performing fast detection of forced emergency brakings during driving based on EEG and electromyography, and discusses the use of such neurotechnology for braking assistance systems. Since the study was conducted in a driving simulator, its significance regarding real-world applicability needs to be assessed. APPROACH: Here, we replicate that experimental paradigm in a real car on a non-public test track. MAIN RESULTS: Our results resemble those of the simulator study, both qualitatively (in terms of the neurophysiological phenomena observed and utilized) and quantitatively (in terms of the predictive improvement achievable using electrophysiology in addition to behavioral measures). Moreover, our findings are robust with respect to a temporary secondary auditory task mimicking verbal input from a fellow passenger. SIGNIFICANCE: Our study serves as a real-world verification of the feasibility of electrophysiology-based detection of emergency braking intention as proposed in Haufe et al (2011 J. Neural Eng. 8 056001).

EEG alpha spindles and prolonged brake reaction times during auditory distraction in an on-road driving study.

Driver distraction is responsible for a substantial number of traffic accidents. This paper describes the impact of an auditory secondary task on drivers' mental states during a primary driving task. N=20 participants performed the test procedure in a car following task with repeated forced braking on a non-public test track. Performance measures (provoked reaction time to brake lights) and brain activity (EEG alpha spindles) were analyzed to describe distracted drivers. Further, a classification approach was used to investigate whether alpha spindles can predict drivers' mental states. Results show that reaction times and alpha spindle rate increased with time-on-task. Moreover, brake reaction times and alpha spindle rate were significantly higher while driving with auditory secondary task opposed to driving only. In single-trial classification, a combination of spindle parameters yielded a median classification error of about 8% in discriminating the distracted from the alert driving. Reduced driving performance (i.e., prolonged brake reaction times) during increased cognitive load is assumed to be indicated by EEG alpha spindles, enabling the quantification of driver distraction in experiments on public roads without verbally assessing the drivers' mental states.

Alpha spindles as neurophysiological correlates indicating attentional shift in a simulated driving task.

The intention of this paper is to describe neurophysiological correlates of driver distraction with highly robust parameters in the EEG (i.e. alpha spindles). In a simulated driving task with two different secondary tasks (i.e. visuomotor, auditory), N=28 participants had to perform full stop brakes reacting to appearing stop signs and red traffic lights. Alpha spindle rate was significantly higher during an auditory secondary task and significantly lower during a visuomotor secondary task as compared to driving only. Alpha spindle duration was significantly shortened during a visuomotor secondary task. The results are consistent with the assumption that alpha spindles indicate active inhibition of visual information processing. Effects on the alpha spindles while performing secondary tasks on top of the driving task indicate attentional shift according to the task modality. As compared to alpha band power, both the measures of alpha spindle rate and alpha spindle duration were less vulnerable to artifacts and the effect sizes were larger, allowing for a more accurate description of the current driver state.