Parieto-occipital Alpha Power Indexes Distraction during Simulated Car Driving a Novel Mechanism for Evoked Responses in the Human Brain

The two major causes of fatal car accidents are distraction of visual attention and lapses in vigilance. Psychophysiological research on real-time classification of driver status aims at reliably identifying cognitive state changes that potentially lead to impaired driving capability. While various vigilance monitors have been suggested, tracking attention non-invasively and with high temporal resolution is still an open challenge. Traditionally, modulation of the parieto-occipital alpha band has been taken as an index of vigilance fluctuations. Recently, parieto-occipital alpha power was found to positively correlate with increasing task demand, which might reflect functional inhibition of irrelevant brain areas during cognitive processing. Here, we characterize EEG signatures indicating competition for attentional resources arising when a primary visual driving task is challenged by a secondary auditory task. Based on 128-channel EEG recordings different driving conditions were compared during four hours of simulated driving in 11 subjects: baseline (fixation cross; no visual flow, no driving), passive driving (the co-driver's perspective on visual flow), and active driving in a lane change test (LCT). The secondary task was kept constant across conditions: Subjects had to press buttons attached to the left or right thumbs reacting to high and low frequency auditory tones presented in an oddball paradigm every 5 to 6 s. Reaction times (RT) to auditory stimuli significantly increased from baseline over passive driving to LCT. EEG data showed a stepwise decrease of alpha power from baseline via passive to active driving. In-depth analysis revealed an interaction of alpha power, RT and driving task: increased alpha power preceded faster RT which was significant only during LCT. Increasing driving task demands are reflected in longer RTs to the secondary task. Classical alpha suppression by visual flow is enhanced by attentive driving. We suggest that for competing tasks, fast reactions to auditory stimuli can be performed only on the expense of a less efficient engagement of visual processing as revealed bya diminishedalpha suppression during the LCT. Even if stimulated by visual flow typical for a driving environment, the alpha rhythm is more than a simple indicator of visual stimulation or vigilance fluctuation: the level of its suppression reflects the relative distribution of attentional resources between concurrent visual and non-visual stimuli. We assume that the visual resources required for driving a car may significantly be narrowed when performing a secondary non-visual task which may be detected by EEG. Acknowledgment: Supported by the BMBF (16SV2243).