Evaluating driver cognitive distraction by eye tracking: From simulator to driving

Abstract Driver cognitive distraction, a critical factor for road safety, is challenging for researchers to evaluate, especially under real conditions. This paper introduces a novel method for simulating involuntary eye movement by combining the vestibulo-ocular reflex model and the optokinetic response. The difference between the predicted and observed eye movements is then assumed to be a measure of the level of cognitive distraction. When this hypothesis was validated under two sets of conditions, in a driving simulator and in a naturalistic situation, our algorithm was able to capture the cognitive distraction event in the naturalistic case. In addition, we also review and discuss the eye-movement sensor, which has a marked effect on the results of the evaluation, and the potential of using eye-movement sensors to evaluate cognitive distraction in drivers.