This study contributes towards the development of an adaptive automation system by focusing on the identification of diagnostic measures for visual distraction. Since previous research has shown a strong association between eye glance measures and driving performance, this study seeks to identify eye glance measures that are diagnostic of visual distraction. In a driving simulator, 14 participants responded to a lead vehicle braking at a^'2 or a^'2.7 m/s2 periodically while reading a varying number of words (6-15 words every 13 s) on peripheral displays (with diagonal eccentricities of 24A., 43A., and 75A.). Results showed that as the number of words and display eccentricity increased, total glance duration and reaction time increased and driving performance suffered. Correlation coefficients between several glance measures and reaction time or performance variables were reliably high, indicating that these glance measures are diagnostic of visual distraction. Reaction time is increased by 0.39 s and standard deviation of lane position is increased by 0.06 m for every 25% increase in total glance duration. These findings can be applied to the development of driver information and warning systems.
Abstract