An Examination of the Relationship Between Attention Profiles and Simulated Driving Performance

This study examined whether attention profiles from a computerized test battery relate to simulated driving performance. Five attention abilities were examined in the study: sustained, divided, selective, switching, and scanning. Participants completed eight tasks in a computer-based test battery and four driving scenarios designed to tap the same attention abilities. Physiological measures were collected during the test battery and the driving scenarios. Principal components analysis (PCA) with varimax rotation extracted seven components from the test battery, including the five proposed abilities along with speed and orienting components. Component scores were used as predictors of simulated driving performance in stepwise regressions and explained a significant proportion of variance (ranging from 7% - 26%) for most measures of driving performance. The speed, visual search, and divided attention components appeared as significant predictors more often than did the sustained, switching, orienting, and selective components. When physiological measures were added to the regressions, they explained additional variance beyond that explained by the component scores, but there was no consistent relation between simulated driving performance and any particular physiological measure.