This report presents the results of a study of driver workload associated with use of in-vehicle systems while driving. The purpose of the study was to develop performance metrics and test procedures to assess the visual, manual, and cognitive aspects of driver workload. A second objective was to develop a “toolkit” of evaluation methods to enable developers to manage the driver workload implications of future products. Driver performance data was collected in three venues: in the laboratory (N = 57 drivers), on an interstate highway (N = 108 drivers), and on a test track (N = 69 drivers). Twenty-two in-vehicle tasks, plus the task of just driving, were examined under a variety of experimental conditions. Analysis of the data focused on selecting metrics that were repeatable, had predictive validity, and discriminated higher workload from lower workload tasks. The results indicated that task-induced workload on driving performance is multidimensional in nature. The effects observed depended on the characteristics of the task. Visual-manual tasks exhibited fundamentally different performance profiles than auditory-vocal tasks. The differences included both the dimensions affected and the magnitude of effects. Driving performance metrics identified for the workload evaluation toolkit included metrics from lateral and longitudinal control, object and event detection, and eyeglance behavior. Laboratory tools included an activity time model, a static task time measurement tool, a visual occlusion method, a peripheral detection task used with and without a part-task driving simulator, and a Sternberg memory task assessment tool. (Author/publisher)
Abstract