What can the task occlusion method reveal about time-sharing of a visual in-vehicle task while driving? To examine this issue, 24 licensed drivers (12 under 30 years, 12 over 65 years) were instructed to plan routes to specified destinations on an electronic map while driving a simulator. Four levels of steering difficulty (parked, straight road, moderate curves, and sharp curves) were examined for two levels of task immediacy (map fixed and map rotating). Subjects also performed the task while parked with the map intermittently occluded. Six combinations of viewing times (1, 1.5, and 2 s) and occlusion times (1 and 3 s) were examined. When driving, total task time increased significantly from 11.0 s while parked to 15.5 s on straight roads and 19.5 s on sharp curves. Total glance time at the display, however, remained unchanged (M=12.0, SD=0.6 s). As road curvature increased, subjects made shorter glances at the display, longer glances at the road, and more glances overall. As a result, driving performance degraded and error probability increased. During occlusion, total task time increased as viewing time decreased and as occlusion time increased. Total glance time, however, remained unchanged (M=10.4, SD=0.4 s) in all 6 timing combinations, and error probability remained constant. These results can be explained in light of four key constructs: time pressure on sharp curves resulted in short and efficient glances to the display; interference of concurrent driving resulted in longer total glance times at fixed maps while driving relative to task occlusion; and postponed processing and the cost of task partitioning balanced each other to keep total glance time unchanged when viewing time and occlusion time were manipulated. (Author/publisher)
Abstract