Testing assumptions implicit in the use of the 15-second rule as an early predictor of whether an in-vehicle device produces unacceptable levels of distraction

Given the proliferation of in-vehicle technologies, techniques must be developed to ensure devices do not produce unacceptable levels of distraction. One approach is to use static time on task (e.g., the 15-second rule). However, this practice makes three critical assumptions: (1) static time on task predicts time on task while driving; (2) time on task measured in a hazard-free environment predicts time on task when drivers expect hazards; (3) time on task predicts perceived distraction, collisions, and driving errors. To test these assumptions, two tasks were compared in 32 drivers using a driving simulator. The tasks were manipulating controls of a radio/tape deck and dialling a hand-held cellular phone. Static time on task underestimated dynamic time on task, though the differences between tasks were roughly consistent across testing conditions, with the cellular task taking more time. Participants who expected hazards required slightly more time on task than those who did not, but the effect was only marginal (p = 0.09) and consistent across tasks. Finally, the device with higher static time on task also produced significantly more lane deviations and perceived interference, though the predicted pattern of results did not emerge for collisions and hazard response time.