Predicting Relative Crash Risk from the Attentional Effects of the Cognitive Demand of Visual-Manual Secondary Tasks 2017-01-1384
This proof-of-concept demonstrates a new method to predict the relative crash risk in naturalistic driving that is caused (or prevented) by the effects on attention of visual-manual secondary tasks performed while driving in a track experiment. The method required five steps. (1) Estimate valid relative crash/near-crash risks of visual-manual secondary tasks measured during naturalistic driving. These data were taken from a prior SAE publication of unbiased estimates of the relative crash/near-crash risks of secondary tasks in the 100-Car naturalistic driving study. (2) Calculate the “physical demand” and “cognitive demand” scores for visual-manual secondary tasks performed while driving on a track. These data were taken from another prior SAE publication that analyzed the physical and cognitive demand dimensions underlying the driver performance of secondary tasks in the Crash Avoidance Metrics Partnership Driver Workload Metrics project. (3) Find similar tasks between naturalistic driving in Step 1 and track driving in Step 2. (4) For these similar tasks, find the linear regression model that best predicts the relative crash/near-crash risk in Step 1 from the physical and cognitive demand task scores in Step 2. (5) Use the model in Step 4 to predict the relative crash/near-crash risk of the track visual-manual tasks that did not have corresponding naturalistic tasks. The results showed that relative crash/near-crash risk had a near-perfect inverse correlation with the attentional effects of cognitive demand of visual-manual tasks, and no statistically-significant correlation with the physical demand of those tasks. Both results are contrary to current assumptions in the driving safety field. This proof-of-concept, which requires validation with additional experimental and naturalistic data, is the first known study in the driving safety literature to establish the external validity of an experimental study in predicting the relative crash/near-crash risk of secondary tasks during naturalistic driving.