Comparison of Primary and Secondary Task Measures as a Function of Simulated Vehicle Dynamics and Driving Conditions

Abstract

In a previously reported experiment involving a moving base driving simulator with computer-generated display, secondary task measures of workload showed significant increases as a function of large changes in vehicle dynamics and disturbance levels. Because the secondary task measures appeared less sensitive than desired, driving performance measures recorded during the same experiment were later analyzed. Particular emphasis in examining the driving performance data was placed on (1) determining the degree of intrusion of the secondary task on the driving task as a function of the independent variables, and (2) on comparing the sensitivity of the primary and secondary task measures. The results showed the secondary task does intrude significantly upon the driving task performance at low workload levels, but that it does not significantly intrude at high workload levels. Also, when the four primary task measures were analyzed for sensitivity to the independent variables, new information was obtained indicating greater sensitivity than is obtained with the single secondary task measure. Steering ratio, for example, is found to affect performance at high disturbance levels—a result not obtained in examining the secondary task by itself. The merits of primary and secondary task performance analysis are discussed, and suggestions are made for future work.