Abstract
This study aimed to clarify the usefulness of surrogate assessment techniques of task demand induced from the use of in-vehicle driver interface. Three candidates of assessment techniques (occlusion, lane change test, visual detection test) were compared in terms of the sensitivity among task differences (visual-manual only tasks, voice control included tasks), and the representativeness of impacts on actual driving behaviors. Obtained results revealed the total shutter open time measure of occlusion technique is superior in both task difference sensitivity and representativeness of actual driving impacts, though the mean deviation measure of lane change test has not sufficient sensitivity and representativeness. Visual detection test used in this study is not sufficiently proper method, but is possibly improvable because the ratio of detection errors showed significant correlations with driving behaviors.
