摘要: |
Distracted driving related to cell phone usage ranks among the top three causes of fatal crashes on the road. Although forty-eight U.S. states currently allow the use of personal devices if operated hands-free and secured in the vehicle, scientific studies have yet to quantify the safety improvement presumed to be introduced by voice-to-text interactions. This study investigates how different modes of drivers' interaction with a smart phone (i.e., manual texting vs. vocal input) affect drivers� distraction and performance in both conventional and semi-autonomous vehicles. The study was executed in a full-car integrated simulator with a sample size of 32 drivers. The study considered two scenarios: (1) conventional manual driving in a suburban environment with intersection stops; and (2) control takeover from an engaged autonomous vehicle that reverted to manual driving at a highway exit. The quality of execution of maneuvers as well as timing and tracking of eye-gaze focus areas were assessed in both scenarios. Results demonstrated that while participants perceived an increased level of safety while using the hands-free interface, response times and drift did not significantly differ from those manually texting. Furthermore, even though participants perceived a greater effort in accomplishing the text reply through the manual interface, none of the measured quantities for driving performance or eye-gaze focus revealed statistical difference between the two interfaces, ultimately calling into question the assumption of greater safety implicit in the laws allowing hands-free devices. |