Citation

Pradhan AK, Fisher DL, Pollatsek A. Transp. Res. Rec. 2006; 1969: 58-64.

Copyright

(Copyright © 2006, Transportation Research Board, National Research Council, National Academy of Sciences USA, Publisher SAGE Publishing)

DOI

unavailable

PMID

unavailable

Abstract

Younger novice drivers are overrepresented in fatal and nonfatal crashes. A major cause appears to be their failure to scan for information where potential threats may appear. Recent studies have assessed whether personal computer (PC) based training in risk perception helps novice drivers to identify high-risk locations while driving in a simulator. In particular, the Risk Awareness and Perception Training (RAPT) program was designed to teach novice drivers about categories of risky situations normally encountered in driving. RAPT, a PC-based interactive presentation with plan views of roadways demonstrating risky scenarios, provides information about inherent risks in each scenario and indicates areas where attention should go, to detect risks. A previous study showed that RAPT improved scanning behavior by novice drivers in a driving simulator immediately after training. The present study evaluated effects of such a training program an average of 4 days after training. One group of novice drivers was trained with RAPT and subsequently tested on a driving simulator; a matched control group was evaluated without previous training. The test contained 16 scenarios, each with information vital to safe navigation. An eye-tracker recorded drivers' point of gaze during drives. Trained drivers fixated significantly more often on predetermined regions where hidden risks could materialize; they were almost twice as likely to scan risky regions as were untrained drivers. Because the size of this training effect was the same as that when the test immediately followed training, it indicates that results of this training are reasonably long lasting. This training also raised performance of novice drivers close to the level of more experienced drivers, as tested in a previous related study.