Use of a Fixed-Base Driving Simulator to Evaluate the Effects of Experience and PC-Based Risk Awareness Training on Drivers' Decisions
Fisher, Donald L., Laurie, Nancy E., Glaser, Robert, Connerney, Karen, Pollatsek, Alexander, Duffy, Susan A., Brock, John, Human Factors
Historically, 16-year-olds have been overinvolved in fatal crashes. Surprisingly, this overinvolvement has almost doubled in the last 20 years. For every 100 000 licensed 16-year-old drivers, there were 19 driver deaths in 1 975 and 35 in 1996 (Insurance institute for Highway Safety, 1998). This increase runs counter to the overall trend. In fact, the death rate overall has decreased from 15 driver deaths per 100 000 licensed drivers in 1975 to 12 driver deaths per 100 000 licensed drivers in 1996.
A number of approaches have been used in an attempt to reduce young drivers' crash rate. Primary among these has been driver education. The first known driver education program was established in 1916 (National Highway Traffic Safety Administration [NHTSA], 1994). However, not until 1976. in a suburb of Atlanta, Georgia, was a full-scale, controlled evaluation of driver education undertaken. The results were disappointing both in the longer and shorter term (NHTSA, 1994), and much of the money for driver education that had come from the federal government and the states disappeared.
As a result, few changes were made to existing driver education programs over the next quarter century.
However, the potential for change has increased severalfold over the last few years--in part because of increasing awareness of what causes young drivers to be overinvolved in crashes, the ability to use personal computers (PCs) to deliver training in exactly those areas that are now identified as critical, and access to high-fidelity, fixed-base driving simulators, which can be used to construct safe and more demanding tests of training programs.
Consider the recent developments in understanding why young drivers' crash rates are so much higher. A 1993 NHTSA study reported that young drivers' lack of experience, increased willingness to take risks, and greater immaturity are the primary reasons for the overinvolvement. More recent studies suggest that a lack of experience is responsible for up to 70% of the crashes among young drivers (Gregersen, 1996). For some time it was felt that the primary effect of young drivers' lack of experience was on their ability to handle the vehicle skillfully in difficult scenarios. However, it is now believed that the primary effect of young drivers' lack of experience may be on their ability to apply the higher-level cognitive skills that are needed for driving in very demanding scenarios (Mayhew & Simpson, 1995; Ranney, 1994).
One higher-level cognitive skill, risk perception, has been singled out repeatedly as especially important. Only 10 years ago it would have been all but impossible to deliver the training necessary to improve risk perception in a realistic and economical fashion. However, with the advanced video capabilities of modern PCs, risk awareness training can now include footage of actual risky scenarios over which a trainee has some control. A number of groups have taken advantage of these and other technological developments to create programs designed to train risk awareness skills (Blank & McCord, 1998; Lonero, Clinton, & Douglas, 1998; McKenna & Crick, 1997; Regan, Deery, & Triggs, 1999; Triggs, 1994; Triggs & Regan, 1998; Triggs & Stanway, 1995; Willis, 1998).
Only the program developed at the Monash University Accident Research Center has been evaluated using a driving simulator (Regan, Deery, & Triggs, 1998; Regan et al., 1999). In particular, an analysis was made of two forms of risk perception training: avoidance learning (Fuller, 1988) and mediated instruction (Wallace & Regan, 1998). The drivers receiving avoidance training were shown a risky scenario on a motion-based driving simulator that required them to take quick, remedial action in order to avoid a crash. The drivers receiving mediated instruction were given a 30-mm PC-based program that was designed to teach people to scan the road ahead effectively, to predict potential hazards, and to make decisions that were safe. …