Young drivers' accident involvement may be explained by a number of different factors, one of which is that they tend to overestimate their skill in driving a car. This study is based upon the assumption that the degree of overestimation is related to the type of training the driver has received. In an experiment, two different strategies for training have been compared with regard to their influence on estimated and actual driving skill, as well as the drivers' degree of overestimation of their own skill. One of the strategies, used in the "skill" group, was to make the learner as skilled as possible in handling a braking and avoidance manoeuvre in a critical situation. The other strategy, used in the "insight" group, was to make the driver aware of the fact that his own skill in braking and avoidance in critical situations may be limited and unpredictable. The experiment was carried out at the Bromma driving practice area in Stockholm. Low friction has been simulated by using "Skid Car" equipment. Fifty-three learner drivers were randomly divided into two groups. Each of the groups was taught on the basis of one of the strategies. The training session was 30 minutes long. One week later, the drivers returned to take part in a test of their estimated and actual skill. The "skill" group estimated their skill higher than the "insight" group. No difference was found between the groups regarding their actual skill. The results confirm the main hypothesis that the skill training strategy produces more false overestimation than the insight training strategy. (Au
Abstract