The thesis presents five different studies with the overall purpose of developing better preconditions for preventing accidents among young novice drivers. The focus of the thesis is on the development of a model of the behaviour of young drivers in which different factors contributing to their accident involvement are explained. In the first study, new strategies in driver training have been developed and experimentally tested, the aim being to reduce the problems of overestimation, poor visual search patterns and unstructured teaching and training by laymen. In the second study, different safety measures were developed and experimentally tested with the purpose of finding the best method for reducing accidents among professional drivers. These two studies, combined with literature surveys, have contributed to the design of the model of young drivers' accident involvement. The suggested model is presented in paper III. The last two studies aim at deeper understanding of two specific relations in the model, the relation between training strategy and young drivers' subjective skill (IV) and the relation between lifestyle and accidents (V). The results of these studies, including the literature surveys of paper III, underline the complexity of the young driver problem. A structure with three main factors influencing accident involvement is suggested: the learning process, social influence and individual preconditions. Through different processes such as training and subjective skill, feedback and motives for driving, skill acquisition and automation, subjective norms or lifestyle, these factors influence driving behaviour and thereby accident involvement. In the two first studies, it was shown that improved driver training, in spite of poor results from many other evaluations of driver training, may contribute to the prevention of young drivers' accident involvement. It was also found that the strategy that is chosen for driver training is important for the driver's own estimation of his skill. The wrong type of training may produce over-confident drivers. Even if we have considerable knowledge about the factors contributing to accident involvement, we cannot yet with certainty identify the best countermeasures. We need to know more about these processes. Since there are still many unsolved problems and many processes in young driver behaviour that are not understood well enough, there is probably still a large potential for improving safety among young drivers. (A).
Samenvatting