The objective of the experiment reported in this paper was to study the role of (a) available risk information, (b) degree of control, and (c) amount of experience in a traffic task. The specific target behaviour and traffic task selected for the experiment is people's risk-taking tendency in blind curves on two-lane rural roads in the Netherlands. The reasons for selecting this particular task were: (a) when approaching a blind curve, the level of risk is largely determined by the driving speed, which can be measured objectively; (b) by selecting a quiet country road as the experimental course it is possible to minimize interference by other traffic and thus create similar task itions for all subjects; (c) a significant proportion of rural road accidents occurs at or near (blind) curves. 64 subjects were recruited; all were under 40 years of age and belonged to one of two experience groups: 'inexperienced' drivers who had obtained their driving licence less than 1 year ago and had driven less than 10000 km; 'experienced' drivers who possessed a driving licence for more than 5 years and had driven more than 50000 km. 39 subjects were male, 25 were female, and the mean age was 25.1 years. The subjects were provided with varying risk information about dangers of blind bends. Each subject drove from De Punt to Norg and back (24 km) and encountered 20 blind curves. All drives took place on off-peak hours, when the traffic intensity on this road is low. The legal speed limit on this road is 80 km/h. Subjects drove an Opel Rekord of the University of Groningen, Traffic Research Centre (TRC). The car was equipped to continuously register the driving speed and to synchronously videotape the road ahead. The experimental results demonstrate that people's risk-taking tendency can be influenced by presenting them with relevant risk information: both frequency information, which consisted of relevant accidents statistics and a probability estimation based on these statistics, and scenario information, which described how blind curve accidents may occur, resulted in substantial and significant speed reductions. This implies that drivers are sensitive to both types of risk information.
Abstract