Overconfidence and optimism bias and traffic safety People show unrealistic optimism and overconfidence with regard to their own potential and performance. This affects individual risk-taking, and can prevent people from taking preventive measures. In the case of transport behaviour, research documents that people overestimate their driving skills and underestimate the risk of being involved in accidents. Moreover, there is no evidence that overconfidence decreases with experience. These biases have direct implications for driving education (which should focus more on risk avoidance), for the framing of safety advertising campaigns (which should put more emphasis on cautious driving) and for the cost-benefit analysis of projects that affect risk. • People overestimate their driving skills and underestimate the risk of being involved in accidents. • There is no evidence that overconfidence decreases with experience. • Driving education should focus more on risk avoidance. • Safety advertising campaign should put more emphasis on cautious driving. • Cost-benefit analyses of projects that affect risk should take into account these biases. There is ample evidence that people show unrealistic optimism and overconfidence with regard to their own potential and performance. This affects individual risk taking, particularly in relation to life and health risks. It can also prevent people from taking preventive measures (Thaler and Sunstein, 2008). In the case of transport behaviour, early research amongst college students showed that they perceived themselves as safer, more skilful drivers, and less likely to be involved in an accident than other drivers. This optimism was not found to be age- or sex- related, but did increase with driving experience (DeJoy, 1987). One obvious limitation of studies surveying college students is that the sample is limited to subjects who are younger, better educated and have less driving experience than the population as whole. However, similar findings have been obtained in other settings. Even professional drivers are subject to overconfidence. For instance, Dalziel and Job (1997) have followed taxi drivers over a 2- year period. The focus of this study was on the effect of fatigue on accident involvement. On all five examined dimensions of optimism bias, taxi drivers self-reported that they considered themselves more competent than their peers, with one caveat: this optimism was less pronounced on the dimension “driving safely when very tired”. According to Dalziel and Job, one possible explanation is that taxi “drivers are frequently reminded of the effects of fatigue on their driving abilities, as a result of the tiredness induced by the long hours they work, and that this leads to a somewhat more realistic assessment”. In a review of the literature, Sandroni and Squitani (2004) found no evidence that overconfidence decreases with learning and experience, and even that it increases with age with regards to the specific ability to drive after consuming alcohol! Sandroni and Squitani (2004) also argue that traditional driving education attaches too much importance on technical skills compared to making drivers aware of their limitations – increased technical skills may even lead to more risk-taking. In other words, current driving education does not sufficiently teach drivers how to avoid taking risks. Techniques to improve cautionary behaviour can include self-evaluation, having the trainee explain to the trainer which risks may arise and what counter-measures should be taken (“commentary driving”), group discussions and personal computer instruction. Harré et al. (2005) point out that “there is an important difference between people understanding that something is risky, and acknowledging that they are personally at risk”, and thus that safety advertising campaigns may miss their objective if the target audience thinks that these campaigns do not apply to them, but only to their (supposedly less competent) peers. They consider specifically self-enhancement bias – the tendency for individuals to take all the credit for their successes while giving little or no credit to other individuals or external factors (http://www.investopedia.com/terms/s/selfenhancement.asp). One particularly interesting finding that Harré et al. found in their studies of young drivers is “evidence of an inflated sense of superior driving ability (…) in the participants who watched advertisements showing dangerous driving resulting in a crash, compared with the control group, who watched advertisements showing people who had been drinking choosing not to drive. This raises the intriguing possibility that the experimental group participants viewed the protagonists as demonstrating poor driving ability, rather than as failing to take appropriately cautious action”. Harré et al. conclude that advertisements demonstrating cautious driving choices would be more effective than those illustrating dangerous driving resulting in a crash. Contrary to DeJoy (1987), they also found that men showed substantially more self-confidence than women about their driving ability, and concluded that future campaigns, instead of presenting “the ideal male driver as being able to handle a vehicle”, should present them as “responsible for the welfare of others”. Harré and Sibley (2007) studied the existence of implicit attitudes. They found that “participants more quickly associated words indicating themselves (e.g., I, me) with words indicating driving ability or driving caution than they associated words referring to other people (e.g., them, they) with these same positive driving attributes”, implying that “people’s beliefs about their driving superiority are probably deeply engrained”. The relevance of optimism bias therefore goes further than the issues of driving education and the framing of safety advertising campaigns, however. If people’s assessment of risks suffers from optimism bias, then the estimated benefits and costs of measures that affect the risk of certain events will be biased as well, and lead to misguided priorities in policy and infrastructure investment. It is an open question whether decision makers should prioritise projects with high ‘perceived’ benefits or with high ‘objective’ benefits. Dalziel, J.R. and R.F.S. Job (1997), Motor vehicle accident fatigue and optimism bias in taxi drivers, Accident Analysis and Prevention 29(4), 489-494. DeJoy, D.M. (1987), The Optimism Bias and Traffic Safety, Proceedings of the Human Factors and Ergonomics Society Annual Meeting, September 1987, 31: 756-759. Harré, N. and C.G. Sibley (2007), Explicit and implicit self-enhancement biases in drivers and their relationship to driving violations and crash-risk optimism, Accident Analysis and Prevention 39, 1155-1161. Harré, N., S. Foster and M. O’Neill (2005), Self enhancement, crash-risk optimism and the impact of safety advertisements on young drivers, British Journal of Social Psychology 96, 215–230. Jiang, L. Y. Li and X. Liu (2008), The effect of motor impulsiveness and optimism bias on risky driving behaviour in Chinese urban areas, Proceedings of the 11th International IEEE Conference on Intelligent Transportation Systems, Beijing, China, October 12-15, 2008, 605-609. Sandroni, A. and F. Squitani (2004), Survey on overconfidence insurance and self-assessment training programs, mimeo, Northwestern University and University College London. Thaler, R.H. and C.R. Sunstein (2008), Nudge, Improving Decisions About Health, Wealth, and Happiness, Yale University Press.