The safer the car, the faster it will be driven

Taken from a talk at the University of Reading delivered by Mark Horswill, a lecturer in the university's department of psychology

It's probable that today in the UK about 10 people will die in car crashes, about 110 people will be seriously injured, and about 780 people will be slightly injured. So, what can we do to reduce car crashes further? One way is to try and understand driver behaviour and the factors that influence it. If our aim is to reduce car crashes, one of our best options would be to understand road-user behaviour and work out ways of influencing it.

It's probable that today in the UK about 10 people will die in car crashes, about 110 people will be seriously injured, and about 780 people will be slightly injured. So, what can we do to reduce car crashes further? One way is to try and understand driver behaviour and the factors that influence it. If our aim is to reduce car crashes, one of our best options would be to understand road-user behaviour and work out ways of influencing it.

Research has shown that most drivers think they're more skilful, safer, slower, and less likely to have an accident than the average driver. This is highlighted by the finding that people are willing to tolerate significantly greater speeds when they're driving than when someone else is driving. This, of course, fits in with the idea that people think other drivers aren't as competent or accident-free as themselves.

These illusory biases leave us with a problem for road safety. If people believe they are safer than average, why should they protect themselves by driving slower? In fact, we've found that drivers with greater biases do drive faster than other drivers.

This is a problem for safety campaigns directed at motorists. Drivers may see safety campaigns as being directed at the "average person" who is less safe than themselves - and therefore ignore the advice in the road-safety campaign, even though they might still agree with the message. One group of people who do not show these illusory biases are those who have been involved in a severe crash for which they are unambiguously to blame. But are there ways of reducing illusory biases without the crash?

One of our research projects at Reading has involved getting people to imagine and describe a severe accident for which they are to blame. What we found was that just getting people to imagine and describe this accident has an effect surprisingly similar to being involved in a real accident. They no longer believed they were more skilful, safer, and less accident-involved than the average driver. Their preferred driving speeds also decreased.

Do fast cars encourage people to drive faster, or do fast drivers simply choose to buy faster cars? We got UK drivers to complete an internet questionnaire. As part of the questionnaire, we gave people a description of a car and asked them how they would intend to drive it. What these drivers didn't know is that there were two versions of this questionnaire. Half the drivers were filling in a questionnaire where the car described was low-powered - a Skoda Felicia. The other half were filling in a questionnaire where the car described was high-powered - a BMW. The differences in intended speed were striking. For example, on a motorway, drivers intended to drive the BMW nearly 10mph faster than the Skoda Felicia.

The only problem with looking at vehicle performance is that a high-performance car is also likely to be bigger and smoother, and have more safety features, and so on. So this is what we looked at in a third study, where we studied different vehicle characteristics independently of one another. Vehicle smoothness and handling did not affect risk-taking, but performance and number of safety features did.

What can we do about this? A first step would be to redesign our cars. The design philosophy behind modern cars is to make them quiet and comfortable, and as well-insulated as possible from the outside world. The problem with this is that you're removing the very cues drivers use to estimate speed and danger in the first place. The problem, of course, is that the danger is still there. We should make cars that give more perceptual feedback when you're going fast. We should similarly educate drivers that sophisticated vehicle safety features do not automatically lead to invulnerability as indicated in an increasing number of car adverts.