BREAKING THE LAW: EDUCATION AND TRAINING FOR YOUNG DRIVER SAFETY Ray Fuller Department of Psychology, Trinity College Dublin Almost a law of nature The first fatal motor accident on record occurred in Ireland. It happened in 1869 to Mary Ward, a cousin of the third Earl of Rosse, as she was being driven around the estate of Birr Castle, in Co. Offaly in a steam-engined carriage (Siggins, 1994). One hundred and twenty-six years later, about nine people die on Irish public roads every week and at least 200 are injured. But not only this: sadly, road accidents are the largest single cause of death in the 18-24 years age group. About 50,000 new drivers take to the roads in Ireland annually. They are more of a risk to themselves and others then than they will ever be again. Over their first 5 years they have an injury-accident rate about five times that of more experienced drivers. In the UK, approximately 850,000 new drivers come onto the roads each year. About 170,000 of these drivers have at least one accident in their first year and approximately 11,000 of these accidents result in injury at an estimated cost of around £1.8 billion in 1991 (Department of Transport, Road Safety Division Research Tender Briefing, 1994). Similarly in the US the "just trained" driver is unquestionably the most dangerous. Those who have most recently passed the driving test kill the most other road users and are most likely to kill themselves (Evans, 1991). This seemingly universal phenomenon has been described by Evans as "almost a law of nature". One conclusion is inescapable from these observations: driver training, which for most people ceases after passing the driving test, and driver testing, are clearly not a guarantee of safe vehicle operation. Thus we are confronted with the problem: is it possible to break this apparent law of nature and release young adult drivers into the system with a greater confidence that they will be safe? Undoubtedly single measures alone to tackle this problem could have only limited success. Nevertheless in this paper I shall confine my discussion to two possible areas for intervention: young driver training and young driver education. Young drivers and their accidents To begin to tackle the problem of young drivers and their accidents, we can first ask if there are features which distinguish them from other groups of road users, in particular older drivers, which might explain their increased accident risk. The recent and fairly comprehensive FERSI report (FERSI, 1995) prepared by members of the Forum of European Road Safety Research Institutes and supported by the European Commission Transport Directorate, identifies the following features. Risk life-style. The first noticeable feature is that, within the group of young drivers, there is undoubtedly a subgroup of drivers who deliberately take risks and as a consequence are overinvolved in both traffic violations and accidents. Their risk behaviour is not just confined to roadway use but characterises their life-style as a whole. They are more likely to engage in drug use, heavy drinking and petty crime. Their poor socialisation and risk seeking may well be the result of both personality and environmental factors. However they are by no means the entire problem. Only about 15-20% of young male drivers fall into this category. There is no evidence that the majority of young driver accidents arises from intentional high risk or antisocial driver behaviour. Risk exposure. A second characteristic of young drivers is that they are more vulnerable because of the conditions under which they drive. Compared with others, they are more likely to drive vehicles which offer less protection in the event of a collision. They are more likely to drive with a car-full of friends, thereby increasing the number of casualties in the event of a crash. They are more likely to drive during darkness and particularly on weekend nights, periods typically associated with a higher incidence of driving with elevated BALs. Late at night there are of course the further risk-contributing effects of drowsiness and the opportunity provided by lower density traffic to drive faster, especially in urban environments. It is also worth noting that the alcohol-related accident risk of young drivers starts at much lower BALs (20 mg per 100 mL) and rises more steeply than in older drivers. Driving with the current legal BAL limit in the UK and Ireland, the involvement in a fatal accident of an 18-19 year old is estimated to be almost five times higher than for a 30-34 year old. Despite this, young drivers are not generally over-represented in alcohol-related accidents, except on weekend nights. Risk seeking. A third aspect of the young driver problem relates to attitudes to driving and in addition the driver's vulnerability to peer pressure. Evidence suggests that for some young adults, high-risk driving, which they see as a demonstration of superior control skills, is associated with social status and is reinforced by peer influences. It can also be unintentionally reinforced by the high-risk driving behaviour modelled by heroes in fictional drama on film and TV. Overestimation of competence. Fourthly, young male drivers typically overestimate their own competence and the possibilities for correction in a developing accident scenario.  To summarise then, young drivers: -include a sub-group of high-risk lifestyle individuals; -tend to drive under more vulnerable conditions; -are subject to peer pressures to adopt high-risk driving styles; -overestimate their ability to drive safely. This propensity to take risks on the road declines with greater maturity and major life events such as getting married. But increasing age is not the only saving factor, the other, and, as we shall see, perhaps the most important, is the accumulation of experience. Learning to be safe What does a driver need to know to drive safely? It boils down to learning what to do under each encountered condition in the road and traffic environment to obtain a safe outcome. This includes knowing how to create a safe road and traffic environment through one's own behaviour. In general, for each condition experienced, the driver needs to know what behavioural option will lead to a safe outcome and what will lead to an unsafe one: in more abstract terms, the relationships between Antecedents, Behaviours and Consequences - what we might call the ABC of driver safety. Unfortunately for the novice driver, a large variety of relatively unfamiliar situations will arise, such as the particular combination of the driver's current speed, a degraded road surface, a bend in the roadway, and the likelihood of loss of road adhesion. Why does it seem to be so hard for inexperienced drivers to know what to do under such circumstances? The learning of A-B-C relationships is difficult for a number of reasons: (1) the road situation the driver has to "read" can be complex, involving the integration of several dynamic variables such as relative and absolute speed, trajectories, properties of different vehicles, road adhesion and so on; (2) the A-B-C relationship can change gradually over time, for example over the long term through tyre wear. What once could be dealt with by braking now leads to a skid; (3) the experienced A-B-C relationship may not be reliable, perhaps because of the "forgiving" nature of the road and traffic environment. Unsafe practices, such as risky overtaking, may creep in because the actions of other drivers saves the day; (4) drivers have little or no experience of low probability hazardous scenarios from which they can learn; (5) learning through vicarious experience, such as seeing high speed crashes in fictional films and television, may be counterproductive, where the consequences portrayed are unrealistic, such as when the heroic driver extricates himself from tangled wreckage unscathed. Further problems for the young driver. Because it is so difficult for the novice driver to learn rules and procedures for safe action, s/he enters the fray with a limited body of knowledge to guide his or her behaviour, a limited set of rules to guide safe action. But not only this, what rules have been learned are more likely to have to be invoked and applied consciously - a process which takes up more of the mental capacity of the driver, leaving less available to deal with what may be a sudden flood of demands as a hazardous scenario develops. And yet s/he is mixing it with more experienced drivers with their sophisticated, relatively vast and largely automatised decision and control skills. These skills include access to mental models of road and traffic dynamics which arm the driver with a kind of in-the-head video. This can be run forward in time in the driver's mind and show the likely consequences of ongoing or alternative scenarios and actions, thereby informing behavioural decisions more appropriately. Given the complexity and difficulty of learning to "read the road" and adjust behaviour appropriately, it is perhaps hardly surprising that, in the process of learning to be safe, the typical young driver has to run the gauntlet of a long series of trial and error experiences. Although a lot of these will involve near-misses (or rather near-hits), unfortunately in a number of them the error may be expressed in damage or personal injury accidents. Those accidents typically involve driving too fast for the prevailing conditions. This is also a problem for older drivers and motorcyclists, but it is especially true for young novice drivers. This lack of an ability to relate speed to conditions (such as negotiating a bend) is also shown up in the relatively high involvement of young drivers in single vehicle accidents. To summarise this section then, young drivers: -have not yet developed comprehensive mental models of the dynamics of the road and traffic environment; -have to rely on trial-and-error experience to teach them about these dynamics; -as a consequence their accidents typically involve driving too fast for the prevailing conditions. Research has shown that after obtaining a license, the young driver's accident liability reduces by 30% in the first year due to the gaining of experience alone. This raises an important question for driver safety. Is it really necessary for novice drivers to have to learn to be safe the hard way, through direct experience of the expensive and life-threatening consequences of being unsafe on the roadway - or is it possible to educate, train and assess the safety of the driver before s/he is, as it were, thrown in at the deep end and left to sink or swim? Training for safety Let us look first at the possibilities for training. To be safe, in principle, we do not have to learn through direct trial and error experience how to act in every possible circumstance. We can be taught both general and specific rules and procedures for correct and safe behaviour. After basic vehicle control skills, such rule learning is a fundamental component, if not the fundamental component, of most training programmes: learning what the rules and procedures are for dealing with situations we are likely to meet. So are training programmes successful in reducing young driver accidents? Well, in the different European countries, varying requirements for driver training and different training methodologies have emerged. Variations relate to features such as (1) the age at which a driver may drive different types of vehicle (for a car, 16 in Norway, 17 in Ireland and the UK, 18 most other EU countries but 16 in Sweden and France as an accompanied driver under a contracted apprentissage or apprentice system) (2) the mandatory requirement for theory and practical training (most countries but not the UK, Ireland or the Netherlands) (3) the opportunity to practice with laypersons (not permitted in Germany, Denmark, the Netherlands, Portugal, Greece, Luxemburg) (4) the requirement for practical training and certification of driving instructors (all countries except Ireland) (5) the requirement of a probationary period for newly qualified drivers (Germany, Spain, Portugal, the Netherlands). What is perhaps surprising is that despite these wide variations in training experience, the FERSI study (1995) concluded that there was no evidence that the differences in national systems produced major differences at the level of national casualty totals. Now although this conclusion should be treated with caution, because a number of recent developments have not yet been fully evaluated, what it may underline is: (i) the importance of the experience of driving on the roadway as a key learning process. My guess is is that the fundamental problem is that appropriate safe behaviour simply cannot be completely taught - it has to be learned by doing and that inevitably means making mistakes on the roadway, and (ii) the effect of driver testing providing a common gateway through which only equivalently competent drivers pass, irrespective of the nature of their previous training. The future of driver training? So in the EU, young qualified drivers are more-or-less as good, or bad, as each other. But does this mean then that driver training can have no substantial part to play in helping to reduce the inordinately high level of young driver casualties? I think not. It has to be recognised that universally in Europe driver training courses have developed on a piece-meal basis and are not based on evaluative research. Neither have they been traditionally designed through the application of established principles of learning or skill acquisition. Thus, to take an example, in the UK, training is controlled by a Government Driving Standards Agency which maintains a register of Approved Driving Instructors who are periodically tested for ability and fitness to give instruction. A recent in depth study of this system (Groeger and Brown) found that less than 5% of training was given to more complex and hazardous manoeuvres such as overtaking and lane changing and less than 2% to elements such as hazard recognition and anticipation.   The nature of young driver accidents suggests that aims of training should emphasise: - developing better cognitive skills in hazard recognition and the assessment of risk - improving drivers' assessment of their own risk-taking and their own skills - developing attitudes of personal responsibility for road safety Improved content of training, then might include: -hazard recognition -perception of own risk-taking Part of our current research has been designed to find out the nature of problems encountered by 'just qualified' drivers. With the cooperation of the DoE and the driver testers around the country, 1000 drivers who have just passed their test have been asked to keep a diary of their next ten near-misses, anxious moments or other important safety learning experiences on the road (such as an accident) . With this kind of information, we hope to be in a position to design more effectively the training components for hazard recognition and sensitivity to risk-taking. Further components of improved training content might include: -the development of responsible attitudes to driving -elements dealing with emotional aspects (impatience; peer pressure; behaviour in competitive situations) and cognitive aspects (decision making; planning) - Germany is currently developing these aspects of training -an element of high speed driving (e.g. at legal maximum). Sooner or later a driver will be tempted to explore the limits of what is possible - and that sometimes means going over the limit of what is safe -the provision of enhanced safety feedback during driving experience - this has a corrective and a motivating function. Lack of feedback enables wrong practices and procedures to creep in. Note that in Ireland a driver age 19 can be on his or her own without having had any assessment. -elements tailored to the needs of the individual (skills, attitudes, personality). Clearly an enlarged training syllabus such as this has implications for the training and other supports for instructors, particularly those who have few teaching skills or who see their primary role as simply getting trainees through the test. What is implied is a new philosophy of training in which the training process explicitly takes responsibility for the prevention of accidents by conferring the required knowledge, skills and attitudes on the novice driver. The prospect may not be as daunting as it first appears. Expert systems using new dynamic multimedia technology and virtual-world simulation could provide essential tools for delivering the knowledge, attitudinal and skills components, including enabling the learner to experience a large range of hazardous situations which are only infrequently experienced in real life: they could provide the driver with experience by doing, but in a safe context. This would free up the instructor to provide feedback in on-the-road training components. This model of training has had demonstrable success in the field of pilot training in civil aviation - so why not on the road? Education Turning now from training to the process of education. Education as an intervention to enhance safety on the road comes in two main forms, media campaigns and road-user educational programmes. Media campaigns. First, media campaigns. Although it is difficult to evaluate the effects on safety of specific media campaigns, it has to be said that until very recently no systematic attempt has been made to evaluate their effectiveness, except in the limited sense of determining whether or not particular media messages were received by target groups. Analyses in terms of changes in attitudes, behaviour or accidents have been notable by their absence. This has certainly been the case in Ireland and, judging by the paucity of evaluations appearing in the scientific literature, does not seem to be atypical of Europe as a whole. Thus the Irish Government's decision to evaluate recent campaigns, and provide a framework for ongoing evaluation of future ones, is to be welcomed. What would also be welcome, however, is a more systematic application of principles for effecting behavioural change through media campaigns. We are gradually learning how to make such media campaigns more effective. In the field of driver behaviour, primary aims are to: - increase the driver's perception of the risks involved in particular situations - motivate the driver to take action to reduce those risks and finally to - instruct the driver in the behaviour which will enable risk reduction. The idea behind these first two aims is to instill in the road user a "mental simulation" which is triggered by a particular situation in real experience on the road and which, by running forward in time to punishing consequences, motivates safer behaviour. The point of the third aim, that of instructing the driver how to avoid risk, is to provide the example of safe behaviour so that drivers do not have to work it out for themselves and are more likely to follow it automatically. Realising these aims, then, requires: -situating the unsafe behaviour in a realistic context -providing a dramatic representation of the punishing consequences of the unsafe behaviour -providing a clear demonstration of the correct, safe behaviour in that same context. The relatively recent National Safety Council (Ireland) campaign concerning fast driving on rural roads (which is seen to lead to a car running into a child pedestrian and then overturning) is a good example of this approach. Regarding the actual content of such campaigns, we now have a fairly good idea about what factors will affect the persuasive effectiveness of a media message (see Wilde 1995). Messages should, for example, exploit the psychological processes of identification and modelling. This means that the source of the message should be credible, knowledgeable, trustworthy and similar to the receiver of the message (e.g. in terms of features such as age, sex, social class and linguistic style). Message content should in part express views already held by the recipient and advocate changes that are within his or her "latitude of acceptability". Finally the creation of a motivational state in the recipient, appealing to what s/he wants to achieve or avoid, should not involve an overly threatening fear appeal. The result may be one of "defensive avoidance" in which the recipient effectively suppresses the information because it is too disturbing. These then are the kinds of principles for effecting behavioural change through media presentations that might be more reliably used in our safety campaigns. Road-user educational programmes. With regard to road-user educational programmes, learning through educational programmes starts with the example, guidance and control of parents with small children, progresses through the support of school-based instruction and facilities and develops further through accumulating direct experience of road and traffic conditions, whether as pedestrian, vehicle occupant or cyclist. Providing information and motivational messages on an ongoing basis, that indicate and support specific safe behaviours, can also be effective. What we might seriously consider, however, is the strengthening of this concept of safety education, through the design and implementation of a national curriculum for safe roadway use, starting with primary school children and continuing right through to early adulthood. Such a curriculum would target not only developing children but their parents and role-models as well. Its fundamental aims would be to develop an attitude of personal responsibility for one's own safety and that of others, a sense of the dangers of unthinking, uncaring and uncontrolled behaviour on the road and a body of knowledge and skills, tailored to developing needs, to engage with the road and traffic system safely and effectively. The future challenge One of the tragedies of modern roadway use is the seeming inevitability of a sizeable annual toll of injuries and deaths, especially amongst young drivers. Ireland is currently in a particularly precarious position in this respect. Compared with many other European countries, Ireland have a burgeoning cohort of youngsters keen to get their first wheels, a population bulge which will continue into the early part of the next century; secondly Ireland is likely to witness a rapid growth in the rate of motorisation. In 1994 the country had only just over 60% the number of vehicles per 1000 population as France and Italy. And thirdly, Ireland has relatively high young car-driver insurance rates, which may force many away from cars to motorcycles, a form of transport associated with a significantly greater risk of injury and fatal accident. For all of these reasons, the frequency of young driver accidents in Ireland is likely to increase. The question is therefore, will Ireland and countries in a simil ar position exploit the potential of training and education to reduce this carnage or, as a community, are we just going to sit back and watch it happen?   References (Evans, 1991) (FERSI, 1995) (Groeger and Brown) (Siggins, 1994) (see Wilde 1995)   Back Home