Reflect

According to the World Health Organization, road traffic injuries are the leading cause of death among children and young people aged five to 29 globally. In Great Britain, pedestrians under 16 are twice as likely to be killed or seriously injured by road traffic as those in the 16-29 age group. This alarming statistic underscores the urgency of proactive road safety education, especially at the start of the academic year when routines are being re-established.

Parental warnings on busy mornings continually evolve over time. “Don’t talk to strangers” and, “Remember to look both ways”, have been commonly used by parents for many years. And this advice is currently supplemented by warnings about not looking down at those digital devices when crossing the road!

In this week’s Wednesday Wisdom, we asked a leading developmental psychologist, Professor Andy Tolmie of University College London to tell us what optimal messaging might look like when it comes to road safety usage. Given the complex nature of road systems in 2025, how should we talk to children at different ages and stages about risk? Is it as simple as warning them about road hazards, or is becoming a sensible road user a complex and life-long skill that requires practice and support over time?

As adults, we don't think twice about gauging when it's safe to cross, though even we can get caught out when we’re distracted or impatient. But children are still developing the brain skills and knowledge that underpin the controlled attention, planning and risk judgement needed to handle busy roads. In fact, road crossing is complex and challenging! Professor Tolmie's work shows that there are four key sets of related skills, built up in layers, that are needed to cross roads safely. Over to Andy...

Motivate

This is often about road ‘furniture’ and layout, such as zebra and light-controlled crossings, pavement build-outs and traffic islands being safer places to cross because the movement of vehicles is controlled or because the road crossing task is broken down into shorter sections. But it’s also about understanding that safe crossing depends crucially on being able to see oncoming vehicles, and on being seen – and expected – by the drivers of those vehicles. For this reason, crossing between parked cars (especially if you’re a small child) or on blind bends is inherently more dangerous.

Even when children know what locations are safer, it’s still crucial to pay attention to the things that matter and filter out things that are distracting. Attending to the right information is key. For instance, at a light controlled crossing, it’s necessary to press a button, monitor the lights that indicate when it’s safe to cross, watch the approaching vehicles to make sure they’ve seen the lights change and are slowing down, and finally cross through the safe section of the crossing. The attentional skills required are even greater at zebra crossings, where pedestrians and drivers need to carefully monitor each other’s behaviour, to ‘agree’ a time for the pedestrian to cross; and greater still again where there is no designated crossing site. It is critical to attend to vehicle movements and factors that might affect these, such as other traffic or rain, to the exclusion of everything else. When broken down like this, it's a lot to think about!

Children also need to learn how to time crossing with the movement of vehicles. To cross safely, they need to accurately judge how fast a vehicle is moving towards them, how much time that allows for crossing and whether it’s enough to get safely to the other side, given the road width and their walking speed. Where there is no designated crossing site and traffic is moving freely, and where the road is two-way and traffic is coming from both directions, making these judgements about relative movements is inherently more challenging. However, there are strategies we can teach our children that will help. Nudge them to look at the gaps between vehicles rather than the individual vehicles themselves (it’s not relevant what colour it is or who’s driving it!). Encourage them to make use of the fact that there is free space between the kerb and the passing vehicles’ position that can be used to start crossing carefully before a vehicle has actually passed. This is called a negative starting delay, which makes it possible to maximise the time available to cross.

Another key skill to develop is reading drivers’ intentions. This involves picking up on the cues to upcoming vehicle movements that are given by speed, road position, vehicle signals, road signals and the actions of other drivers. Understanding how to interpret these makes it possible to predict the vehicle’s likely arrival with greater certainty and use this to further increase the accuracy of timing judgements. For instance, if a vehicle is approaching but has an indicator flashing, it may be turning off before it reaches you, meaning it can be ignored (this should always be confirmed by watching whether it is also slowing down and changing direction slightly ahead of turning, since the indicator may just have been turned on by accident!). This skill comes closest to the ‘reading the road’ employed by drivers, though all the other skills feed into safe driving too, and indeed form part of a crucial ability developed by all skilled road users: building a working mental model or simulation of the traffic environment you’re in, how it operates and what you need to attend to in order to move through it safely and smoothly. In many ways, then, becoming a skilled pedestrian lays the foundation for becoming a competent and safe driver later on.

Children only develop these skills gradually as they progress through the primary school age range. Those aged five years and younger are typically poor at all four sets of skills, partly because deliberate control of attention (one aspect of executive function) is not yet well developed. However, it’s also because young children simply don’t know what to look out for. At best, they have a very limited model of the traffic environment that may not extend much beyond recognising kerb boundaries and different types of vehicle. Because their road crossing experience is restricted and characteristically managed by adults or older siblings, they have little opportunity or incentive to develop the necessary conceptual framework. Asked to map out a safe crossing route, five year olds will typically choose the most direct route between start and end points regardless of location or whether this means traversing the road diagonally and therefore spending longer potentially at risk. And asked at the roadside what they need to be looking at to cross safely, they’re as likely to talk about the cat in the next door garden as they are to mention cars.

They get notably better at identifying safe crossing locations and looking out for the right information by the age of eight. Both are assisted by better executive function, especially working memory, making it possible to take in, retain and co-ordinate more complex strands of information. Children in this age range also improve because of the greater experience gained from regular journeys to school, feeding into better conceptual knowledge of traffic environments. Past the age of nine, timing of crossings starts to improve, as executive control becomes more developed, and innate perceptual systems, which allow ‘time to contact’ of approaching objects, become more attuned to the patterns of movement exhibited by vehicles. Reading drivers’ intentions starts to get better by the age of 10, as a result of further accumulation of experience, including from being a passenger in the front seat of cars, and the development of a more consolidated model of traffic environments. On balance, then, 10- to 11-year-olds are starting to approach adult levels of competence, especially on safe crossing location and attending to the right information and timing of crossings, though they still have some way to go, as much as anything via sheer volume of practice.

Despite this, pedestrian accident rates peak later, at around age 14 for boys and age 16 for girls, which seems on the face of it surprising. In part, this is because the shift to secondary school (often on busier roads) typically means younger teenagers have to engage with more challenging traffic environments and their journey to school is longer for most. However, a bigger issue is that teenagers characteristically think concerns with road safety are for younger kids, overestimate their skill levels, and pay more attention to their social environment even when engaged in road crossing, leading to critical reductions in attention to traffic. Novice drivers, especially males, have a greatly heightened accident rate for the same reason. They have not yet sufficiently harnessed executive function to be able to strategically switch attention between the cognitive demands of the traffic environment and their social world, which is the hallmark of performance by older adults.

Let’s remember that this pattern of skill development is what happens in the absence of any specific support for growth. Amazingly, most children learn to become highly competent at road crossing by their mid-teens without any obvious training (the penalty is that a small percentage will be killed or seriously injured along the way). Certainly, guidance traditionally involved in school-based road safety instruction like, "Look both ways", is completely unhelpful because it’s too general and vague. Which ways and what for? And what do you do with the results of the looking? However, research has shown clearly that even with the constraints imposed by only partially developed executive function, it is possible to substantially improve children’s road crossing skills, giving them a crucial head start. The key to this is the use of various forms of practical training.

Support

So, what does practical training in road crossing actually involve? Early research emphasised taking children to real roadsides, pointing out the differences in their view of oncoming vehicles and their visibility to drivers at different locations; getting them to talk about what they can see that’s relevant to crossing and correcting them where necessary, so they can assess relative risk; and getting them to make ‘pretend’ crossings, where they view vehicles from the kerbside and either raise their hand or take a single step forward when they think it’s safe to cross, again receiving feedback on their performance.

Although this approach was shown to be effective, improving children’s skills by up to two years, it suffers from a number of problems, and also potentially misses a crucial dimension. First of all, it means actually taking children to the roadside, which for schools involves unfeasible amounts of staff time, since for safe supervision only a maximum of three children can be taken out at a time, and useful road environments may not be immediately adjacent. In many parts of the world, opportunities to take children out may also be restricted by poor weather. And even when roadside sessions are feasible, they tend to default to children working one at a time with the trainer, crucially limiting the kinds of dialogue that can take place and the conceptual consolidation that this produces.

Attempts to address some of these problems led researchers to explore the value of setting up pretend roads in school playgrounds or using tabletop models of road layouts. These work well for more static skills such as identifying safe crossing locations, but lack the dynamic movements necessary to build attentional and timing skills. However, I’ve been exploring other creative ways to address these issues and help children become more rounded safe pedestrians. One of our standout innovations was developing computer-based training tools that simulate real street-crossing scenarios. These let children practice and build confidence in key decisions – like where to stand, how to scan for traffic, when to cross and what cues to tune in to more generally – in a safe and controlled environment. The research showed that training using them produced substantial improvements in performance at the roadside itself.

Training using simulations also makes it possible for children to work safely with an adult in small groups of up to three, allowing them to talk between themselves as well as with the trainer. Dialogue with an adult has clear value if managed appropriately. For example, asking a child, "Why did you wait for that car?" or "What made you decide it was safe to cross?" helps them to make stronger mental connections, turning road safety into a thinking habit and is substantially more effective than simply telling them what to do. Guided reflection empowers children to understand the why, not just the what. However, children engage in richer and more extensive exploration of the rationale underlying their crossing-related decisions if they are allowed to discuss these with each other, and only receive occasional strategic input from the adult. Dialogue between children and adults tends to shut down more quickly, and for this reason, perhaps surprisingly, one to one adult-child training has been shown to be less effective than one to small group.

Where does this leave parents? The key message of the simulation research is that practical training in road crossing skills doesn’t have to take place solely at the roadside – or in schools! – provided it involves decision-making (not necessarily implementation) and dialogue about the rationale about underlying those decisions, preferably involving other children too. It is this that helps children build up the conceptual model of the traffic environment that is really the end goal.

As parents, we can use all our encounters with traffic environments as opportunities for practical training, and if this involves siblings or friends at the same time, so much the better. Perhaps to begin with, model thinking out loud. When crossing the street with your child, narrate what you're doing: "That car’s moving fast, so we’ll wait until it passes." Building on this, different types of real situations can then be used as learning moments. School runs can be turned into conversations about safety: "What do we check before stepping into the road?" As part of this, encourage practice and questions and give children agency to make decisions under supervision: "Do you think it’s safe to cross now? Why? Why not?" You might also look for interactive resources which support road safety through simulations, though using encounters during driving or even on television are likely to be as effective. Think about the four key areas of skill that I outlined above, and how these build up in layers, and use this to steer the focus of crossing-related discussion. If done regularly, these kinds of interactions will give them an all-important head start which can be carried into the teen years. Remember too, that although it may often not seem like it, parental influence on safe behaviour continues to be substantial through that period of growing independence. We can provide children with the right kind of supported practical experience when they are younger, continue to model safe behaviour by sticking to it when they get older, and sometimes allude to it when opportunity with teenage children arises (especially with regard to balancing social demands on attention with environmental ones).

A one-off chat clearly isn’t going to work. Teaching road safety is a process. And during back-to-school time, when children’s routines shift and independence often increases, that process becomes even more crucial. Thanks to this kind of research, we better understand that children need space to explore, reflect, and gradually build their skills. Whether you're a parent guiding your child down the street, or a teacher covering pedestrian safety in the classroom, remember: patience, conversation and practice go a long way.

And finally, a reminder for drivers as children return to school. As schools reopen, we all need to take extra care on the roads. Children walking or cycling to school are more vulnerable than adults and their smaller stature makes it harder for them to see and be seen. Unfortunately, many risky driving behaviours are seen far too often near schools. This includes parking on zig-zag or double yellow lines, reducing visibility for children crossing, parking on pavements; forcing families and children into the road; blocking pedestrian crossings after lights change, leaving children to navigate around cars; failing to give way to children crossing when turning into a road; driving while distracted, especially by phones and simply driving too fast. Mornings can feel rushed and stressful, but no journey is worth endangering a child’s safety!