What speed is appropriate for traffic in urban environments?

This is a hotly debated issue. We have recently tackled this issue by studying what human pedestrians are capable of judging, with a particular focus on children from 6 -11yrs.  When you stand at the side of the road and try and judge the distance and speed of the approaching traffic, many of the cues from road features can be shown to be unreliable.  One of the few reliable cues to distance is the visual size of the vehicle and one of the only cues to speed is the rate of expansion of the visual image, known as “looming”.  But all humans have a limit or “threshold” in their ability to detect looming and contrary to what you might expect if you are trying to cross the road, faster vehicles loom LESS (because a vehicle that is at 40mph vs. 20mph, needs to be twice as far away so its image size is smaller and the looming rate lower) so you may not detect fast vehicles as approaching, but see them as distant vehicles that are not approaching.

When we measure detection ability in adults and children we use very precise downward descent psychophysical procedures to get very accurate measures of thresholds.  But here are some demonstration examples for you to try. In each case we present brief stimuli of the type you would get with a brief glimpse (saccade) down the road, this would be the last sample you get before you make your step-out decision.


Click image for demo of stimulus

Here the images appear on your fovea with no additional motion and you should find this task quite easy.  But in this case you are not really detecting “looming”, so can just respond if you see any edge of the car move, 'catch trials' are included, where the car does not move forward, so we can monitor whether participants are guessing or not attending. Note that we are interested in basic motion detection abilities against the type of backgrounds you get in a road setting, so the car images are deliberately not made to travel down the road, but try the other conditions leading up to the “time gap” experiment and you should understand why.


Hopefully you could make the judgment in the first demonstration right up to 70mph, but this could be detecting any edge moving.  In natural scenes there is often additional lateral motion of the vehicle, or the observer moves a little at the roadside.  So now try the next 2 demonstrations and see if you can detect “looming” towards you when there is additional scene motion.  If looming is above threshold it should “pop-out” for you.  In all cases the additional lateral motion is less than 1 degree of visual angle, so the type of scene motion you might encounter in many natural settings.

Click image for demo of stimulus

Click image for demo of stimulus


Click image for demo of stimulus

The other problem that can occur is that you glance briefly down the road but may not fixate directly on the approaching vehicle.  To what extent does a looming vehicle “pop-out” when it’s just a few degrees outside of your central vision (fovea).

Click image for demo of stimulus

What we can then do is explore children's use of looming information is a real road scenario, where the vehicle does follow the road. We have measured children's walking speeds and find that children in primary school need approximately 5 seconds to safely cross the width of an average urban road. So 4 seconds would be "cutting it close". We can then test what time gaps children select in a simulation. Here we show you some examples of cars approaching that are 5 seconds and 4 seconds away. The problem of faster vehicles once again means that faster cars are visually smaller and are looming less so often look less immediate than slower cars.

Combined our findings across all these tasks strongly supports the argument for a wider use of 20mph zones around areas where there may be children attempting to cross the road.


The research was supported by UK ESRC award RES-062-23-0842



Links to Related Research

London Science Museum

Road crossing and developmental disorders

  Department of Psychology, Royal Holloway, University of London, Egham, Surrey TW20 0EX
Tel/Fax : +44 (0)1784 443526/434347