Support road.cc

Like this site? Help us to make it better.

news

Invisible cyclists: Eye-tracking experiment finds drivers don't see more than 1 in 5 riders (+ video)

Younger drivers, women and sat-nav users most likely not to see cyclists

More than one in five cyclists go unseen by motorists on the road, according to an experiment using eye tracking technology conducted for the insurance company Direct Line, confirming the extent of the ‘SMIDSY’ – ‘Sorry mate, I didn’t see you. Younger drivers missed spotting nearly one in three riders, and female motorists one in four. By contrast, just 4 per cent of what Direct Line terms "jaywalking" pedestrians were not seen, and 15 per cent of motorcyclists.

Motorists who took part in the experiment wore “specialist glasses that pinpoint the exact focus of the eye by tracking microscopic movements in the cornea,” said the company, adding that film footage “enabled researchers to establish exactly where drivers focus their vision, which was often at clouds, buildings and passers-by.” Here’s a short video of it in operation.

The experiment was conducted in three cities – London, Oxford, and Sheffield – and according to Direct Line the issue is most prevalent in the capital, where motorists fail to see three in ten cyclists.

That’s despite the growth in cycling in the city in recent years, that suggests we’re some way from seeing a ‘Safety in numbers’ effect kick in there, whereby the more people there are on bikes, the more motorists are likely to register their presence and drive accordingly.

In Oxford, which has the second highest levels of cycling in England after Cambridge, 20 per cent of riders went unseen, and in Sheffield, 15 per cent.

Researchers found examples of motorists taking their eyes of the road to adjust sat-nav devices and in one case navigate using a hand-held smartphone, and Direct Line says that 24 per cent of riders are “invisible” to drivers using a sat-nav device, compared to 19 per cent where the motorist does not use one.

The biggest difference in the proportion of drivers registering the presence of cyclists was by age.  Some 21 per cent of cyclists were unnoticed by those aged 50 or over, but 31 per cent among motorists aged between 20 and 29 years. Again, that’s a cause for concern given that younger people have better eyesight on the whole.

Vicky Bristow, spokesperson for Direct Line car insurance said “For the first time we know exactly where people focus their eyes when driving and the results are frightening.

“UK roads are busy and congested and as a result millions of cyclists are going unseen.

“Blaming motorists seems like an easy option, but this issue can only be really addressed if both motorists and cyclists accept responsibility.

“Encouraging all road users to be extra vigilant will certainly improve road safety but tackling an issue of this scale really requires top-down change.

“Successive governments have encouraged local authorities to adopt policies to make cycling safer in the past but our research highlights that this issue is still widespread.”

Drivers                 % that failed to spot cyclists

Sat nav drivers                     23.7
Non-sat nav drivers                 19.0
Female drivers                      25.6
Male drivers                        17.1
Drivers aged 20-29                  31.1
Drivers aged 30-39                  20.7
Drivers aged 40-49                  21.6
Drivers aged 50-59                  20.9
All drivers                         22.0 

Source: Direct Line Motor Insurance

One thing we wondered was whether the cyclist wearing hi-viz clothing had any impact on their visibility to motorists – a subject of some debate in comments to stories here on road.cc - so we asked Direct Line whether the clothing cyclists sported had any impact.

The company told us that the study considered a lot of data, including speed cameras, pedestrians, road signs etc, and the lack of vigilance motorists display towards cyclists was what it chose to focus on.

It added that had the survey been commissioned specifically into cyclists, then that would have been one of the areas it would have looked at, and that it is likely to undertake such research in the future.

As for that comment in the first paragraph about "jaywalking" pedestrians, the term of course is widely used in the United States where there are much more severe restrictions on where pedestrians can legally cross a road compared to England, Wales and Scotland; here, pedestrians are not allowed on motorways, but other than that can cross the road except where a specific 'no pedestrians' sign is in place, although official advice is for them to wait until it is safe to do so. Jaywalking is an offence in Northern Ireland, but one that is rarely enforced.

 

Simon joined road.cc as news editor in 2009 and is now the site’s community editor, acting as a link between the team producing the content and our readers. A law and languages graduate, published translator and former retail analyst, he has reported on issues as diverse as cycling-related court cases, anti-doping investigations, the latest developments in the bike industry and the sport’s biggest races. Now back in London full-time after 15 years living in Oxford and Cambridge, he loves cycling along the Thames but misses having his former riding buddy, Elodie the miniature schnauzer, in the basket in front of him.

Add new comment

45 comments

Avatar
mintimperial | 11 years ago
0 likes

Blaming motorists seems like an easy option, but this issue can only be really addressed if both motorists and cyclists accept responsibility.

I give up, really I do. Jesus H Christ.  22

Avatar
spatuluk | 11 years ago
0 likes

"Blaming motorists seems like an easy option, but this issue can only be really addressed if both motorists and cyclists accept responsibility."

Try telling that to a judge! oh.. wait..

Avatar
felixcat | 11 years ago
0 likes

It would be interesting to do the experiment with cyclists. I would guess that cyclists are much less likely to fail to see motor vehicles. If so, why?

Avatar
rggfddne replied to felixcat | 11 years ago
0 likes
felixcat wrote:

It would be interesting to do the experiment with cyclists. I would guess that cyclists are much less likely to fail to see motor vehicles. If so, why?

'cos they're bigger, and so less likely to fall into a 'saccade': http://www.londoncyclist.co.uk/raf-pilot-teach-cyclists/

Avatar
Simon E replied to rggfddne | 11 years ago
0 likes
nuclear coffee wrote:
felixcat wrote:

It would be interesting to do the experiment with cyclists. I would guess that cyclists are much less likely to fail to see motor vehicles. If so, why?

'cos they're bigger, and so less likely to fall into a 'saccade': http://www.londoncyclist.co.uk/raf-pilot-teach-cyclists/

The cyclist is usually travelling more slowly, expecting to give way or stop and, being more vulnerable, watching for danger.

Slowing cars allows drivers more time to look, which means they are more likely to see cyclists and peds. Forcing drivers to take longer at junctions, as described in Chris Boardman's article in The Times (and illustrative Street View link in a tweet), means they have to look more than if they just glide through. Everyone does it - it's easier if you can keep rolling and not stop completely. But it is more difficult to scan the scene effectively while the car is moving than if you stop completely.

Riding further into the road means you are more obvious to other road users, including those on side roads. It also reduces (but doesn't prevent) the chance of too-close passes by impatient, selfish idiots. I usually ride at least 18" and sometimes 2 feet from the edge of the road.

Wobbling is apparently a good way to get the attention of following traffic - your trajectory becomes unpredictable.

I have been saying it for a long time now, but there should be a restriction on in-car gizmos, touch-screens and most definitely satnav, but sadly manufacturers think these are great selling points and buyers are seduced by gadgets. Many drivers also have a false sense of just how good (i.e. safe) their driving is.

Avatar
ConcordeCX replied to rggfddne | 8 years ago
0 likes

nuclear coffee wrote:
felixcat wrote:

It would be interesting to do the experiment with cyclists. I would guess that cyclists are much less likely to fail to see motor vehicles. If so, why?

'cos they're bigger, and so less likely to fall into a 'saccade': http://www.londoncyclist.co.uk/raf-pilot-teach-cyclists/

i would guess that it may also be because cars pretty much drive themselves (the clue is on the word 'automobile') so do not require as much attention as riding a bike, which isn't going anywhere without some effort from the rider, and because in general cyclists are looking out for themselves more than drivers because of the greater damage likely in an unpremeditated shared space situation.

Avatar
CarlosFerreiro replied to felixcat | 11 years ago
0 likes
felixcat wrote:

It would be interesting to do the experiment with cyclists. I would guess that cyclists are much less likely to fail to see motor vehicles. If so, why?

I think the most interesting test would be "are drivers who cycle a lot noticeably better at picking up cyclists visually"?
i.e. can noticing/not noticing be improved with "training". If so, then obviously that training is required for all drivers.

Avatar
bohrhead | 11 years ago
0 likes

Surely you don't need to look directly at a cyclist or even completely focus on them to know that they are there. I focus on things other than the road when I'm driving (e.g. spring lambs) but my field of vision is quite enough to passively take in the road and what's on it too.

Avatar
HKCambridge replied to bohrhead | 11 years ago
0 likes
bohrhead wrote:

Surely you don't need to look directly at a cyclist or even completely focus on them to know that they are there. I focus on things other than the road when I'm driving (e.g. spring lambs) but my field of vision is quite enough to passively take in the road and what's on it too.

But human peripheral vision is not good, and we have a tendency to make up things which we expect to see (see pilot link above). You cannot, and should not rely on peripheral vision while driving. It only take a moment to move your eyes or head.

In any case, it doesn't explain the discrepancy with pedestrians.

Avatar
CraigS | 11 years ago
0 likes

The fact it's based on eye tracking shows it's not about cyclists not making themselves visible, it's about drivers not looking.

Riding defensively puts you in a driver's eyeline, no amount of hi-vis will get you seen if you're riding along the kerb.

Avatar
alotronic | 11 years ago
0 likes

No surprises but good to see it studied. I have done exactly the same kind of testing on websites.

http://en.wikipedia.org/wiki/Saccade

Short version - your brain makes stuff up as your vision skips around. And mostly makes up a picture between states of vision and *may* completely skip the cyclist/pedestrian/motorbike in the middle 'frame'.

Once you know that everything kinda makes sense - they really didn't see you! However good and careful driving can compensate for this obviously.

Avatar
md6 | 11 years ago
0 likes

quote: Blaming motorists seems like an easy option, but this issue can only be really addressed if both motorists and cyclists accept responsibility.

How exactly can a cyclist be responsible for a driver watching a cloud, looking at someone waling by instead of the road? If i'm wearing lights and/or bright clothing, or it is day time. What else can I be expected to do? seriously, I am interested...any ideas other than drive a car?

Avatar
mad_scot_rider | 11 years ago
0 likes

Sorry still trying to align these two phrases in my head

"... Blaming motorists seems like an easy option ..."

and

"... drivers focus their vision, which was often at clouds, buildings and passers-by ..."

It's just not working

Avatar
notfastenough replied to mad_scot_rider | 11 years ago
0 likes
mad_scot_rider wrote:

Sorry still trying to align these two phrases in my head

"... Blaming motorists seems like an easy option ..."

and

"... drivers focus their vision, which was often at clouds, buildings and passers-by ..."

It's just not working

That's what I was thinking. What are we supposed to do? Other than disguising myself as a cloud, obviously.

Avatar
YorkshireMike | 11 years ago
0 likes

Funny, I've often thought about how/when an experiment like this would be conducted while I've been out riding - the results are pretty much identical to what I expected!

Pages

Latest Comments