The Hurt Report

[thundercatsho]

Thirty years ago, a man called Harry Hurt (really!) published a ground-breaking study into motorcycle accidents.

Harry and his expert team spent 1979 and 1980 racing to the scene of motorcycle accidents in southern California to try to work out why and how the riders had crashed. When they got there, they talked to witnesses, the attending police, any involved drivers and where possible the riders themselves. They followed up over 900 accidents on site and reviewed a further 3,600 accident reports.

The findings were presented in 1981 as “Motorcycle Accident Cause Factors and Identification of Counter-measures”, better known today as the “Hurt Study” or “Hurt Report”.

It was the first half-way scientific study of motorcycle accidents.

The full report is over 300 pages long, and Hurt and his team drew no less than 55 main conclusions. I’m not going to repeat them here (a search on Google for “Hurt Report conclusions” will find them for you) but I will summarise one key area:

Approximately three-quarters of these motorcycle accidents involved collisions with another vehicle, which was most often a passenger automobile. Intersections are the most likely place for the motorcycle accident. In multiple-vehicle accidents, the driver of the other vehicle violated the motorcycle’s right-of-way and caused the accident in two-thirds of those accidents. The failure of motorists to detect and recognise motorcycles in traffic is the predominating cause of [multiple vehicle] motorcycle accidents. The driver of the other vehicle involved in collisions with the motorcycle did not see the motorcycle before the collision, or did not see the motorcycle until too late to avoid the collision.

Before the ’80s were over, a UK-based study conducted by Booth looked at nearly 10,000 motorcycle accidents in the predominantly urban Metropolitan Police area.

It concluded that nearly two-thirds (62%) were primarily caused by the other road user. Half of the accidents were caused by car drivers, and 10% by pedestrians. The report found that two-thirds of motorcycle accidents where the car driver was at fault were due to the driver failing to anticipate the action of the motorcyclist.

Notice the similarities?

As the authorities and rider groups themselves picked up on the problem that motorcycles simply weren’t being seen in traffic and that many accidents happened when drivers pulled out in front of riders, one obvious suggestion was that motorcyclists needed to make themselves more visible

The result was that the use of day riding lights (DRLs) and other conspicuity aids such as high visibility clothing (hi-vis) were suggested in books like the Highway Code and in some cases became mandatory (permanently-illuminated turn signals in California, headlights-on in France and Australia, CBT instructors and trainees required to wear day-glo bibs in the UK – just a few examples).

Yet there were warnings that this might not be a fully effective strategy. From “Motorcycle Safety – a scoping study” published by the Transport Research Laboratory in 2003:

“The conspicuity problem appears to be partly associated with car drivers learning visual strategies that are not very effective at detecting motorcycles, and there is potential to address this by training – although it is not clear how effective this would be in the long term given that the basic problem may be the relative rarity of motorcycles.”

So what are these visual strategies they are talking about? Somewhat simplified, it’s down to three things:

1.The way the eye works (a very narrow zone of clear focus right in the very centre and a large “fuzzy” area where what seems like a clear image is actually filled in by the brain from memory, rather like the way software interpolation on a digital camera “invents” detail from the surrounding pixels)
2.The way the brain interprets what the eye sees (it detects what it sees using movement or light/dark contrast, but then identifies what it sees from a “database” of shapes it knows from prior experience are important – thus drivers tend to be aware of cars and trucks, but not bikes and cycles)
3.Where and how drivers look (research shows they look for gaps, not vehicles, because that’s what they need to pull out into, and they look for an average of just half a second, which isn’t long enough to scan and focus on the full area between that gap and close up to their car).
So how does that lead to “detection failure”?

The first thing to understand is that your brightly-coloured and illuminated bike isn’t moving relative to the background until you are very close to the driver – this is called the “looming effect”. So you’re not picked up by the “motion detection” system, particularly when drivers are glancing rapidly to the left and right.

The second thing to understand is that your brightly coloured and illuminated bike only stands out if there is a strong contrast against the background (incidentally, hi-vis vests are too small to be effective, particularly behind a fairing – so use a sleeved hi-vis jacket). An increasing number of vehicles now also use DRLs, so the “contrast detection” system fails too.

In both these cases, even if the driver appears looks straight at you because you’re between them and the gap they intend to use, they tend to look behind you so you’re out of focus. Combined with the failure to detect you via motion or contrast, it means from their perspective the road ahead of the gap appears to be clear (even though there is a bike in it) so they pull out.

This kind of behaviour even has a name – it’s called “looked but did not see”.

And that’s the point I want to make about the effectiveness of DRLs and hi-vis.

Thousands of riders across the world are still injured or killed in this self-same accident that was identified 30 years ago.

If hi-vis and DRLs really worked, we’d find a big shift in the location of accidents since the 1980s when those first serious studies in accidents were done (when riders didn’t use hi-vis and DRLs). But we don’t.

We’re still having virtually the same proportion of junction accidents 30 years on. There’s been a slight drop in the frequency of accidents per vehicle mile, but the same accidents are happening in the same places in more or less the same ratios as ever.

So, the main benefit of hi-vis and DRLs is to help the driver to see you when you’re five or even ten seconds and more from collision (which is miles away in urban riding) and to remember you’re there when you enter the “killing zone”, which is where YOU cannot avoid a collision if the driver starts to emerge.

But the main drawback of hi-vis and DRLs is that just when you’d imagine they would be most effective (because you’re so close to a driver it seems obvious he can’t fail to see you), is when they don’t work and that’s when you’re at risk of a “looked but did not see” accident.

Kevin Williams

© Survival Skills Rider Training 2011

[Fuzzrr]

Harry and his expert team spent 1979 and 1980 racing to the scene of motorcycle accidents in southern California

I sort of decided to take this with a pinch of salt after reading that first sentence......... bloody americans!! 

But I read it through....... Interesting if it is correct. It now proves that the only reason the 'Hi Viz' brigade wear that clothing is not for safety....... But to try and look important. 

[snoopy]

Car drivers need educating to not just ''look'' but to actually SEE ... not just the cursory glance ...   

[thundercatsho]

  My brother is a driving instructor and a biker so tries to teach his students some awareness when it comes to bikes.
But he would tell you there is nothing specific to bikes in the training given to new drivers.
We are just classed as other road users and therefore no emphasis is put on looking for bikes unless the instructor see's fit.

[blueracer]

 I have to agree with the above we are classed as other road users

[thundercatsho]

Its not fair I tell you I demand special treatment 

[snoopy]

Its not fair I tell you I demand special treatment 

Me too   

[blueracer]

Didn't your mum ever tell you ,you were their special one 

[snoopy]

Didn't your mum ever tell you ,you were their special one 

    
Yeeess and we all know your special too     

[blueracer]

Stop it you will get me all emotional 

[snoopy]

Stop it you will get me all emotional 

        

[ja6on]

hmmm interesting read, nice post 

[lewy]

hmmm interesting read, nice post 

for more of the same look at the site that it came from
You are not allowed to view links. Register or Login

[Lunkhead]

Interesting, I agree with much of it. The problem is finding the answer. I'm of the opinion that you can rant all you like at a learner driver, broadcast adverts etc, but once the driver is settled down and relaxed with Ken Bruce and 'Pop-Master' on radio 2, all that education is forgotten. The driver goes on autopilot.

I think that we have to take responsibility for our own survival. Assume that the driver waiting to emerge from a juction as not seen you, even when looking directly at you. Get on autopilot youself as regards this, one day it will save your life.

Many, many vehicles have pulled out on me over the years, even when I'm wearing hi-vis, headlight on and riding a bloody great Pan with the most violent colour scheme known to man.

I'm not sure how we can make drivers notice us, but I do know that if I can't cope with it then it's going to hurt.