With Mrs V's test looming and her theory to a good standard we though it best to try out the Hazard Perception Test. For those of you who don't know it's comprised of 14 video clips of assorted road scenes. 13 of the clips contain 1 developing hazard and the other contains 2. You need to click on the button when you identify a hazard which could result in you having to slow or change direction. Throughout each clip there are any number of minor potential hazards which might cause you to do this or where you'd be right to see at least the potential for danger and you can indicate these but what they're really looking for are the major hazards. You need to click to register the major hazard during a particular time slot in order to gain marks. IIRC the pass mark is 40/75. Easy eh? Well not so. It appears possible to register a hazard too early, think you've done all that's required but come away with a big fat zero! It seems that in this situation too much anticipation is just as bad as none and that can't be right can it? For example I clicked when I saw road works signs immediately ahead but didn't click again when I reached the actual location just a short distance ahead. For this I got 0. The dilemma is that if you click to early you get 0 but if you click again and it's deemed too late (even though you've aleady recognized the hazard) it's assumed that you haven't acted in sufficient time.
Also, the software has a habit of assuming cheating if you click too regularly and awarding 'nil points' too. This happened to both of us even though we weren't trying to cheat! Of course there needs to be a method of preventing people just clicking repeatedly but I can't help thinking this isn't very effective.
All in all though it's a good test and worth trying (if you dare). There are some problems which need to be better adressed however. It's true that the test software we're using isn't exactly the same as that used officially but is supposed to be a very close match.
|
You'd think they would be able to create the software so that you had to click on the hazard itself eg the roadworks sign rather than just click to say 'there is a hazard'. Otherwise it is a bit of a waste of time, especially if it encourages people to look just in the 3 or 4 car lengths in front of their bonnet rather than at the whole distance they can see ahead.
teabelly
|
I agree with what your saying teabelly and it would be a fantastic way of doing the hazzard test, rather then the too many clicks and your out method.
But this would take an age to programe, as the co-ordinates for the hazzard would have to be pin pointed for each frame, this would cause a lot of coding for just one clip let alone a database of clips. And with how todays computer systems are poorly designed, it would have a lot of bugs in it that would take an age to fix.
|
Agree with Teabelly - that would be a much better idea. On one of the clips we had, the test car was driving down a very busy high street. There were all the potential hazards one would expect (pedestrians, parked cars, buses etc.) all of which were present most of the time. Quite how they determined whether you'd clicked for the designated 'hazard' or something on the periphery you saw at the same time I've no idea. The problem is that wondering how they want you to respond wastes valuable time and if you click again, just to be certain you've made the point you can get marked down it appears.
One aother aspect of viewing all this on a screen is you get a very small 'window' to view and are deprived of the normal panoramic view you have when driving. This is quite unnerving at times and makes the whole process a lot harder IMO.
Imagine driving along with a cardboard box over your head and just a small slot to view the road from - it's a bit like that.
Anyway, where are all the BR driving instructors when you need them? :)
|
Anyway, where are all the BR driving instructors when you need them? :)
>>
Will an ex instructor do?
The problems you had with your software is exactly the same on the software used for the real test. When I first had a go I scored 38, when I admitted this to another ADI he owned up to only 36! Former police traffic drivers who are qualifying to become instructors have also got a very poor record on the Hazard Perception test, and the problem is as you've suggested, if you click immediately you spot the hazard you are outside the scoring window.
The whole idea is a bit ridiculous anyway, the only way to tell if someone is hazard aware is to sit beside them on the road and see how they react to real life situations, not how they perform on a glorified computer game. Most instructors will echo my own experiences, that the ability of a pupil to score well on the theory test bore no relation to how they dealt with hazards on the road, in fact the reverse was often true.
A classic case of education for educations sake, with no advantage to ones driving ability whatsoever.
|
I had a go at the motor show DSA stand a couple of years ago.... As an IAM member, ROSPA Gold (Class 1) driver and previously an Advanced Driving Instructor I thought I may do "OK" but failed pretty dismally for all the above reasons (mainly anticipating too early).
As it was a preview day with no competition and queues on the DSA stand I hogged the Hazard Perception Test carried out an experiment and found that if you shut your eyes and click randomly you can actually do quite well and pass, as long as you do not click too many times.
I believe instructors teach more how to pass the test than how to anticipate hazards......
|
|
I took my theory test about two years ago and was concerned about whether you actually had to click on the hazard or just click anywhere on the screen when a hazard appears. At no stage in the instructions was this explained, it was particularly confusing as the cursor remained on the screen during the test so you may have been expected to move it and click on the hazard. I ended up having to ask one of the supervisors- you just had to click anywhere on the screen when a hazard appeared. I don't think it is sensible to expect people to click on the hazard as this would give people who use computers often a clear advantage over someone who is not used to using the mouse. This extra time taken to position the mouse before clicking would lead to the "reaction" time being reduced. This would result in a test on people dexterity with the mouse and not the time it takes them to react to road hazard situations.
|
|
proper simulator as used for fast jets is the only way, too expensive no doubt, otherwise a waste of time
|
|
|
|
>>The whole idea is a bit ridiculous anyway, the only way to tell if someone is hazard aware is to sit beside them on the road and see how they react to real life situations, not how they perform on a glorified computer game.
I am not an instructor but fully agree. experience doesnt come on a computer screen
--
Steve
|
A thought I have had for some time is that all this stuff may be just to stop people becoming drivers, and hopefully to make them public transport fanatics!
|
|
|
|
Phew Tom - I'd thought it was just me for a while there :)
My marks on the test have ranged from 44 to 57 so I haven't failed one yet but I don't feel at all comfortable after half a dozen attempts.
We reviewed the official DSA explanatory CD Rom last night and it appears that the official test will be judged in a very similar way but the video quality is probably a bit better. It seems they want recognition at the earliest point you see the hazards plus additional recognition during the time window when you have to anticipate and take action. The critical question is when these windows are and for me they don't seem quite right as you've said. If you've correctly identified a hazard very early and (as I would do) have already eased off a little, say, the last point atwhich you actually need to brake/swerve will be different (later) from that of another person who's seen the hazard at the same point but carried on at the same speed and will be have to take more aggressive action when the hazard develops. The software doesn't seem to recognise this. Possibly a steering wheel and brake pedal connected to the computer (just like on computer games) would be a better way of assessing the individual's response. Perhaps the truth is that because it's designed for learners it's never going to suit more experienced drivers and doesn't need to.
|
|
Just as an afterthought, it's also quite odd trying to react to what's going on when you're not actually 'driving' the car but effecively reacting to someone else's driving. It's more like being in the passenger seat of a car than anything esle. The more I think about this the more I feel an advanced interactive setup when you're asked to 'drive' through a series of virtual scenarios would be a better test. Given the quality of current DVD games I can't see why this would be too difficult but presumably cost is the major factor.
|
Agree with VM. I'll admit, I like my racing games and some of them are surprisingly realistic to the point where if you brake too hard, the wheels lock up. If you brake hard in the wet, you go sliding. With the rear-wheel drive cars, you even lose it on a corner if you give it too much gas. Something like this would be idea
-OR
Here's an idea. How about a REAL life hazard perception test. We could make people drive cars with an examiner in them and get marked on their abilities. We could call it a driving test.
--
Adam
|
|
Yeah Adski, the DSA could pay an army of 'idiots' to appear on cue and do stupid things, like jumping out in front of you from behind a parked bus or cutting you up with no warning. Seriously though, I think the computer test is valid as it gives everyone an even playing field when the road test doesn't. It just doesn't feel like driving though.
|
>>It just doesn't feel like driving though.<<
That's my point exactly. Don't get me wrong - despite what it sounds like, I agree with you completely. Honestly I do but the problem is, it doesn't feel like driving. People are sitting at a computer so are going to act nowhere near like they would do behind the wheel of a 2 tonne hunk of metal.
I think I'd think differently if the software was spot on and by that I mean registering your early anticipation of the roadworks as oppsoed to waiting for you to be dead on top of them. When is real life ever like that?
>>Yeah Adski, the DSA could pay an army of 'idiots' to appear on cue and do stupid things, like jumping out in front of you from behind a parked bus or cutting you up with no warning<<
Sarcastic as that was, (and seeing as it's you I'm willing to overlook it ;-)), that's not a bad idea. I'm not for one minute suggesting that they should employ real people but what about a track with parked obstacles and "dummies" being thrown out. You're in a real car, and early anticipation would be awarded.
Fifth Gear can manage it (as can "Britian's Worst Driver") and you never know, you may come away actually learning to do something different.
--
Adam
|
I was joking Adski - don't know when I might need your free 24hr computer support hotline service do I :)
Don't forget the real test does allow your actual driving to be tested on the road and, yes, with the sort of hazard we all face every time we set off. Short of the scenario you describe in which a test track is used to assess driver reactions (is this feasible on the required national scale?) I believe the hazard perception test via computer is the next best thing. Although it can never realisticaly (for reasons of cost) emulate a real situation I think it would feel more realistic if you weren't being asked to click on a mouse button but had very basic car like controls with which to register your reactions. These could be much more finely measured and judged too - let's face it, it's hard for a computer to differentiate between an accidental or tentative mouse click and a carefully considered/ measured one.
|
Agree with you there. I have the Steering wheel and pedals for my games which let you change gear with a real lever, brake using pressure sensitive pedals and so on. The most realistic I think you can get in terms of what you suggest. I know I'm a big kid really!
Also, still waiting for that pint VM ;-)
--
Adam
|
Within five minutes of a driving test starting an examiner will have formed a far more accurate picture of a drivers ability to identify and react to hazards than any computer test ever will. The only reason airlines use simulators is even the multi million pound ones they use only cost a fraction of a Jumbo jet.
I would not lend my car to someone on the basis that he could get through Grand Theft Auto without crashing, anymore than England managers are selected because they are good at FIFA 2004. Computer simulations are fun, but they are not the real thing.
|
But the point is that if you can't succesfully pass the hazard warning simulation in the theory test they won't let you do it for real in the road part of the test.
I have to add my voice to the critics as I can't believe that a better hazard warning test could not be developed fairly easily.
I wait for this to become a problem at home as my middle child will be doing the theory / hazard test in the first part of next year.
|
Well I think we need both the computer based HPT (which could be better) and the 'real thing'.
Adksi - sorry about the pint, I'm e.mailing it as an attachment now via DD et al. If it doesn't arrive one of them will have had it :)
BTW had a bit of real life hazard awareness experience this morning. Like so many things it arose out of almost nothing. I'm proceeding down the main road and a guy in a Hyundai Coupe emerges slowly from a side turning as I approach. I brake slightly, he eases off and as I proceed ahead he follows along what is now a one way road. Next thing I know, white van man has roared up behind us both at high speed, overtaken the Hyundai on the left, cut in front of him and then attempted to pass me on the right despite having no room to do so. Hyundai man of course gives loud blast on the horn and WVM responds by slamming on the brakes. What follows is a stupid and senseless cat and mouse game of 'see if you can get past me then pal' on my right followed by the inevitable face to face confrontation and wrestling match! Fortunately I was turning left just ahead and managed to stay out of their way. Pathetic really.
|
|
|
