Are older drivers less safe than younger drivers? Does just having a licence indicate you are more able to drive? What is the point of certification?

For more information on Aleph Insights visit our website https://alephinsights.com or to get in touch about our podcast email [email protected]

Hello and welcome to the Cognitive Engineering Podcast produced by me, Frase McGruer for Aleph Insights. In this series of podcasts, we take a look at interesting topics and discuss what we think they tell us about analysis and decision making. I'm here with Peter Coghill, Nick Hare, and Chris Wragg of Aleph Insights. And this week, we're discussing Chris's mother-in-law. Chris, tell us about your mother-in-law.

Well, it's not so much my mother-in-law, who I obviously love dearly. It's more about the fact that she has recently reached the age of 70. And that means she's having to reapply for her driving license. And so this got me thinking about kind of what a driving license is, why we have them. And also, this idea has been going through my head because it's been a recent TV series on ITV, which has been looking at drivers close to the age of 100 and sort of assessing their ability to drive. And that's the whole program. They've been following them around as they go through. And you're no spring chicken yourself. Yeah, obviously.

And Chris has got a preoccupation with death and age and all sorts of that sort of thing.

Yeah. So no, I've just been just thinking about this process. And of particular interest to me is the fact that as a 70-year-old, you reapply for your license, but you don't have to take a test again.

I think it's called renewal, I've checked. Yeah. And it's called renewal, which sounds a bit like a cult you can join, somewhere in San Francisco.

I checked on Saga, which is definitely a cult you can join. In fact, you join just based on age. And yes, it's just an administrative process. There's no test.

Yeah, so every three years, you basically send off a form saying, can I have a new driving license, please? And they say yes.

And I think the idea is that it's to make you reaffirm and make you reassess whether or not you need to drive anymore and to sweep up any of those drivers, those very infrequent drivers who might jump in a car and they've not been in a car for 10 years, and they are more likely to have an accident than someone who is at least a regular driver. Yeah, but why is that age?

I mean, there is because obviously even normal people, not old people, but normal people have to have to renew their license every 10 years anyway. Yeah, I think I think I don't know.

I think the DVLA would probably like to do it to everybody, but it would be too much of a burden on them in terms of the administrative costs. But also, there'd be a huge uproar of people's inverted commas, the right to drive is being threatened. I mean, you have to.

Sorry, go on. Well, I was just going to say, you know, fundamentally, the question is, does having a driving license make you a safer driver or does everybody having driving licenses reduce risk on the road? That was essentially, you know, the question that sprang to mind.

Do you mean literally someone giving you a driving license, does that make you safer? I think the logic, isn't it, is that the thing you have to do to get the driving license, which is learn to drive, is the thing that makes you safer.

But at the point at which you have a driving license, does renewing that driving license and having to go through that process, which obviously has an administrative cost associated with it, does that make our roads safer?

Yeah, I mean, so I've tried to work out, because I know there's always those hilarious stories, aren't there, about an 85 year old driving the wrong way down the M1 and stuff. Hilarious, unless you're on the M1, yeah. And I naturally assume that old people were demonstrably worse at driving, but I have to say, there isn't any evidence of this. The people who definitely are the worst drivers are people between the ages of about 15 and 25, well, let's say 17 and 25. Bit like us. There isn't. Now, old people do get into marginally more crashes, and you might think, you know, that they are probably less likely to be on the road in the first place. So we might think the rate was higher. But as a cause of death, you know, something like, if you're 20, 25% of 20 year olds who die, die from traffic accidents. Whereas a tiny proportion, it's like 1% or something of, you know, 80 year olds die from traffic accidents. That's because they're all dying of something else. And of course, you know, but then on the other hand, they're more likely to die in a traffic accident anyway, like, you know, a minor, a minor graze of the bumper might kill some of them, but it doesn't. So the point is that there isn't, even though there are hints that they might be more dangerous, there is no, there is no sort of body of statistics I can find that's demonstrably show that they are, you know, more dangerous as drivers. So there's that. So we can't, and obviously, organizations like Saga are very keen to point this out. But honestly, it's not there in the stats. You can't, there's no stats there, which unambiguously show that old people are dangerous.

But I think it was due to some sort of grandfather clause, you didn't have to apply for one if you already had one. That's right, yeah. So now I come to think of it, that could be an explanation for why traditionally we think of old people as bad drivers because, you know, 30 or 40 years ago, old people didn't have to have done a test at all. They might not have learned to drive.

Yeah, but I don't think that's why we think of them as being bad drivers.

Well, let's just say that there might be a cultural memory of old people being crap drivers because they actually used to be. But what you're about to say is, no, we think of them as bad drivers because they're all demented and doddery and can't change the gear shift and they stay in second gear in the middle lane. But what I'm saying is there's no evidence for that. No, no, no. There might well be evidence. It sounds plausible, but it's not.

Hold on. This is something that James Bond said, actually, which actually he said something terribly sexist, which is that women are often safe drivers, but they're rarely good drivers. Right. We're talking about the same sort of thing, I think, which is the doddery old people, which I'm not that far from. Maybe they're already. They're just practically not terribly good at it, but they certainly don't take the risks that you would usually associate with accidents.

Sorry, what's the definition of good here? What does good constitute? Not dying seems to be quite a key part of it.

Yeah, I mean, he's ruined quite a few cars in his time, hasn't he, James Bond? So let's not make him the...

But also his car, if he drives it off a cliff, it turns into a boat. So it's zero risk there. Yeah, it's not fair.

But also logically, logically, from what we've been saying so far.

I love it when you say logically, because it's guaranteed that something completely illogical is about to come out of your gob.

You're right. So logically, we shouldn't let people drive until they reach 60 or something, because we've already said that they're the safest people. We shouldn't let young people. And so, yeah, young people should just not. No one's allowed to drive until they're 60 odd.

Well, I think, you know, if someone said, well, the problem is, are they bad drivers because they're young? Or are they bad drivers because they're inexperienced? Because they haven't accrued any experience. Yeah, I mean, I would say... I think it's because they're young as well, though. Well, we don't know. We can't tease the two things apart. What I can tell you is that I didn't pass my driving. We can't. How?

We can. Because how you would do it is you divide it up by sex. So if we sort of put aside youth for a moment, so most accidents, they're not just by young people, they're by young men. And so therefore, we can separate it from age, because men do have different characteristics. They do drive differently.

I know, but young men and young women have a vastly higher rate of accidents than middle-aged men and middle-aged women. That's a much bigger gap than the gap between men and women. Right? Men and women are not... The difference... Are you sure? Are you sure? Yeah, they are safer, but they're not massively safer. But middle-aged people are massively safer than young people at driving. Right? So, you know, that's all. That's what the stats say.

I mean, the essence of this was whether or not making somebody do a test, accrediting somebody effectively, makes them safer at doing something. And, you know, this is something that's quite difficult to look at, because we've always had the test, so we can't remove it as a, you know, as a sort of a factor in our analysis. I mean, there have been times during the Second World War they stopped doing driving tests, because all the people doing driving tests went off to ration fuel, you know, be involved in fuel rationing. And the same was true after the Suez Crisis. But those... And actually, accidents went up during the Second World War, but that is... Wait, what?

Wait, hang on. You're saying that all of the people who were driving instructors or driving examiners... Examiners, yeah. ...all got drafted into the Fuel Rationing Corps? Yes. And then that happened when OPEC formed and during the fuel crisis in the 70s?

No, in the Suez Crisis. Oh, sorry, the Suez Crisis. Yeah. So, yes. And I imagine the reason for that was because the examination process was, you know, they were sort of civil servants who could be reallocated somewhere. But the point is, you know, you wouldn't have necessarily expected a blip in accidents. Accidents did go up in the Second World War, but that's more attributed to things like the blackout and, you know, massive changes in conditions. But if you look at countries that have more or less stringent driving recommendation and then look at the effect that that has on driver safety, apparently Brazil is somewhere where... I haven't found any methodical studies of this, but looking at places where they have more stringent licensing processes and using those as case studies, Brazil is apparently very difficult to go... It's a very robust testing process and... No, not true? No, it's not actually. Okay.

It might have changed recently. Right. But actually, I was horrified to learn when I lived there that when you take your driving test and actually up until your driving test, you're not actually allowed to go on the roads. You're not allowed to drive on the roads. You can only drive on a public highway once you have passed your test. Right. And so where you do all your learning is on essentially what is a huge car park with bollards on it.

Maybe they're measuring stringency wrong.

Yes, yes. I think that must be the case that they... Whereas we have provisional licenses and you get, yes, that you're not permitted to drive until... But Brazil has a much higher road traffic incidents, accident incidents than countries like the UK. In South Korea, apparently, they are very rigorous on the health check aspect of it, which is analogous to us, I suppose, getting 70-year-olds to reapply with the being a sort of health implication there about eyesight or hearing, perhaps. And South Korea has, again, higher levels of traffic accidents than countries like the UK or European countries.

I'd never thought of this, but it does... I'd never even considered the idea that getting rid of a driving test might actually not have a big impact on road safety. But if actually thinking that it's experience which is the biggest predictor of whether or not you're a good driver, which the statistics kind of suggest, maybe actually the difference between getting tested or passing a test and not passing a test compared to the difference between not having driven before versus having driven for 10 years is so massive, maybe it really doesn't make that much difference.

And I think the sort of broader thing that I started thinking about is accreditation processes in general, where you have to behave in a certain way to pass some test, and you then get given a license to do something. And this applies to us in particular for our sort of IT security. We went through a process recently, and it made me start thinking, does that make us more cyber secure, that process? Does driving make you a better driver, or does it just make you behave during an assessment in a particular fashion? And then after that, you then start driving with your eyes closed or whatever. Yeah.

I mean, I think this brings us back to our old friend signaling. Not mirror signal maneuver type signaling, but signaling where you show that you can do something. And that process of showing that you can do it isn't the thing that actually makes you better, but you might as well act like it is. So classic example being something like getting a degree, where people who have degrees, if you look at their productivity, and you do a sort of comparison, life-like comparison, people who've actually got a degree and people who are exactly the same, but just didn't go and get a degree, but have all the other characteristics, they're equally productive. But the thing is that what it does do is show that you are able to go to lectures and write essays and do all these other things, which are correlated with the things that people want. Well, I've been to university and I can't do either of those two things. But I mean, I know it's a catastrophic failure of the education system. But yeah. I mean, and I think in the context, if we think about that in the context of the driving renewal, it's not the renewal process particularly. Maybe it's simply the fact that someone is compostmentous enough to be able to go through the renewal process. The fact that someone is able to fill in a form and send it off tells you quite a lot of information about the fact that they're still with it, they're still able to manage their diary, still able to fill in a form. That might be pretty useful information.

Yeah. I just want to bring in Peter at this stage. Anything to comment at this stage, particularly on this signalling idea?

Yeah. Well, Nick's rather stolen my thunder on that. I was going to try and broaden the subject to having a particular license or a particular credential. What does that tell you about someone's competence? But wouldn't it be nice though if you could spot all those self-educated, extremely productive people that send in CVs to your company with some kind of test, some sort of method that would tease out those people who are potentially at a lower price point in terms of the labour market because they haven't been to Oxford and Cambridge and places who are just as good if not better than those art historians from Oxbridge who seem to be highly self-entitled.

Yeah, but the thing is about signals is they only work if they're costly. There isn't a cheap way of getting signals. I mean, that's the thing.

But I'm suggesting, can we engineer a test that provides, gives somebody an opportunity to signal in a different way that's more telling?

But doesn't that sort of happen with standardisation, standardised tests for the civil service, for example?

Exactly, yeah. And job interviews are a good example. So you interview somebody about their competence. So rather than say, oh, what school did you go to and who's your dad? You ask them, give me an example of when this sort of thing has happened and what did you do in response to that? Which I've always thought was really rubbish kind of interviews.

And then you say, when I went to this school, my dad told me that. Yes, yeah, yeah.

Give me an example of when you're really, really disappointed. My dad didn't give me the job at his company.

But I think competency-based interviews are always awful because you just make stuff up.

Have you actually had to interview people?

I have interviewed people, yeah.

Yeah. No, I mean, the good thing about competency-based interviews from an interviewer's point of view is you are able to collect information in a much more structured way. You know, you ask the same five questions to all the candidates and then you have your evidence there and you can go, right, this guy said this. And, you know, that is better than a kind of, you know, what sort of a chap are you type interviews where you really don't have. It's much harder to lie about. I mean, this is why this is, again, signaling. It's just harder. It is possible to lie in a competency-based interview, but it's harder to lie. You know, if you have someone who's self-deluded and narcissistic, they will tell you that they're amazing and brilliant at everything. And the point about narcissists is they're often quite plausible. Whereas someone who is a narcissist and has never done anything of note in their life has to then also put in the effort of making things up and lying. And it's much easier to be caught out in an interview where you're having to deliver facts. That's the idea. Now, look, it's not perfect. The problem with these kinds of situations is that there isn't ever going to be a signal which is going to perfectly divide people. But, you know, but it's better. It's an improvement. I've heard, actually, recently that the civil service are abandoning them again. Are they? Yeah.

What, the competency-based questions or the… I think so. I don't have enough information.

I shouldn't have raised… I've got no information about this, so I shouldn't really have brought it up.

Maybe they demand too high a standard. They're not recruiting enough people.

OK. So let's wrap up. I want to talk about driving. Nick, are you a good driver?

No. I'm a terrible, terrible driver. And I hate driving. But, see, crucially, I passed when I was 30. And I have this feeling that people who learn to drive early haven't learned enough about their own immortality and their own cognitive limitations. And so that sort of total misplaced confidence gets them through the first few years. And then at the end of it, they've got enough experience to compensate for however good they are at driving. Whereas when you're 30 and you learn to drive… Well, I mean, I learned to drive when I was 17, but I quite rightly failed my test twice and gave up until I became a dad. And I thought, you can't be a dad who can't drive. That's illegal in this country. So I decided that I had to pass my test. And then, you know, since then, I've really had to force myself to drive down to the shops and things. But I hate it. I mean, I can't stand driving. I find it unbelievably stressful. I feel like I have to pay attention to everything. I haven't ever got near that stage where you can sort of relax and just go, well, I'm in control here. And I have to think about everything. You know, every time I change lanes, I've got to think, oh, what should I be looking at? You know, so, yeah. So I would say that's entirely a lack of experience. It's an experience that I'm just not ever going to be able to gain because I'm so deterred from driving.

Yeah. No, you don't surprise me. You know, your bearing is that of a person who should be kept away from three tons of vehicle, you know, moving metal. So, yeah. We're not getting him to drive HGVs just yet.

Chris?

Am I a good or bad driver? Yeah. Well, I'm definitely a very different driver than the driver I was as a 17-year-old boy. Yeah. And having children. Doing donuts in the car park. Yeah. Having your genetic inheritance in the back of the car is definitely something I think would, yeah, temper young men's driving.

Yeah. Okay. I mean, I'd like to say the same. And I'd like to think I'm a good driver and a very different driver to the one that I once was. However, just this very morning, I was in a rush to get to the station and I had my, you know, three offspring in the back of the car with me. And I did make a foolish sort of pulling out at a junction right in front of a white van man who made his feelings about me and my driving expertise very clear to me. And, yeah, no, I still sort of do silly things, I guess.

Yeah. At least you didn't let them drive.

True. Yeah.

That would be irresponsible.

Yeah.

Peter, good driver? I think I'm exceeding the average. I think I've had a similar sort of a very mean number of accidents and tickets and all sorts of other things. I don't like driving either. I think it's incredibly stressful and tedious activity. I can't wait for driverless cars. I can't wait just to outsource that and so I can read a book in the back. Yes. But I am consciously cautious and dry and my girlfriend is forever bemoaning the fact that I'm not driving fast enough or ushering on the car in front more. But I just don't see the point because you never get there much quicker.

Yeah. I have this feeling that there's basically there's nervous drivers and then there's deluded drivers. I don't think you should ever not feel nervous when you're driving. If you're relaxed and comfortable and you feel in control, you're wrong because you're not. And you're going to die. You're just stopping. You're just an accident waiting to happen.

Yeah, exactly. Despite all the statistics of the show that I'm travelling by.

I'd say when we sort of, you know, it was turning to the Aleph Insights Cognitive Engineering Podcast Roadshow, probably the people I would want driving the car would either be Peter or Chris probably.

You wouldn't want me or Nick. Well, we'll get a driver. We'll get a driver. Because I love driving, but I don't love driving. I love being driven. I love staring out the window.

We'll be on our own private jet.

I don't trust anyone who's my age or younger to drive cars. I want it to be a 55-year-old man called Reg who got aviators on. That's the only acceptable driver as far as I'm concerned.

Reg, we're waiting for your call. The blue blazer. Yeah, exactly.

Polished shoes. Okay. We'll stop there. Okay. Well, look, thank you, as always, for listening to the Cognitive Engineering Podcast. So I'm Fraser McGruer. We've been here with Peter Coghill, Nick Hare, and Chris Wragg of Aleph Insights. Until next time. Goodbye.