Hacker News Reader

Comment by jschwartzi | Hacker News Reader

When I was taking driver's ed as a teenager the instructor showed us a video that illustrated a driver's gaze from the first person as they drove using a small circle. The difference between a skilled, competent driver and a student was that the competent driver only fixated on things that were likely to matter. So mostly their gaze was on the vehicle in front of them, followed by any place that could hold moving objects and pedestrians, then briefly signage. In contrast, novice drivers' gaze was all over the road, the signage, pedestrians moving parallel to the road--in short they had no mental filter for their attention.

View on HN

Intriguing. A friend of mine just learnt how to drive very recently. He told me he was able to pick up driving much faster than an average person as he was able to apply the skill he developed in many years of playing Counter-Strike Global Offensive in driving: Instead of looking everywhere he would focus his attention on things that likely to matter and react to it like he was playing the first person shooter game. He once managed to reach top 1000 in the country for the game and he took roughly half the number of lessons an average person here needed to get his license. But I wonder how many of what he believed are true. Have there been any studies done on things like this?

Will have to source it but there is a video game where you can play to win a chance to be a race car driver. One of the guys won the contest and then won the actual race with a real car. The skills apparently do translate [1].

[1] https://www.theguardian.com/technology/gamesblog/2014/apr/07...

Note that GT Academy winners get extensive race driver training as part of their prize. I agree that the skills translate (somewhat), but people should be aware that he (the guy in the article) didn't go from playing Gran Turismo directly to racing an actual car and even then it took a while before he started to be competitive.

One study found that surgeons who regularly play video games are better at laparoscopic surgery.

https://www.ncbi.nlm.nih.gov/pubmed/17309970

Correlation doesn't imply causation, of course.

He may have picked up driving quickly not because of Counter Strike, but for the same reason he was capable of becoming a skilled Counter Strike player- some genetic or psychological advantage that is independent of anything else. He may just be better than the rest of us at filtering and focusing!

In contrast, I played a bit of counter strike in high school but I never got even remotely good at it. I just didn't have any knack for it. And while I am now a fairly skilled driver (no accidents in 16 years on the road!) I suspect those early years it was a matter of very good luck that I didn't get myself killed behind the wheel.

Experience teaches you the noise to ignore. Same thing applies to all sorts of disciplines, even debugging or fixing a code.

An experienced dev can look at a large code base and spot a bug or just from the structure of the code tell if it's "badly written" code.

An experienced sysops can look at large streams of system log files and tell you what's wrong with the system because they can ignore the noise. Someone else might be fixated on the high cpu load when the issue is a bad device.

Experience is knowing what qualifies as noise, and when.

Just because a system is running doesn't mean it's running well. Just because code is orderly in appearance doesn't mean it's effective at performing the desired task.

Perfection is the enemy of good.

Sometimes a shambling mess of spaghetti is that way because of a lot of underlying and tangentially-related issues with the operating environment. Code can be perfect looking yet fail because developer expectation of the compiler/toolchain and the underlying OS don't actually function as documented. Of course this can also lead to things being brittle when the underlying issues change. Experience is knowing when those issues are involved.

I had something like the opposite issue. I've got my permit and not a license yet. Often when I drive, I feel I'm not paying much attention. I'm almost spacing out. I felt a bit guilty about it. I was leaning toward undiagnosed ADD/ADHD being a factor, but by your logic it sounds a bit like I've adopted the mindset of a more seasoned driver right out of the gate. I have driven small vehicles like ATVs off-road, so actually maneuvering isn't a major challenge, but just the environment of urban roads feels very different. Having to care about rules and such adds a lot of mess to the experience of driving.

This is a pretty universal thing in complex tasks. As you get used to things and your brain translates them into familiar patterns, they take less effort to anticipate (or produce,) so you can string lots of them together without being overwhelmed. Like when you see a piece of piano music that seems to be completely black from so many notes. Or when you watch olympic diving and wonder how a person is going to go from doing a perfectly still handstand 10m up, to launching into multiple flips and then somehow be calm enough by the bottom to think about not splashing at the end.

So you learn to recognize the set of signage you're expecting in a given pattern, and you recognize where the likely spots are for surprises, or you remember to watch for heads in the cars along the side so you know if someone's about to open a door.

Was that real eye tracking or a fake for pedagogical purpose?

It's the same with everything. After a while, in college, one learns what the important things to pay attention to are in class, and what aren't. It makes one a much more efficient learner.

Motorcycle classes with the Maryland MVA focus on SIPDE.

https://www.google.com/search?q=sipde

Scan, Identify, Predict, Decide, and Execute

Don't know if that correlates with 'quiet eye' but it does put emphasis on the eye being engaged in a specific activity (the scanning). Looking for what's out there, with the intention of knowing about it, versus just randomly looking around. Or worse blankly staring without focus.

I agree. I have long thought that a large part of learning to drive is learning where to look and when.

Funnily enough, attention is one of the things that help deep neural networks learn.

I think I watched that same video. I actually took it to heart and that's been the basis for where my eyes go ever since I've been driving.

Fascinating comment. Makes me want to read random eye tracking studies.

My driving instructor (in the car) is really against this. He talked about becoming fixated on the car in front of you and then lots can go wrong.

Mostly on car in front is not exclusively on car in front.

The car in front of you is by far the greatest risk to your safety that you can control. the car in back is a large risk but you have minimal control.

Your eyes should be scanning at all times. Those 'experienced' drivers have literal blind spots and are likely to enter a mental state akin to a sleep state.

That’s not really what allows people to drive safely. Without filters youd treat stationary objects with the same deference as moving ones.

Well it works for Tesla's autopilot right? /s

I never said you should drive without mental filters, I said that your eyes should be scanning at all times. Lots of strawmen today.

Your advice is for an experienced driver who wants to become even better. The original commenter's scenario talked about novice vs. experienced. If you just tell a novice driver to scan everywhere, it's not actionable advice. Since, as was pointed out, they have not built a good filter yet. They'll end up being stressed out looking all over the place, when the simpler advice that driver's ed gives like "check these 3 places every 30 seconds", while far from perfect, will get them looking around a little bit and training that filter without becoming overwhelmed.

> The original commenter's scenario talked about novice vs. experienced.

It talked about novice vs. "skilled, competent" drivers, which suggests the latter meeting some threshold beyond just experience. (Making their behavior even more likely to be a demonstration of skill and safe technique.)

Nope, my advice is for all drivers, as is taught to motorcyclists. Checking three places every thirty seconds is appalling advice.

Sadly, sites like this are full of ignorant, unskilled, overconfident motorists who don't like correct advice.

So what you’re saying is, learn from those most likely to get into car accidents, rather than the least likely?

https://article.images.consumerreports.org/prod/content/dam/...

What an absurd strawman.

That's what properly aligned mirrors and looking over your shoulder are for..? Any sleep state you could be referring to is caused by (a) exhaustion or (b) being unaware of being a bad driver.

A book on safe driving I read said in normal driving your eyes should mainly be on the car ahead, and every 5 or 10 seconds glance at your mirrors to see what else is going on and also do a broader scan. And of course in different circumstances like turning, you should look around much more.

Reply on news.ycombinator.com