Using Apple CarPlay Impairs Driver Reaction More Than Alcohol, Study Shows (appleinsider.com) 85
U.K. based road safety charity IAM RoadSmart has released a study showing how in-vehicle infotainment systems can substantially impair driver's reaction times even more than alcohol and cannabis. AppleInsider reports: The test involved two experimental trials focusing on Android Auto and Apple CarPlay. Twenty Android and twenty Apple users would be subjected to the same simulated test route. Each driver drove three times: A control drive, where they did not interact with the system, a voice-enabled drive where they used the voice system only, and a touch-enabled drive, where they used the touch screen of the system only. As part of the test, users would be required to follow another car, navigate an erratic motorway, and perform a figure-eight loop. During these tests, users would be told to perform a task, such as accessing music on Spotify, input data into a navigation app, or read texts and take phone calls. Additionally, participants would be asked to flash their headlights whenever a red bar appeared on the screen, which measured their reaction time to external events.
Regardless of the infotainment system, all users showed significantly slowed reaction time. Undistracted drivers typically showed a one-second reaction time. Those who used the voice-controlled Apple CarPlay saw a 36% increase in their reaction time, which rose to 57% when they used the touch interface. Android Auto users faired only slightly better -- a 30% increase in reaction time when using voice control, and 53% when using touch controls. For comparison, those who drive under the influence at the drink-drive limit showed a 12% increase in reaction time, and those who used cannabis saw a 21% increase. Those who used Android Auto saw a 1.73-foot (0.53-meter) deviation on their lane positioning when performing navigation tasks with Android Auto. Those who used Apple CarPlay saw a deviation of 1.64 feet (0.50 meters.)
Regardless of the infotainment system, all users showed significantly slowed reaction time. Undistracted drivers typically showed a one-second reaction time. Those who used the voice-controlled Apple CarPlay saw a 36% increase in their reaction time, which rose to 57% when they used the touch interface. Android Auto users faired only slightly better -- a 30% increase in reaction time when using voice control, and 53% when using touch controls. For comparison, those who drive under the influence at the drink-drive limit showed a 12% increase in reaction time, and those who used cannabis saw a 21% increase. Those who used Android Auto saw a 1.73-foot (0.53-meter) deviation on their lane positioning when performing navigation tasks with Android Auto. Those who used Apple CarPlay saw a deviation of 1.64 feet (0.50 meters.)
Well, duh. (Score:5, Insightful)
I've been saying for the better part of a decade that touchscreen interfaces in cars are inherently unsafe unless the car can drive well enough to let you take your eyes off the road safely for several seconds, because realistically you will be doing so every time you want to change anything. It is the worst possible way to build in-vehicle controls unless you have fairly solid self-driving tech.
Re: (Score:1)
Stuff like this just makes me feel old. I remember when the only electronics that cars had was an AM radio. The phone stayed at home. My most recent job involved a 25-mile commute each way. I can't begin to tell you the stupid shit I saw on the road particularly in the wealthier and university areas. Like people who would go 10 miles without paying any attention to WTF... looking in my mirror
Re: (Score:3, Insightful)
Whatever happen to tactile feedback? I can switch sources or channels or whatever on my car by feel, or mostly so for something complicated. Even when I'm not entirely sure my hand is on the right control, I can glance down and back to the road to confirm, and never have to stare at my control while I fiddle with them. And most of the time, it's just muscle memory, even when it's not a steering wheel control.
I sure as Hell won't by a car that doesn't have an actual volume knob on the dash. Quickly turn
Re: (Score:1)
I sure as Hell won't by a car that doesn't have an actual volume knob on the dash.
A volume up/down toggle on the steering wheel is even better. Recent Toyotas have both.
In general, the only safe app interface in a car that isn't self-driving is voice control. If you have to look away from the road at a screen, adjusting visual focus accordingly, you're already halfway to hell.
Re: (Score:2)
Well it is pretty clear, all controls that are active whilst the vehicle is in motion, should be law be mounted to the steering wheel or steering column. All touch panel controls should only function whilst the vehicle is stationary and set in park with handbrake on.
Re: (Score:3, Insightful)
And then the passenger can't do anything. I've been in cars that do this. It is worse than the problem they're trying to solve. The right solution is to do a lot less by visually aided touch, and a lot more with actual tactile interfaces, unless your car can drive itself, or at least mostly so.
Re: (Score:2)
Have a look at the Honda e. Two touch screens, one for the driver and one for the passenger. The driver has a single button that switches the two displays, so say they have navigation up and want the passenger to add a waypoint they can pass it over and let them do the fiddling around.
It's also got physical buttons for the radio and a volume knob, as well as steering wheel controls and voice commands.
Re: (Score:2)
And then the passenger can't do anything.
The passenger can use Toyota's physical volume and tuning knobs for the same functions that I can do on the steering wheel.
Re: (Score:2)
That doesn't help, sorry, the problem isn't touch screens.
My car has a screen with complex menu options. It's not a touch screen, and you can do everything though steering-wheel mounted controls. It's just as dangerous, because you have to switch your attention to the screen and fiddle with the damn controls to e.g. check your tire pressure while driving. I think it's more dangerous than a touch screen, because it's more time consuming selecting the right menu option.
Re: Well, duh. (Score:2)
Re: (Score:3)
A volume up/down toggle on the steering wheel is even better. Recent Toyotas have both.
No, it's not better - it's a nice supplement. I can instantly mute the radio with a knob, which is the most important thing one might do with the volume. Steering wheel controls are nice for normal adjustments while you drive, though.
In general, the only safe app interface in a car that isn't self-driving is voice control.
Voice control is garbage, and there's evidence it's a bigger distraction than other inputs. Glancing at your speedometer has never proven to be an issue. It's fiddling with controls that's dangerous. As the BMW controls were once described "you need to win the video game t
Re: Well, duh. (Score:2)
Voice control is only garbage on iOS. On Android, it's great.
Re: (Score:2)
Re: (Score:2)
Uh-oh, that radio is about to hit mach 3.
Re: (Score:2)
My car has a touchscreen. It disables itself whenever the car is in gear.
Re: (Score:3)
Re:Well, duh. (Score:4, Insightful)
Touchscreens should be banned. Bring back tactile controls. There is a reason for them.
Re: (Score:2)
I'd rather have a touchscreen than a physical keyboard for entering sat nav destinations.
Touch screens are fine for some stuff, like configuring stuff while the car is stationary. The problem is over-use and replacement of all physical controls.
Uh... don't touch the radio in a figure 8! (Score:3)
This is nonsense. 2-way figure 8 loops are unsafe to begin with... no wonder there's crashes.
Re: (Score:2)
That would be the flaw in this study, imo. The difference between someone using Android Auto, say, and someone hammered is that the person using Android Auto can choose not to look down at the head unit at an inappropriate time. The hammered person has lost their ability to discern when it is and isn't appropriate to stop paying attention to the road.
Re: (Score:2)
Observing other drivers, I wouldn't be to sure about people's choice when to look down as it seems to be whenever they have the urge rather then when save. All I can say is that it's amazing there isn't more accidents.
Re: (Score:1)
In short, a user distracted by the infotainment system is distracted while they interact with it. A drunk or drugged driver is impaired all the time.
Garbage study (Score:4, Informative)
Being forced to interact with infotainment system while doing difficult maneuvers is not a real world test.
People (with sense) interact with their systems when in a straight line, red light and when there are no maneuvers to do.
No one is choosing the next song to play when zig zagging through cones and being told to flash lights at the same time.
Re: (Score:2)
Those infotainment screens are an eye magnet, all these automated safety features are needed simply be
Re: (Score:2)
I do not disagree that infotainment systems are a distraction. And no distraction is better than any distraction for sure.
I just don't think this test/study has any merit at all.
Re: (Score:3)
No one is choosing the next song to play when zig zagging through cones and being told to flash lights at the same time.
I know people well enough that while that specific instant may not necessarily happen, something close enough will happen for a certain "select" group of drivers. There is also the fact that sometimes things can happen that the drive doesn't expect. Maybe the driver in front of you swerved around a rock in the road at the last minute or something.
Now, that isn't to say I think infotainment systems are bad or anything. I'm also not saying the test environment wasn't a bit extreme. But, situations can occur
Re: (Score:2)
I agree with you. They can be distracting and some peoples judgement is certainly flawed.
My reaction was mainly on the test and how it was conducted.
Re: (Score:2)
Re:Garbage study (Score:5, Insightful)
and when there are no maneuvers to do.
Potentially, there are always maneuvers you might have to do, sometimes with almost no warning.
Re: (Score:2)
Agreed. But for that matter, temperature knobs, radio stations, ECO buttons, etc.. are all distractions in the same sense.
I'm not saying infotainments arent a potential for distractions and that some people dont have bad judgement on when and how to use them, but the test parameters where poorly designed to give any valuable data.
Re: (Score:2)
Re: (Score:2)
I'm always amazed at the signs that say "prepare to stop" as you should always be prepared to stop if moving.
Re: (Score:1)
Potentially, there are always maneuvers you might have to do, sometimes with almost no warning.
Not when I'm standing at a red light...
Re: (Score:2)
Then you have to pay attention to the damned light so that you don't just sit there blocking traffic when it turns green.
Re: (Score:2)
> Then you have to pay attention to the damned light so that you don't just sit there blocking traffic when it turns green.
No.
a) If you enter the intersection the moment the light turns green, you risk being hit by someone blowing through their now-red light.
b) You have a couple seconds, therefore, where it is still reasonable to be sitting at the light.
c) Even if you are blocking traffic when it turns green, you STILL are not going to cause a collision.
Honk at me if you like, but you're not going to ra
Re: (Score:2)
When the light turns green, you use those couple of seconds to look for red light runners in the cross lanes. You don't use them to finish up your damned text message.
Just because people can't physically ram you for being a selfish prick, it doesn't mean you're still not a fucking asshole.
Re: (Score:2)
Never said you might not be a selfish prick. ... ... well...
But if you're calling people assholes because you lost 3 seconds when the light turned green,
there are many full-flavor decaff brands on the market that you should try.
Re: (Score:2)
It's not three seconds. As often as not, the selfish prick eventually looks up and goes, while most of the people behind him get stuck until the next cycle.
I have no idea why you think that's OK.
Re: (Score:2)
Your response makes it seem like you have never been stuck behind some moron on their phone, putting on makeup, reading a magazine, etc. while the light has been blazing green for several seconds. It happens to me at least a couple times a week.
Re: (Score:3)
Re: (Score:2)
The ones that used to scare me the most were women doing makeup at 100km/h on the Eastern Freeway using the vanity mirror in the driver's sunshade. You have to take your eyes off the road to do makeup - it inherently prevents you from looking at what's ahead of you.
Re: (Score:2)
I'll agree there. I poorly worded it and concede that there are people with bad judgement that will interact with their systems at the wrong time for sure.
Re: (Score:2)
This becomes a measure of how quickly a driver can respond and it applies equally to a figure 8 maneuver as it does to driving straight. It is difficult to measure this impact when driving straight so they use a figure 8 maneuver for the test - which makes sense.
In real life, a user will probably be driving straight when interacting with their infotainment system and a deer jumps out in front of them. This test provides clues as to by how much the technology impairs the driver's response time. At the
Re: (Score:2)
Sure. So why do they insist on comparing to people who are drunk, but otherwise concentrating on the road?
Compare to someone trying to get a screaming baby to shut up instead.
Re: (Score:2)
Or to someone having a conversation, seeing a plane fly by, putting in a CD or a cassette tape or an 8 track. Or fiddling with the temperature, sound level, etc...
Basically, if your distracted, you react slower...
Re: (Score:2)
Err, I drive on roads with signs that say to watch for low flying planes, so yes, watching for planes can be part of driving. Seems there were even 2 cases of planes making emergency landings on roads in the last year, with one on a busy highway. Nothing bad happened partially due to people paying attention.
Re: (Score:2)
Agreed. Additionally, chemically impaired drivers can't get unimpaired because they find themselves in a challenging driving situation. Distracted drivers can.
Re: Garbage study (Score:2)
I think it is also problematic that the drivers are using a system they are unfamiliar with. I notice that people with temp tags are always the worst drivers. I assume it is because they are driving a car they have not yet become fully comfortable with. It takes a long time to become well attuned to a vehicle. When I have to drive rental cars I never mess with the radio and all the extra features. You have to be quite familiar with the car to do so safely.
Re: (Score:2)
Thats a good point. Once used to your car many things are done by feel. Not all, but some.
Re: (Score:2)
except that certain interfaces now control everything and almost have no buttons or controls at the steering wheel left. Tesla & Volvo are brands that pop to mind.
in the model 3 you must even adjust the airflow of your vents using a touch screen interface, insane.
it's not feasable to stop or hope for a red light for every little things you want to do.
Re: (Score:2)
Good point.
Bullshit. (Score:1)
Don't freak out, dude (Score:3)
The cannabis users definitely could've reacted faster if they wanted to, man... but that would've harshed their vibe.
Re: (Score:2)
The thing with cannabis and I believe alcohol is that a heavy user doesn't get affected as much.
Re: (Score:2)
The thing with cannabis and I believe alcohol is that a heavy user doesn't get affected as much.
So, you're advocating heavy use to improve performance?
Re: (Score:2)
No, I'm saying a heavy user who lightly uses isn't as affected. Like the heavy alcoholic who actually calms down after a drink or two. Remember they were testing people at the legal limit, 0.08% I believe.
Re: (Score:2)
Re: (Score:3)
I had the impression from the summary that the alcohol case was the legal limit rather then pissed, so basically one or two drinks.
My scariest driving has been when overtired, and its like drinking, you don't realize how impaired you are.
Knobs and buttons (Score:5, Insightful)
Touchscreens suck in the car because the driver *must* take their eyes off of the road for several seconds in order to use them. The trend of removing all mechanical knobs and buttons and replacing them with soft buttons on a touchscreen is a safety nightmare. Give me my old-school volume knob -- or even better, an encoder knob that does the same thing digitally!
And oh yeah, CarPlay blows too.
Re: (Score:2)
Re: (Score:1)
3rd Party infotainment sys block OEM voice control (Score:3)
Once I activate CarPlay (and it is really convenient to use it for Waze and other apps) I lose all access to native functionality. Anything I then want to do natively either requires that I manually do it (removing my eyes from the road) or that I disable CarPlay (again manually taking my eyes off the road.) This makes it inherently less safe.
I don't understand why there isn't a simple, steering wheel button controlled process for talking to the underlying OEM system.
cannabis (Score:2)
Re: (Score:2)
I don't know about that, but I do know that it tends to make drivers less pushy. At least, it does me. I'm more patient when using it.
I wouldn't drive under the influence right now because I'm no longer a frequent user, but when I was I was a much better driver overall because of the mellow factor. I was also slightly less likely to overanticipate a turn, although I think I've mostly trained that out now.
Re: (Score:2)
Re: (Score:2)
Just saying something ridiculous, you mean. Cannabis isn't alcohol. The two are substantively different in a variety of ways. Both can impair your judgement, but alcohol makes you unaware that your judgement is impaired. That's why people drive way too fast on alcohol, and way too slow on cannabis.
Bogus research (Score:3, Interesting)
This is bogus research because it assumes that drivers are stupid and inexperienced. Any experienced driver would have a sense of the current situation, including having a feel for the likelihood that a situation requires extra attention, such as non-straight roads, other cars or objects in close proximity, approaching traffic signals, unfamiliar routes, etc. In these situations, the experienced driver would not only block out all interaction with the navigation or entertainment system, they would also give their undivided attention to driving. That might mean pausing a conversation with passengers, stopping listening to the radio, and even avoiding looking at the speedometer. The experienced driver would sense that in situations where a quick reaction time may be needed, actions that may be okay in less stressful situations should be avoided.
The research basically found that if you do stupid things at inopportune times, bad things happen. Just because a researcher forces a driver to act stupidly in a restricted situation doesn't establish the inherent undesirability of the target action in all situations.
Re: (Score:3)
You never know when something is going to happen that you have to react to.
Re: (Score:2)
You never know when something is going to happen that you have to react to.
True. The experienced driver can't know with 100% surety that something weird won't happen. However, the probability of a safety-critical event happening when the experienced driver lets down his guard is very low, and it's the product of this probability and the occurrence rate of safety-critical events that largely determines the accident rate.
In practice, knowing when to be on heightened alert and when to relax are key skills for the experienced driver because staying on heightened alert for extended p
Re: (Score:2)
Even when relaxed, nice straight stretch etc, it still seems like a good idea to keep my eyes on the road excepting glances at mirrors, speedometer and such rather then paying attention to a touch screen.
Re: (Score:2)
Even when relaxed, nice straight stretch etc, it still seems like a good idea to keep my eyes on the road excepting glances at mirrors, speedometer and such rather then paying attention to a touch screen.
There's a continuum of ostensibly risky behaviors that are less risky due to the current driving situation. I do research into self-driving car safety as a job, and my research has shown the significance of the scenario in affecting conditional accident probabilities. That is, if there is nothing to hit, then accidents are not likely. This is sort of obvious, and it's the basic intuition that all drivers have. However, experienced drivers are better able to discern truly low-risk scenarios.
Re: (Score:2)
I've had too many experiences of deer, bears and even dogs unexpectedly running in front of me on relaxing parts of my drive to feel comfortable not watching the road. Hitting a bear at 80 km/h could be fatal, especially to the bear.
Re: (Score:2)
I've had too many experiences of deer, bears and even dogs unexpectedly running in front of me on relaxing parts of my drive to feel comfortable not watching the road. Hitting a bear at 80 km/h could be fatal, especially to the bear.
In 46 years of driving (and zero accidents with vehicles), I've hit a deer, dog and coyote, along with a couple birds. In every one of those incidents there was little to nothing I could have done to prevent it, short of driving off the road, and likely causing myself and passengers injury. Each one ran/flew right in front of my vehicle, from places that hid them from view. That's not going to happen on an open highway in Nebraska (where I lived for a couple years).
Again, it's about risk assessment. If
Re: (Score:2)
So, while I will agree with your basic statement, I'll disagree that the risk factor is equal. Driving down an empty open rural highway on a sunny day holds extremely little risk when compared to rush hour in icy conditions in an urban setting. In the former, I'll gladly hold conversations with passengers, fiddle with the radio, nav, etc. and look around at the scenery. In the latter, I shut up, and have both hands locked on the wheel with full attention.
We all take risks every day, and it's not that shi
Re:Bogus research (Score:5, Interesting)
This is bogus research because it assumes that drivers are stupid and inexperienced.
Actually that is a very good assumption to make.
Most drivers are horribly incompetent and unaware of it to begin with. Adding in distractions that take their limited attention away from the task of driving makes them even more unsafe.
Re: (Score:2)
This is bogus research because it assumes that drivers are stupid and inexperienced.
Actually that is a very good assumption to make.
Most drivers are horribly incompetent and unaware of it to begin with. Adding in distractions that take their limited attention away from the task of driving makes them even more unsafe.
The assumption that stupid and inexperienced drivers are easily distracted is obvious and is a dubious research topic.
Are most drivers horribly incompetent? Now that would be a great research question! It's not at all obvious to me what the distribution of driving experience would be among real-life drivers. Insurance companies already assume that driving skill increases with driving experience and has a significant knee for drivers 25 years and older. It would be very interesting to test this hypothesi
Actually most accidents happen near home (Score:2)
Most people are stupid for limited periods of time, like when they are drunk for example... but also when multitasking (divided working memory and time limited focus shifting.)
When you drive your usual route near your home you let your guard down and run on a human form of autopilot, you also think about and do other things shifting your focus away because of the routine habit trained into your brain. It makes you more stupid (like any other trained animal) because your smart brain evolved to optimize how m
Misleading headline is misleading (Score:2)
Android Auto: 30% increase, 53% increase 0.53 meter deviation.
Given the ridiculously small sample size, these are the same numbers between the two systems. But the headline only calls out Apple.
Don't make me sic the Macalope onto you.
Task selection (Score:2)
Task selection is key to this study. Which ones are people likely to actually do while driving?
- Select music on Spotify? Yes. But on screen while doing a figure 8? Probably not.
- Input destination data? Maybe by voice, but certainly not on screen. If you can say "Navigate Home" I don't see a problem. But if you've got to search for a restaurant and choose from among many choices on the screen, and you try this while driving, you really are an idiot.
- On screen red car? Really? This part of the test is high