"it does advise accelerated standardization and introduction of features including automatic emergency warning, blind spot monitoring and lane departure warning systems"
If these things actually made a difference wouldn't vehicle deaths be falling unless some other change is offsetting the difference?
I'd be surprised if automatic emergency warnings didn't increase fatalities. The absolute last thing I want my car to do in an emergency driving situation is add an additional distraction.
The idea is they let you know when an emergency situation is occurring. Otherwise, particularly if you're distracted, the first indication you may have that you have left your lane is when you strike another vehicle.
I just bought my first new car in over 14 years and loaded it up with all the safety features: lane keep assist, advance collision detection and braking, blind spot detection.
I've been driving for 30 years and haven't had an accident in over 25, and I can tell you: I love these features. The blind spot detection mostly offsets the poor visibility on the new car, but the other features are really helpful.
unless some other change is offsetting the difference?
FTA:
"Many studies and observers cite the rise of mobile devices and their growing role in our lives as a possible cause behind the recent spike in traffic fatalities."
Anecdotally, in my area, teens often receive the current family car (which is often 8+ years old), or one of the families secondary vehicles. Then another vehicle is purchased for the parents, often new or a recent used vehicle. So the vehicle with less up-to-date safety options goes to the family member that is the least experienced driver.
But that's always been the case, and today's 8 year old car is safer then it was 10 or 20 years ago, so I wouldn't think this was causing an upward trend.
That is a concerning recommendation. How would this be achieved? GPS tracking? Speed limit cameras? Both methods can have serious privacy implications.
Automated enforcement of speed limit violations would be fine, as long as our speed limits were adjusted for strict enforcement versus human enforcement.
Almost everyone will drive 35-40 in a 30, but you could make a case that anyone breaking 45-50mph for more than a few seconds should automatically get a ticket. Even better, the automated enforcement could account for actual danger, using road conditions, traffic density, and pedestrians: you can do 15+ if the road is empty, but you had better slow way down for a stroller in the crosswalk, etc.
Every speeder ever thinks they can speed until, well, they can't. As air drag increases quadratically with speed so does kinetic energy and pedestrian survival rates in a collision rapidly approach 0% for speeds above even 25 mph.
So no, widespread violation of speed limits is not in fact an argument towards allowing the violation of speed limits. Most countries are already much more lenient here than physical reality allows for.
I think you're talking about something else, because we should be designing for pedestrians to not be hit by cars, not making sure that pedestrians might survive the impact. At the same time, limiting modern cars to 25 mph everywhere is ridiculous.
In some high density locations, that might mean strictly enforcing a 15mph limit instead of loosely enforcing a 10mph limit, building bike lanes, public transport, pedestrian bridges, and guarded crosswalks, etc.
I'm sure we can do both. It was you who suggested being lenient on people going 35-40 mph in a 30 zone; 30 suggests the likelihood of encountering pedestrians, hence the mention of the survival rate.
I'm quite with you on what we can do in high density locations, but I think you have not quite thought it through. In high density locations, there should simply not be any cars. The huge compromises are not justified where pedestrians outnumber them 5:1 anyways.
Let's be clear that what you're saying is, "In high density locations, there shouldn't be anyone who won't pay the rent to live within walking distance."
This is a great reason not to speed in residential or downtown areas where there are pedestrians around.
We're most likely to get speed cameras on major roads and freeways, where pedestrians are rare to nonexistent and have no expectation of being yielded to, as you might with a crosswalk on a 25mph road.
You do not have an expectation of privacy while operating a vehicle upon the public highway. That is why your car has a giant number plate on it, why you must carry and present a license to operate said vehicle, and why you provide implied consent for a search of /your own blood/ by driving one.
Most US law enforcement already routinely [1] use automatic number plate recognition (ANPR) with no oversight whatsoever. That is a much worse situation than speed cameras in most countries, where a picture is only taken if you did in fact exceed the limit.
also conducted a survey that indicates some possible causes for the steady increases, including speed, texting via touch input or voice, as well as driving while under the influence.
I think most of it can be attributed to cellphone usage while driving.
And now that most drivers choose to use their smartphones as GPS rather than a dedicated GPS device, I think it's going to get much worse.
I think GPS is secondary: you can easily get a mount for it.
Based on experience, the most likely explanation is texting or voice calls. When I got a new car with bluetooth calling, all of a sudden I was fumbling a lot less. A call is now about the same level of distraction as having a passenger or 2.
Texting however, I have no idea. Maybe something that moves texting to voice, but I haven't seen anything actually make that work. And, it's distracting enough that I just don't text while driving: sending ETA updates or w/e isn't worth risking my life over.
> When I got a new car with bluetooth calling, all of a sudden I was fumbling a lot less. A call is now about the same level of distraction as having a passenger or 2.
Most studies that have actually tested it have found that talking with someone on a phone, even if hands free, is much more distracting than talking to a passenger.
A big factor is that a passenger is much more likely to be aware of road and traffic conditions and stop the conversation during situations where the driver needs to pay extra attention to driving.
Sounds plausible that it's more distracting, but I'm not convinced that the difference is more varied than found among different types of passengers (back-seat driver/kids/dumb friends/audiobook(?) etc)
If you have a few studies that show more than that, I would be glad to check it out.
I don't know of any that have looked at different types of passengers. They have usually just compared an adult passenger talking with the driver to an adult on a phone talking with the driver.
I think some of it must be an unintended consequence of usage bans. People still do it, they just keep the phone out of sight, and this keeps their concentration off the road for longer. (And of course you can't type by feel on a virtual keyboard.)
This. I text while driving very rarely, but when it was legal I would with my arm straight ahead so I could see the road and my phone. My phone's now been relegated to below the dash.
But it's not out of sight! Perhaps you don't realize how obvious it is! If you're in an SUV, you can easily see the phones on people's laps. If you're in a car, you can still tell because people don't normally stare at their crotch while at a red light. Even if they hold it higher, you can still tell they're not looking around them. ...
Finally if all else fails, at night you can see the reflection of light from their phone.
In LA I would guess something like 75% of drivers use their phones at red lights. It makes the delay of cars starting to move when a light turns green super annoying.
Smart phone penetration in the US has been above 75% for many years now, so you'd have to explain why there's a jump in fatalities now but wasn't as smart phone penetration went from 0% to 75%.
Secondly, this data should be controlled for miles driven. Otherwise a growing population and a growing economy can be misconstrued as roadways becoming more dangerous.
Anecdotally there are a lot more people glued to their screens than there used to be. I think one factor is younger people who become drivers after already have been addicted to their smart phones.
It used to be that you could tell from a distance who was talking on their phone while driving. They would drift off their lane and correct randomly. At least while talking you have some % of your attention on driving, if low. While texting or interacting with the device's screen you have just about 0 attention left for other tasks. And you often don't even realize it...
There must be some data specific to distracted driving you'd think...
> I think most of it can be attributed to cellphone usage while driving.
This. My car insurance went up 25% recently b/c of this trend. That's right, OTHER PEOPLES behavior is costing me money. I guess thats how insurance works though right. I mean the insurance companies need to be guaranteed to make money right???
You might check out New Hampshire. I don't have a reference handy, but I recall reading that they don't require insurance yet have more people insured than most other states.
That only helps as long as there aren't many insurance companies. If there are many insurance companies, competition between them drives the costs down to marginal cost plus normal profit, in theory.
Honestly let's not refer to the problem as cellphone usage, but touchscreen usage. I can't fathom how builtin touchscreens on center consoles are even blessed by the DOT (especially with the horrendous input lag "modern" implementations generally have), but for their being "hip".
My 2016 Mazda disables the touchscreen when the car is in motion. I thought perhaps that was a NHTSA rule, but given cars like the Tesla S (where the screen operates everything not just the IVI) I guess it's not a firm requirement.
Driving while distracted is a huge problem. Denial of that problem by people who have it is in itself also a huge problem. Apparently current levels of laws and penalties aren't sufficient to deter it, so something stronger may be needed.
Awful controls and interfaces in cars are also a problem. The number of gadgets and displays and distractions of all sorts in modern vehicles is disturbing. Laggy touchscreens are a prime example. Speak-out-loud systems and voice activated systems with poor speech recognition are other weak areas. I find it amazing that when cars are probably the second most expensive thing most of us ever buy, we accept such crude and poorly designed control systems, but there seems to be little concern in the industry because they're all just about as bad as each other. Again, at some point stronger regulation may be needed.
Tacking stronger legislation onto a social problem is why the War on Drugs exists, which hurts more than it helps. I'm not saying we shouldn't add stronger legislation, but maybe we should gather data on whether or not it would be more effective than an education & awareness campaign.
I'm all for evidence-based policy-making, so I would have no objection in principle to considering alternatives like a public education campaign.
That said, in this particular case, we had plenty of campaigning when existing laws were introduced here in the UK, and that doesn't seem to have been very effective. The near-total lack of enforcement means those who find using their phone while driving convenient continue to do so, despite the mountain of evidence about how dangerous it is. Worse, because we didn't also ban hands-free use, despite the mountain of evidence that it is almost as bad as hand-held phone use in terms of impairing driver performance, vendors of hands-free kits were literally putting up six foot high stands at the entrances to major stores the week after the ban came in with captions like "Stay safe -- go hands-free!" all over them.
If we're going to solve this one, I suspect what we need is a combination of education and enforcement, both much stronger than we have seen so far. The number of people I come across online who maintain that they drive while using a phone all the time and are still perfectly safe is genuinely alarming.
Oh definitely. But my point is touchscreens add to that distraction, in an imperceptible way.
Texting while driving wasn't such an epidemic with T9, which had much societal adoption. I admit the sheer volume has gone up since then, but personally I recall it being relatively easy to enter a message only having to look at the screen occasionally to verify input.
A touchscreen requires one to look at the screen to make input. This insidiously creeps into one's OODA loop, sapping attention in a way that felt much harder to be aware of (since the "atomic task" of entering a letter now has a variable delay).
Likewise, I've done two cross country round trips with a handheld GPS unit (Garmin 60CSx), which my right hand operated just fine, including shifting. I wasn't using it for routing, but a "you are here map" with crow-flies distance. Most critical use was in denser cityish areas, that require more shifting and navigating traffic, and I still never felt distracted. Meanwhile even just starting Waze seems like a massive distraction that I'd rather only do while stopped - nevermind navigating any of its modal dialogs.
I haven't found a link to less generalized data. What were road/traffic/driver conditions in the fatal accidents?
Were there more accidents, or were accidents themselves more fatal? What percentage of fatalities involved a semi? Happened on freeways?
What if climate change is the culprit? Road conditions have deterioriated as a result of changing weather patterns. Drivers have deteriorated as mobile devices become more central to our lives. I doubt that the 8% increase in fatalities is best explained by a 3% increase in miles traveled.
Unfortunately there isn't a standard way to collect this data on a national level. Believe it or not its up to that local jurisdiction, or State (in the case of state patrol) to dictate what information they gather and how it is gathered. This makes comparisons and real statistical analysis quite difficult.
There is no federal database of traffic fatalities? In terms of human resources we lose on the order of 150 billion (40000 <lives>*7,000,000 <$/life> /2 ) dollars a year in traffic fatalities. That's not including the lost work from injuries. That's like 1% of the GDP. You would think that this would be a good enough reason to at least compile a database of causes.
Even an incomplete one could be useful.
Wonder how much of this can be tied to the economy getting better and more people buying/driving (used and less safe) cars when they previously couldn't afford to.
That's pretty convincing to me -- more miles driven means more driving deaths. Also, it's not clear to me whether the statistic on motor vehicle deaths includes collateral damage: cyclists and pedestrians killed by vehicles. Either way, yikes, cars kill lots of people.
Also, how much ridership do Zipcar and Uber take away from public transportation since they mostly operate in cities? Uber goes further in that it creates additional driver-miles while drivers are getting to or waiting for passengers.
The last recession in the US ended officially ended in 2009. I know the recovery has arguably been sluggish and uneven, but it doesn't seem like the slow & steadily rising GDP over an eight year span should explain much of this shorter-term jump in vehicular deaths.
Does anyone know if (and where) it's possible to download a data set of all car accidents (or just fatals) for the last few years? I'd like to see things like date and time, weather conditions, lat/lon, make and model of cars involved, etc.
Overall these data sets have been pretty well studied for the past few decades. The results are well documented: e.g. 6pm to midnight is the deadliest time, young males are the riskiest, older drivers are also risky, etc. Weather can actually be good for safety because people often drive slower and there are fewer cars on the road in bad conditions. But yes, weather can be a factor too.
It's kind of funny how the article is using a picture of Toronto for a US topic. Also, I wonder if Pokémon GO can be blamed for some of this increase in motor vehicle deaths.
Need stricter tests for getting a license. There's people up here in Portland that just change lanes whenever they feel like it. It's like people just don't care enough to drive safely. If you leave a big enough gap someone will just race in to take it.
Along with the absolutely horrific case where the driver uses a smart phone WHILE driving, from my personal observation, another component is the one where a lot of drivers will immediately hop on their devices as soon as they reach a temporary stop, be it at a street light or an intersection.
This creates a lot of delays and confusion in traffic due to the fact the driver is actively incapacitated in participating in traffic.
In this environment, having a car that has active collision control/prevention mechanisms is more and more a necessity.
Watch a busy street at night during heavy traffic. You can see from the light shining off driver's faces that 40%~ of them are either using their phones or another system while driving.
Given the proliferation of front-facing cameras and velocity sensors in all iPhones and most Android devices, it is unconscionable that these devices remain usable in these circumstances.
How can you write an article about this without any discussion of the trend in automobile accidents (including nonfatal ones) as well as the trend in miles driven per year? Counting traffic deaths without any explanatory context simply raises a lot of questions to which I don't have time to research the answers.
80 comments
[ 4.0 ms ] story [ 186 ms ] threadIf these things actually made a difference wouldn't vehicle deaths be falling unless some other change is offsetting the difference?
I've been driving for 30 years and haven't had an accident in over 25, and I can tell you: I love these features. The blind spot detection mostly offsets the poor visibility on the new car, but the other features are really helpful.
FTA:
"Many studies and observers cite the rise of mobile devices and their growing role in our lives as a possible cause behind the recent spike in traffic fatalities."
Anecdotally, in my area, teens often receive the current family car (which is often 8+ years old), or one of the families secondary vehicles. Then another vehicle is purchased for the parents, often new or a recent used vehicle. So the vehicle with less up-to-date safety options goes to the family member that is the least experienced driver.
Today's car may have collision avoiding systems, which could protect pedestrians.
That is a concerning recommendation. How would this be achieved? GPS tracking? Speed limit cameras? Both methods can have serious privacy implications.
Almost everyone will drive 35-40 in a 30, but you could make a case that anyone breaking 45-50mph for more than a few seconds should automatically get a ticket. Even better, the automated enforcement could account for actual danger, using road conditions, traffic density, and pedestrians: you can do 15+ if the road is empty, but you had better slow way down for a stroller in the crosswalk, etc.
So no, widespread violation of speed limits is not in fact an argument towards allowing the violation of speed limits. Most countries are already much more lenient here than physical reality allows for.
In some high density locations, that might mean strictly enforcing a 15mph limit instead of loosely enforcing a 10mph limit, building bike lanes, public transport, pedestrian bridges, and guarded crosswalks, etc.
I'm quite with you on what we can do in high density locations, but I think you have not quite thought it through. In high density locations, there should simply not be any cars. The huge compromises are not justified where pedestrians outnumber them 5:1 anyways.
I used to, until I learned this [0]:
> If someone is hit by a car at 40 mph they are 90% likely to be killed.
> If someone is hit by a car at 30 mph they are 50% likely to be killed.
> If someone is hit by a car at 20 mph they are 10% likely to be killed.
[0]http://www.safespeed.org.uk/killspeed.html
We're most likely to get speed cameras on major roads and freeways, where pedestrians are rare to nonexistent and have no expectation of being yielded to, as you might with a crosswalk on a 25mph road.
1: https://en.wikipedia.org/wiki/Automatic_number_plate_recogni...
I think most of it can be attributed to cellphone usage while driving.
And now that most drivers choose to use their smartphones as GPS rather than a dedicated GPS device, I think it's going to get much worse.
Based on experience, the most likely explanation is texting or voice calls. When I got a new car with bluetooth calling, all of a sudden I was fumbling a lot less. A call is now about the same level of distraction as having a passenger or 2.
Texting however, I have no idea. Maybe something that moves texting to voice, but I haven't seen anything actually make that work. And, it's distracting enough that I just don't text while driving: sending ETA updates or w/e isn't worth risking my life over.
Most studies that have actually tested it have found that talking with someone on a phone, even if hands free, is much more distracting than talking to a passenger.
A big factor is that a passenger is much more likely to be aware of road and traffic conditions and stop the conversation during situations where the driver needs to pay extra attention to driving.
If you have a few studies that show more than that, I would be glad to check it out.
Here was a recent study that was pretty interesting: https://news.illinois.edu/blog/view/6367/204502
Here's an older one: https://www.sciencedaily.com/releases/2008/12/081201081917.h...
You have my permission to use this as your next start-up idea.
Finally if all else fails, at night you can see the reflection of light from their phone.
In LA I would guess something like 75% of drivers use their phones at red lights. It makes the delay of cars starting to move when a light turns green super annoying.
Smart phone penetration in the US has been above 75% for many years now, so you'd have to explain why there's a jump in fatalities now but wasn't as smart phone penetration went from 0% to 75%.
Secondly, this data should be controlled for miles driven. Otherwise a growing population and a growing economy can be misconstrued as roadways becoming more dangerous.
https://en.wikipedia.org/wiki/List_of_motor_vehicle_deaths_i...
It used to be that you could tell from a distance who was talking on their phone while driving. They would drift off their lane and correct randomly. At least while talking you have some % of your attention on driving, if low. While texting or interacting with the device's screen you have just about 0 attention left for other tasks. And you often don't even realize it...
There must be some data specific to distracted driving you'd think...
EDIT: Some data:
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2951952/
http://www.fleetfinancials.com/article/story/2011/09/5-year-...
http://www.cbc.ca/news/canada/british-columbia/distracted-dr...
This. My car insurance went up 25% recently b/c of this trend. That's right, OTHER PEOPLES behavior is costing me money. I guess thats how insurance works though right. I mean the insurance companies need to be guaranteed to make money right???
At least on average anyway. Or they will stop writing insurance policies.
It'd be more expensive for the people that did carry it, but it'd be a large business.
And of course, in many states there is the option to post a bond to prove ability to compensate.
Unfortunately Mazdas are the only economy cars that seem to care about this. For many others, it's a laggy touchscreen well below the vents.
Very happy with the old fashioned tactile controls on my 2012 VW.
Driving while distracted is a huge problem. Denial of that problem by people who have it is in itself also a huge problem. Apparently current levels of laws and penalties aren't sufficient to deter it, so something stronger may be needed.
Awful controls and interfaces in cars are also a problem. The number of gadgets and displays and distractions of all sorts in modern vehicles is disturbing. Laggy touchscreens are a prime example. Speak-out-loud systems and voice activated systems with poor speech recognition are other weak areas. I find it amazing that when cars are probably the second most expensive thing most of us ever buy, we accept such crude and poorly designed control systems, but there seems to be little concern in the industry because they're all just about as bad as each other. Again, at some point stronger regulation may be needed.
That said, in this particular case, we had plenty of campaigning when existing laws were introduced here in the UK, and that doesn't seem to have been very effective. The near-total lack of enforcement means those who find using their phone while driving convenient continue to do so, despite the mountain of evidence about how dangerous it is. Worse, because we didn't also ban hands-free use, despite the mountain of evidence that it is almost as bad as hand-held phone use in terms of impairing driver performance, vendors of hands-free kits were literally putting up six foot high stands at the entrances to major stores the week after the ban came in with captions like "Stay safe -- go hands-free!" all over them.
If we're going to solve this one, I suspect what we need is a combination of education and enforcement, both much stronger than we have seen so far. The number of people I come across online who maintain that they drive while using a phone all the time and are still perfectly safe is genuinely alarming.
Texting while driving wasn't such an epidemic with T9, which had much societal adoption. I admit the sheer volume has gone up since then, but personally I recall it being relatively easy to enter a message only having to look at the screen occasionally to verify input.
A touchscreen requires one to look at the screen to make input. This insidiously creeps into one's OODA loop, sapping attention in a way that felt much harder to be aware of (since the "atomic task" of entering a letter now has a variable delay).
Likewise, I've done two cross country round trips with a handheld GPS unit (Garmin 60CSx), which my right hand operated just fine, including shifting. I wasn't using it for routing, but a "you are here map" with crow-flies distance. Most critical use was in denser cityish areas, that require more shifting and navigating traffic, and I still never felt distracted. Meanwhile even just starting Waze seems like a massive distraction that I'd rather only do while stopped - nevermind navigating any of its modal dialogs.
Shows the per population and per mileage driven death rates have increased as well.
Were there more accidents, or were accidents themselves more fatal? What percentage of fatalities involved a semi? Happened on freeways?
What if climate change is the culprit? Road conditions have deterioriated as a result of changing weather patterns. Drivers have deteriorated as mobile devices become more central to our lives. I doubt that the 8% increase in fatalities is best explained by a 3% increase in miles traveled.
Source: I'm in the traffic safety industry.
Also, California has their own database that includes all crashes in the state down to crash/victim level: http://iswitrs.chp.ca.gov/Reports/jsp/userLogin.jsp (You have to create a free account)
There are other datasets too: http://opendata.stackexchange.com/questions/1755/automobile-...
https://fred.stlouisfed.org/series/M12MTVUSM227NFWA
If had to guess, I would attribute the rise to mobile devices distracting drivers - hence all the campaigns to stop texting+driving.
All California crashes, fatal to no injury, goes back a few years at least: http://iswitrs.chp.ca.gov/Reports/jsp/userLogin.jsp (You have to create a free account)
Overall these data sets have been pretty well studied for the past few decades. The results are well documented: e.g. 6pm to midnight is the deadliest time, young males are the riskiest, older drivers are also risky, etc. Weather can actually be good for safety because people often drive slower and there are fewer cars on the road in bad conditions. But yes, weather can be a factor too.
In this environment, having a car that has active collision control/prevention mechanisms is more and more a necessity.
Given the proliferation of front-facing cameras and velocity sensors in all iPhones and most Android devices, it is unconscionable that these devices remain usable in these circumstances.