Infotainment too complex, says one study; safety features really work, says another.
So it’s good to know that advanced driver-assistance systems (ADAS), like blind spot monitoring and lane departure warnings, are saving lives, according to the findings of another study, this time from the Insurance Institute for Highway Safety (IIHS). The National Highway Traffic Safety Administration just announced, however, that the number of road deaths in the US has increased for the second year running.
The first study was conducted by David Strayer and colleagues at the University of Utah and is part of a long-running research program funded by the AAA Foundation on distracted driving. The report, which was published this week, sought to answer several questions about current infotainment systems, which it notes are getting ever more complex. Among these were which types of tasks are the most distracting and if it’s more work to interact with controls in the center of the dashboard (either buttons or a touchscreen, the study did not discriminate), controls between the front seats, or voice commands.
Each of the 30 different vehicles was driven by 24 different research participants, each of whom had to perform four different tasks (audio entertainment, calling and dialing someone on the phone, text messaging, and navigation) while driving on a residential road. The first two of these were equivalent in cognitive workload, but texting required much more attention, and entering a new navigation destination was the most demanding of all.
Unsurprisingly, the study found a great deal of variability between different vehicles and even between different models from the same manufacturer. As for which mode of interaction was the most distracting, if you guessed “poking a screen or buttons on the center of the dashboard,” it will surprise you to learn this was actually the least demanding. Fiddling with buttons on the center console (the bit between the driver and front passenger where you also usually find a gear lever) was the most demanding way of controlling an infotainment system, though only four of the vehicles (an Audi Q7, Infiniti Q50, Mazda 3, and Nissan Maxima) actually had their infotainment controls located there.
Interestingly, voice commands kept drivers’ eyes on the road but kept them distracted for longer, offsetting any benefits, thanks to much longer interaction times. If that isn’t proof that more OEMs need to copy BMW’s voice command system (which lets you interrupt), then I don’t know what is.
There are a few caveats to this work, most notably the participants’ lack of familiarity with each vehicle and its particular infotainment UI. Although they were given a brief acclimation period in each vehicle before being tested, I’m not sure they’d find the same levels of cognitive or visual demand associated with some tasks after a more lengthy period of time to develop muscle memory or find effective shortcuts. For instance, several of the functions tested can usually be controlled via a multifunction steering wheel that keeps the driver’s hands on the wheel (usually in combination with the main instrument display).
As infotainment systems aren’t going away any time soon, the report recommends that automakers think long and hard about whether all their infotainment features really need to be available while in motion. It notes that NHTSA guidelines recommend not letting people text, browse the Internet or social media, or enter navigation addresses while they are driving, nor should cars display long messages unrelated to driving. “Greater consideration should be given to what interactions should be available to the driver when the vehicle is in motion rather than to what features and functions could be available to motorists. With the best intentions, we will put some technology in the car that we think will make the car safer, but people being people will use that technology in ways that we don’t anticipate,” Strayer said.
If you do plan to enter a lot of addresses from a moving car, at least try to do it in one with some good, modern safety features. IIHS’ study found that lane departure warnings reduced the number “of single-vehicle, sideswipe, and head-on crashes of all severities by 11 percent” and that injures in those kinds of crashes were down 21 percent. A less rigorous analysis of those types of crashes (which didn’t control for driver demographics because of the smaller sample size) showed an 86-percent decrease in fatalities when lane departure warning was fitted.
“This is the first evidence that lane departure warning is working to prevent crashes of passenger vehicles on US roads. Given the large number of fatal crashes that involve unintentional lane departures, technology aimed at preventing them has the potential to save a lot of lives,” said Jessica Cicchino, IIHS vice president for research. Interestingly, the study notes that these findings are more modest than a pair of earlier works that found lane departure reduced crashes by nearly 50 percent in US truck fleets and injuries in Swedish Volvos by 53 percent. One possible reason for this is that many people disable the system, which nags you when you veer out of your lane.
The study also notes that lane departure warnings then require action by the driver, and more than 30 percent of lane-drift crashes occurred when drivers were incapacitated. That’s reason enough to want some form of lane-keeping assist, and indeed, the NHTSA investigation into a fatal crash in 2016 found that Tesla’s crash rate dropped by 40 percent when it added that function to Autopilot. (However, it also added automatic emergency braking at the same time, and it’s difficult to tease out the relative contributions of different ADASes.)
Blind spot monitoring was similarly useful, reducing the number of lane-change crashes by 14 percent. If every vehicle on the road was fitted with such systems, we could avoid around 50,000 police-reported crashes each year, according to the study.
Unfortunately that isn’t the case, and on Friday, the NHTSA published an overview of 2016’s fatal vehicle crashes that makes for depressing reading. For the second time in two years, the death toll on US roads has gone up. In 2016, 37,461 people died in vehicle crashes, up 5.6 percent from 2015. Perhaps the sole comforting fact is that the rise in death toll wasn’t quite as great as the delta between 2014 and 2015. There were more passenger vehicle occupant deaths and more motorcycle deaths than at any time since 2008. Pedestrian deaths haven’t been this high since 1990 and cyclist deaths since 1991. While there was a small increase in the number of vehicle miles traveled, the fatality rate increased to 1.18 per 100 million vehicle miles traveled.
People: take care out there.