Driven to Distraction

Is the spread of cellphones, navigation systems, and other auto-based gadgets endangering motorists? We try our luck in a driving simulator–and crash.


Automotive Editor Dan McCosh climbs into Ford's driving simulator to test how on-board electronics can become a hazard. McCosh quickly and inadvertently goes off-roading as engineers record his eye movements on videotape for later analysis.Rachel Holland

My passenger in the Ford Taurus I'm driving suggests we skip through Bob Dylan's "Positively 4th Street." It's not my favorite song either. I start to fiddle with the CD player in the instrument panel. Then, out of the corner of my eye, I see that I'm running off the road into a ditch at 60 mph.

There are no screams of alarm. In fact, my passenger, Jeff Greenberg, a group leader at Ford's science labs, grins. No one is getting hurt here. We're sitting in Ford's new $10 million simulator, designed to study how people act while driving. Think of it as a big video game and you've got the idea. Ford hopes the simulator will bring scientific data to the nationwide debate over how distracting cellphones and other electronic devices are for drivers.

"How long do you think you can drive straight without moving the wheel, before the car leaves the lane?" asks Greenberg, bemusedly.

"About 4 seconds?" I guess.

"About 4 to 6," confirms Greenberg, promptly.

"Sloppy steering," I comment, lamely trying to shift the blame for the crash to a problem with the car. When in doubt, always blame the equipment.

"We're working on the steering," Greenberg says with a slight frown. The new simulator is realistic, nonetheless, particularly in the way it responds to minor steering and braking. During the virtual drive on I-94 between Kalamazoo and Chicago, the sensation of a slightly irregular road surface is compelling, the tar strips bumping a rhythm as the miles reel off. It's hypnotic, and you have to concentrate to keep the car in its lane.

Driver distraction gained national attention when a car crash critically injured supermodel Niki Taylor early this year. The accident was attributed to a friend who, while driving, reached to answer a cellphone and subsequently lost control of the car. The incident intensified the debate over cellphone use by motorists. Many states have passed or have pending legislation that would prohibit drivers from talking on handheld cellphones. Accessories that allow drivers to talk on the phone without holding it are becoming increasingly common and may eventually be mandatory.

Lots of things distract from driving -- putting on makeup, eating lunch, glancing at newspaper headlines. But operating the new generation of electronic devices that automakers are installing in cars is something different.

Cellphones are just the beginning. Navigation systems and e-mail are already available. Audio systems are becoming more complicated to operate. And loads more potential distractions are making their way into cars -- from customized information services that shower drivers with stock quotes, news items, and shopping tips, to TVs in the back seat.

The question is: When does a moment of inattention become a hazard to the driver and others on the road?

Automakers point out that they can't keep motorists from acting irresponsibly; all they can do is try to make sure the design of automobiles doesn't contribute to the distraction problem. "We can't control all the things people attempt to do while driving," says Greenberg. "But we can determine whether the equipment we install is safe to operate."

Even before Taylor's accident, the National Highway Traffic Safety Administration (NHTSA) had begun taking testimony on the subject of driver distraction as a prelude to federal regulation. But despite the attention being paid to inattention, much of the debate so far has been anecdotal rather than scientific. In fact, there is surprisingly little hard data about what people are looking at while they drive. The Ford simulator, as well as another being built by NHTSA, is among the first of several new machines that are designed to compile that information.

The Taurus I'm virtually driving is very real, but the car is bolted solidly to the flat floor of a spherical pod. The roof overhead is about the size of a military early-warning radome and is made from the same material. Eight huge hydraulic jacks that support the pod some 20 feet in the air generate the sensation of motion. The hydraulics are quick acting, making short, fast, lateral moves, then tilting the pod smoothly. The occupants inside feel the force of gravity simulating acceleration, braking, and turning. To observers on the outside, the pod's gyrations make it look like some giant insect learning the chicken dance. But seated inside, with a computerized view of the road coming at you and the bobbing dome contributing the physical sensations, it feels like you are driving a real car on a real road.

We are not alone. In the next room, technicians are watching several views of the interior on a bank of monitors. One camera shows them your feet. Screwed down to the instrument panel is another tiny camera, this one taking digital images of your iris. Low-power infrared lasers also focus on your eyes. Another electronic array worn on your head completes the complex system that has a single purpose -- determining where your eyes are focused at any given moment.

Documenting exactly where your eyes are looking is a key part of the studies being conducted here. The time spent gazing ahead is compared with the time spent looking elsewhere. The machine notes that it takes about one-and-a-half seconds to glance at your speedometer, for example -- during which you have traveled about 150 feet at 60 mph.

Greenberg calls such events "glance time," and they are surprisingly consistent in at least some respects.

Most drivers are generally reluctant to take their eyes off the road for more than a few seconds, but individuals differ in their habits. Some studies, moreover, have shown that the tolerance for taking your eyes off the road diminishes with age. Young males, for example, seem to be more apt to look away. Greenberg once recorded an individual who "drove" for more than a hundred yards without looking at the road.

Many tasks -- such as punching a single button on the radio or checking the speedometer -- can be accomplished with a single glance; more complex tasks require multiple looks.

The machine also documents the total time per task; newer technologies are generally more distracting. Dialing a cellphone, for example, takes about three times as long as punching a radio. Computer displays that require selecting from a menu sometimes accumulate 15 seconds or so worth of glances.

Ford and other auto companies are using such data to develop engineering standards for equipment installed in their cars. A 15-second limit for the total time needed to operate a device was proposed as a universal standard, but hasn't been adopted.

But the focus on time spent looking away from the road begs the bigger issue: Can people safely accomplish other tasks while driving, period? Some recent scientific research suggests that the brain has built-in limits when it comes to performing multiple activities. The research seems to indicate that you really can't walk and chew gum at the same time. At least, not do each task quite as well as one done alone.

One such study, conducted at the University of Utah and released in August, tested reaction times in a simulator while drivers were operating various devices and talking on cellphones, both with and without hands-free options. The results seemed to indicate that phone conversation interfered with reaction times and attention, even when the driver did not have to dial the number. Another recent study, at Carnegie Mellon University, examined brain activity while people were attempting to multitask. The study did not specifically involve cellphones or driving, but the results showed reduced brain activity in the area associated with each task when two were attempted at once.

An accident study by NHTSA indicates that driver distraction causes as many as 20 percent of accidents. But the study takes a very broad definition of distraction -- most of the "distractions" consisted of something happening at the roadside rather than inside the vehicle. "We are just beginning to study what happens when we add cognitive tasks (to driving)," says Greenberg.

The research has implications far beyond studying the potential for onboard devices to distract. One is that the basic perceptual mechanisms at work can be harnessed to design a whole new, and safer, way of driving. A new array of sensors has already been developed that provides the driver with potentially lifesaving real-time information. Examples are night vision, rearward-warning radar, and traffic-monitoring devices that calculate the speed and actions of each car approaching an intersection. Some sensors, such as distance-controlling, radar-activated brakes, intervene automatically. Others, however, require the driver's attention.

"The question is, how do we get this information to the driver in a way that (lets him) act appropriately?" Greenberg says. It's not a simple issue. A warning beep? If so, how loud, and at what frequency? What effect will an automatic braking application have on a driver? Driving simulators that consider how sudden sights and sounds affect a driver's reactions are designed to answer these questions. They should yield a system of alerts that most drivers react to intuitively.

Meanwhile, the immediate reaction to the issue of driver distraction is to reduce or eliminate activities that induce motorists to take their eyes off the road or their hands off the wheel. Voice activation is already available, not only for cellphones but also for other activities such as changing radio stations and operating air-conditioning. A voice recognition system that makes e-mail reasonably safe and practical has yet to be introduced, though GM claims it has taken steps toward such a system with its OnStar Virtual Advisor, which presents most of its information, including e-mail, audibly, over a phone line. OnStar relies on a huge phone bank to deliver its information, which has led to a study of whether any OnStar subscribers have suffered accidents while online. A review of some 8 million phone calls found only two accidents severe enough to deploy an airbag.

Still, the proliferation of onboard electronic systems continues to cause concern. The installation of such systems, whether approved by the manufacturer or not, is outstripping the research that studies how they are used -- research that ideally will provide for the sound, safe use of electronics in the cars of the future.

One of the best ways to keep your eyes on the road and your hands on the wheel is to invest in a hands-free cellphone for the car, an option that many states may soon require. But there are a few ways to go about it. Here's what you need to know:

Many luxury car companies offer built-in hands-free cellphones that use the stereo's speakers to provide excellent sound quality.

You can get a cellphone like Nokia's Cark-91 for about $175 plus installation, and have it retrofitted into any car. This integrated set-up enlists the car's antenna to boost reception.

A more portable option is a speaker/microphone cradle for your existing cellphone that plugs into the car's cigarette lighter. One model from Sharper Image lists for $129.

Some newer cellphones with voice activation and speakerphone capabilities can be used in and out of the car. Some models simply clip to your car's sun visor for easy use. One good example is Motorola's Timeport 270c, which lists for $349.

Driving Data

There were 6,393,409 vehicle accidents in 2000.

20% of those accidents were caused by distracted driving.

20% of drivers report steering with their legs.

32% of drivers read and write while driving.

16% of the arguing done while driving causes an accident or near miss.

13% of the kissing done while driving causes an accident or near miss.