Robert Gray uses an advanced driving simulator to study driver attention and perception. The simulator’s rearview mirrors show animated displays of what is “behind” the driver. The device is set on a moving base that makes the car rock when the driver accelerates, stops, or drives over a bump.
Home / Stories / Articles / Driven to distraction
by Diane Boudreau
October 01, 2009
Share This Article

Driven to distraction

On the whole, driving is pretty easy. Press the gas, turn the wheel, hit the brake. It’s so simple that most of us do it while mentally planning dinner, tuning the radio, yelling at kids in the back seat or even talking on a cell phone.

But driving is also very difficult, if you consider that car crashes kill about 40,000 Americans each year. Automobile accidents are the leading cause of death for children and young adults. How can something so easy be so deadly?

Ironically, it’s deadly because it’s so easy, according to Robert Gray, an applied psychologist at Arizona State University.

“Driving is hours and hours of boredom intermixed with moments of terror,” he says. “Most of it is not very demanding, especially if you’re on the highway. Then all of a sudden the car in front of you gets a flat tire, or a child runs out into the street. It’s really bad for human performance. People can’t be vigilant for a long, long time when nothing’s happening.”

Gray studies human attention and perception and how they affect driving. Attention is an important factor in driving safety. During those hours and hours of boredom, people get distracted. And distraction is a factor in 80 percent of all car crashes, according to a 2008 study from the National Highway Traffic Safety Administration and the Virginia Tech Transportation Institute.

Modern technology complicates the problem. Today’s cars come equipped with an alphabet soup of possible distractions. There are MP3 players and GPS systems and DVD players for passengers. And of course, don’t forget the ever-present cell phone.

Gray says that there are many ways to bring a distracted driver’s attention back to the road. Flashing lights on a police car, honking horns and “rumble strips” along the side of a highway all serve this purpose. But sometimes dangers appear without warning. When the car in front of you brakes suddenly, or a dog darts into the street, there are no flashing lights or sirens to capture your attention.

Some automakers are developing warning systems that notify drivers when a crash is imminent. A few high-end vehicles already come equipped with visual or auditory warning systems. But Gray says these aren’t the most effective ways to get a driver’s attention.

“Visual warnings are pretty much useless when you’re driving. You’re too visually engaged when you drive. You don’t notice them,” he says. “Auditory warnings work pretty well in most cases, but we recently showed—not surprisingly—that when you’re talking on the phone you miss those as well.”

Gray uses an advanced driving simulator to study different kinds of warning systems. His research shows that the most effective warnings use tactile signals—strong vibrations delivered through the seat belt, steering wheel, or seat.

“These are really effective because you’re not doing anything with, say, your shoulder when you’re driving. You don’t have any stimulation there, unlike your ears and your eyes,” Gray says.

But even tactile warnings lose effectiveness when drivers overwhelm their senses too much. Psychologists theorize that people have separate pools of resources for the different senses. For example, you can listen to music in the car and still keep your eyes on the road. However, when the demands on a particular sense exceed a certain level, the other senses become impaired.

Gray’s research bears this out. He has conducted experiments in which drivers were asked to conduct either simple or complex cell phone conversations.

“In one experiment we had people talk about really simple things, like the weather. When we did that, the tactile collision warning worked just as well as if you weren’t talking at all. But another condition we did was to have a complex conversation—kind of a simulated business meeting. When you do that, then the tactile warning is better than nothing at all, but its effectiveness goes way down.”

Gray’s research will help automakers design better safety features for cars. But he notes that most of his research would be obsolete if people always followed a few basic rules:

  • Always wear a seat belt
  • Never drive under the influence of alcohol
  • Obey the speed limit
  • Don’t talk or text on the phone while driving

He also emphasizes defensive driving.

“The good drivers I’ve seen are always anticipating what’s going to happen. So if you’re in the right lane and someone’s merging via the on-ramp, and you are potentially going to be there at the same time, move over. You can avoid a lot of problems if you are anticipating what people are going to do,” he says.

He also says that simulators like the one in his lab could vastly improve driver education programs.

“There are basically two kinds of knowledge: declarative knowledge, which is facts, and procedural knowledge, which is how to do things. Doing tests where you have to remember what distance to put between you and the car in front of you uses declarative knowledge. And that doesn’t transfer very well to actually doing it. It doesn’t mean anything when you actually get on the road—you have to experience it. That’s where a simulator can come in,” he says.

For Gray, the research isn’t just a job—it’s personal. Growing up in rural Canada, where driving long distances was a necessity, Gray saw his share of roadside tragedy. He says that every year he was in high school, at least one student died in a car accident.

“Driving is something that we created and we could make it safer—we just have to commit to it,” he says. “I believe that driving accidents are one of the most serious problems in society and also one of the most solvable. There are so many things we could do to make the situation better.”

Driving safety research is funded by the National Science Foundation and Nissan.