Texting while driving is a very real problem. The cause of the problem, however, isn’t recklessness but an incorrect perception of safety on behalf of drivers who feel little risk in texting. We can write all the anti-distracted driving laws we want but, at best, we will only displace the problem, replacing texting with some other distraction. To really address this problem, we need to be willing to incorporate driver psychology, including risk response, into our engineering approach.

A headline in my home state’s paper of record called for what is quickly becoming consensus in this country: grieving Minnesota families plead for stricter distracted driving laws. The article went on to describe a number of horrible tragedies that have happened when people were texting, talking or otherwise engaging their mobile phones while driving. Here’s a sample:

Andrea Boeve, a 33-year-old nurse, wife and mother, tucked 4-year-old Claire and 1-year-old Mallorie into their bike stroller the morning of June 30 and went for a ride along the quiet rural highway that connected the family farm in Steen to the girls’ grandparents’ home next door. On the road behind them was 25-year-old Christopher M. Weber, who told crash investigators he was looking at his phone, waiting to see which number he should press to advance to the next step in his bank’s automated phone system. He said he had no idea that anyone was in front of him until he felt a thump and saw bike wheels in his rearview mirror.

That is undeniably horrible. It’s painful to even read. If we could enact a law that would prevent, or even reduce, this kind of thing from happening, how could we not do that? A stricter set of laws combined with more driver education seems like the rational response for concerned government officials.

But is it? Before we enact any more laws, I think it is important to understand exactly why people text while driving, or change the radio station, look at a map, clip their nails or — like I did in my first job nearly 20 years ago when I had to drive to a job site ninety minutes each direction every day for four months — read the newspaper. They do it because they feel safe.

Feeling safe is also the reason why a lot of people don’t. My mother-in-law is a beautiful, wonderful and kind person. She is also a rather paranoid driver. She avoids snow, ice and rain (in Minnesota, no less), keeps her hands at 10 and 2 and would never do anything that would divert her attention from the road while driving.

When it comes to assessing risk, each person has a different threshold. I have no qualms driving through most blizzards, even in my little Honda Fit. My mother-in-law has an all-wheel drive Subaru and she won’t go out if there is a threat of snow. I would never even take a ride on a motorcycle, let alone drive one myself, yet I know lots of people feel very comfortable with the risk of riding a curvy country road at 60 mph on a cycle, the threat of a deer running across the road omnipresent.


So let’s acknowledge that everyone has a different threshold for risk — let’s label it ‘R’ — and that this threshold may change depending on time, place, conditions, age, etc. For example, I may develop a higher threshold for risk when I’m running late for work than when I’m on my way to a meeting I’m dreading.

What happens to our threshold for risk (R) when we go out and improve a road to make it “safer”? What happens when we widen lanes, flatten hills, straighten curves, remove obstacles and add recovery zones? Does the risk threshold go down? Does it go up? Does the teenager suddenly realize their mortality? Does my mother-in-law suddenly become more reckless?

No. Our threshold for risk doesn’t change. What changes is our perception of how risky the environment now is.

Let’s say that a driver on a road posted at 55 mph drove exactly the posted speed and felt neither danger nor safety but instead perceived that the road, at 55 mph, met their exact threshold for risk. Then the roadway is “improved” so that it is theoretically safer; wider, straighter, flatter and greater clear zones. Now that same driver perceives that the roadway at 55 mph is less risky than the R they are willing to bear. There is a gap between the risk the driver is willing to bear and the risk they perceive. How do they respond?

Before I answer that, I know some of you are lost. You think, What's wrong with just feeling safer? But that’s not how people react. Everyone assumes some degree of risk — from the teenager to my mother-in-law — when they choose to drive. That degree of risk is acceptable — it is below their threshold for risk — or they would choose not to drive. At a degree of risk below their own personal R, they feel safe. What happens when we make their driving environment far below their personal R?

The answer is that they utilize that gap in some other way. Let’s use an extreme example to help visualize the effect. Say you are driving across Nebraska on the interstate. You can see for miles in every direction. The lanes are wide. The road is straight. There are wide shoulders, no trees if you go off the road and no cars anywhere in sight. Do you feel comfortable looking down to change the radio station? How about talking to someone in the passenger’s seat? Talking on your cell phone? Texting? Speeding? Reading a newspaper?

The interstate across Nebraska is below most people’s R threshold. A driver has a tolerance for risk that is unlikely to be reached there and so they will feel comfortable doing other things. Some will feel comfortable changing the radio station. Some will speed. Some will text while driving. We’re not talking about whether or not this is reckless but how the driver perceives the risk.

Now ponder a busy parking lot at a mall. How many people text while driving in such a place? Talk on a cell phone? Speed? Read a newspaper? Likely very few and I think it is clear as to why; the driver is likely to perceive greater risk — even though the speeds are slow — due to the complex and random nature of a parking lot environment.

People text and drive because they perceive little risk in doing so. And when we make the driving environment feel safer, we increase the number of people who are going to perceive little risk in texting while driving.

Does this mean we don’t pass a law against texting while driving? Not necessarily, but it is important to understand the limits of such a law. In his book Risk, John Adams points out that a lot of safety legislation doesn’t eliminate risks but simply displaces it. The book was released in 1995, before the widespread adoption of cell phone, so he uses the example of a ban on motorcycles, another dangerous activity:

All risks can be displaced. If motorcycling were to be banned, it would save about 500 lives a year. Or would it? If it could be assumed that all the banned motorcyclists would sit at home drinking tea, one could simply subtract motorcycle accident fatalities from the total annual road accident death toll. But at least some frustrated motorcyclists would buy old bangers and try to drive them in a way that pumped as much adrenaline as their motorcycling, and in a way likely to produce more kinetic energy to be dispersed if they crashed. The alternative risk-taking activities that they might get up to range from sky diving to glue sniffing, and there is no set of statistics that could prove that the country had been made safer, or more dangerous, by the ban.

So if we banned texting and driving, am I suggesting that some people would therefore get high and drive for the adrenaline rush? No. I think the public information campaigns raising awareness of texting and driving have done a lot to make people aware of the risks. Their risk threshold hasn’t changed, but their perception of the risk has increased, and a law might do that to a greater degree. What's likely to happen, however, is that people may reduce their texting, but they will add something else. Maybe they'll drive at higher speeds or do their make up or eat a sandwich. 

So if we’re really serious about reducing driving accidents, we won’t stop with public service announcements and a law that may ultimately do little. If we’re serious, we’ll actually design our roads and streets to be safe. That means not simply more efficient engineering but actually pondering driver psychology and asking difficult questions about how people respond to our designs. We don't want people to perceive the road as safe; we want it to actually be safe.

When we build a roadway wider, straighter and flatter, people perceive less risk with that roadway. They respond to that perception — they fill the gap in R to their personal risk threshold — with other risky activities. Almost all the time there is no negative consequence to that response, but in the small instances where there is, the results are tragic. Put enough people on the road in these environments and you have fatality statistics that are appalling.

Where we use Forgiving Design techniques — wider, straighter, flatter — then we have to make sure that the roadways are totally free of the randomness that causes collisions. We can’t have highways we're all familiar with: 75 mph design speed, 55 mph posted speed and driveways and accesses on average every 150 feet. It is safe 99% of the time. The other 1%, it is an absolute death trap; fast moving cars with randomly turning traffic, bikers and pedestrians on the shoulders and kids playing in the yards. The problem is, it feels safe 100% of the time. There is no way to identify that 1% until it is too late.

Counterintuitively, the way we make streets safe is to make them dangerous. That is to say, when people perceive the real danger, they respond by being more careful. So, our task as engineers is not to design roads to be wider, straighter and flatter — in most instances that is making them more dangerous by giving the illusion of low risk — but to design them so drivers correctly perceive the level of risk they are assuming.

In places with pedestrians and cyclists, that means we’re doing our job if drivers feel safe only at speeds less than 20 mph, if their perception of risk meets their risk threshold at that low speed. On high speed roadways, that means we’re doing our job when there are no random or unforeseeable ways for drivers (or anyone else) to interfere with the traffic stream as random acts will not be perceived. For the stroad environments in between, we need to pick either a road or a street design or, if all else fails, do what we can to lower speeds by helping the driver perceive the real risks.

(Top photo source: U.S. Air Force/Staff Sgt. Andrea Thacker. A different version of this article was originally published in 2014.)

Related stories