Tell Me A Fairy Tale

By John Carr, NMA Massachusetts Activist

A federally funded study (released earlier this month) warns that texting while driving is more dangerous than previous studies may have led you to believe. If the PR cycle follows its typical course this will be followed by a federally funded campaign for more federally funded enforcement.

Here’s the headline from the Texas Transportation Institute: “New study says texting doubles a driver’s reaction time.”

What does that mean? Read the real report rather than the press release.

Drivers were asked to read or write a fairy tale via text message while driving 30 miles per hour along a straight test track. Sensors measured speed, weaving, and time to notice a light had turned on.

As an aside, this is boring by historic standards. A man was recently awarded the Ig Nobel Prize in Public Safety for a 1960s study on distracted driving. He drove on Massachusetts Route 128 with a helmet blocking his view of the road most of the time. Lawyers and human subject ethical standards have taken the fun out of modern safety studies.

The first problem with distilling this study into a press release is it deliberately does not test realistic driving conditions. In ordinary traffic most drivers aren’t in the act of texting and nothing surprising is happening around the car. It takes a lot of time to get enough data to quantify how well texting people drive. The TTI study tried to distill texting and driving down to its essence, like measuring intersection safety by sending drivers through line after line of cross traffic without break.

Subject were given two unusual tasks in addition to driving. They were asked to read or write a fairy tale. The long message had to be broken into four separate texts. The length ensured that texting kept drivers busy. TTI researchers were supposed to remind them of this task as needed. Meanwhile, they were to look for a small, green LED on the front of the car and press a button when they saw it lit. The button was mounted on a stalk near the steering wheel, like the button on the end of the turn signal or wiper lever.

This is not a realistic test of driving. It’s like activating your windshield washer when a car in the next lane uses its turn signal. You can do it. You have to think about doing it. You aren’t going to be as fast as you are at braking or swerving to avoid a crash.

In driving, two important events to watch for are brake lights and cars moving in from the side. Our vision is well designed for these tasks. Once upon a time detecting motion or changes in light in the corner of our eye saved us from predators. The green LED in the study was designed so that the driver had to look at it to notice it. (p. 27, second full paragraph) How this was done was not stated; presumably the LED was dim compared to ambient light.

The button was placed so the driver had to use a hand to press it, even though the hands were busy with the phone. You can brake with your foot even if you are texting with one hand and holding your coffee in the other.

Not surprisingly, even when drivers were not texting it took longer to press the button in response to the green light than it would take a driver to brake. Most drivers fell in the range 1-2.5 seconds. The time range used for highway design is up to 1.5 seconds to react to the simplest possible event. (Source: “Green Book”, the standard reference for U.S. highway design.)

When writing a text message most drivers took over 4 seconds to press the button. Sometimes they didn’t notice the light at all before it turned off 15 seconds later.

This result is the core of the publicity. Texting is like closing your eyes for four seconds, or 15. Or is it?

These drivers were not completely oblivious to the road for 15 seconds. In the simulated work zone, texting drivers maintained position at least as well as non-texting drivers. They were ignoring the least important task: watching for the light. Texting and driving (or driving and texting) took priority.

The report itself has some reservations; see pages 27-28. (If you want a DOT grant, you put the agency’s position at the beginning.) The measurement may show for how long a driver becomes reactive instead of proactive. Proactive behavior includes “periodic scanning of the driving environment for emergency vehicles or latent driving threats.” Reactive behavior includes “braking in response to a lead vehicle.”

Four seconds not checking for cars in driveways is much different than four seconds not looking at the road or traffic ahead. Becoming reactive makes texting drivers more vulnerable to mistakes by others. If you hit somebody who ran a stop sign, society blames the guy who ran the stop sign. If you rear end the car ahead, society blames you. Or if you don’t like the at fault vs. not at fault dichotomy, consider not watching the car ahead to be more risky behavior than not watching the car waiting to enter the road.

Reaction time is not simply a number. It always has to be interpreted in context.

As a cognitive science study this is interesting. I don’t believe the numbers translate to real driving. The evidence does not justify the press release. But “texting drivers become reactive rather than proactive” doesn’t sound as scary and it doesn’t bring in the grant dollars.

Not an NMA Member yet?

Join today and get these great benefits!

Comments are closed.