Driving Miss Dangerously

California’s new law, which prohibits talking on the cell phone while driving unless the cell phone is hands-free, went into effect earlier this month.* I am relatively certain that I am now a more dangerous driver.

I’m not saying it’s totally safe to talk on the phone while driving. But for me, talking on the phone is no more distracting – indeed, probably less distracting – than talking with someone in the passenger seat. And driving with one hand? I often do that when I’m not on the phone. I know this is a self-assessment and therefore possibly biased, but I think – for me at least – there are only two dangerous phases of talking on the phone while driving: (1) before the conversation, when I’m fumbling to find the phone and either dial a number or answer before the call gets shunted to voicemail, and (2) after the conversation, when I’m hanging up and stowing the phone somewhere. These are the phases when I’m most likely to take my eyes off the road.

The new cell phone law has increased the inconvenience of phase 1. Now I have to pick up the phone, locate the earphone hanging at the end of a cord somewhere, stick it in my ear, and finally answer or dial. (Phase 2 is also slightly more inconvenient, though it’s not too hard to pull the earphone out.) The end result is that I spend more time distracted from my driving than if there were no law.

This development reminded me of an episode of “Mythbusters” I happened to catch about a week ago. “Mythbusters” is a reality show where they use “science” to test popular empirical claims. This episode’s claim was that using a cell phone while driving is more dangerous than driving while intoxicated. The testers ended up concluding that the claim had been confirmed by their experiments. Now, the testing method suffered from a problem that afflicts all episodes of this show, namely that they reach conclusions based on sample sizes of just one or two. But setting that objection aside, the testing method suffered in a couple of other ways as well. First, it’s apparently illegal in California to drive drunk even for experimental purposes on a closed course. So for the test of driving while intoxicated, they had the two drivers drink enough alcohol to reach a BAC almost 0.08. This necessarily biased the test in the direction of saying that cell phone usage is more dangerous. Second, the test for driving while on the phone was really hard. The person on the other end of the conversation was, among other things, giving the drivers word problems that involved geometrical visualization (e.g., “You have three shapes, a circle, a square, and a triangle. The triangle is to the right of the square, and the square is above the circle. Is the circle (a) left of the triangle, (b) above and to the right of the triangle…” etc.). It was not clear whether getting the wrong answers to these questions hurt the driver’s score. But even if not, it seems they chose exactly the sort of conversation that is most likely to distract the driver and least like real-life phone conversations, thereby again biasing the test toward saying cell phone use is more dangerous.

Supposedly there are some serious studies that have found cell-phone-driving to be dangerous. I’m too lazy to look them up right now, but I believe they exist. What I wonder, however, is what the studies have used as the relevant control group. If they’re composed of drivers who are not distracted in other typical ways – by talk radio, by conversation with other people in the vehicle, etc. – then the control group could be too ideal to provide a relevant point of comparison. And the experimental group’s performance will also depend, I suspect crucially, on the type and intensity of conversation they are expected to carry on.

* UPDATE: Apparently I was mistaken about the implementation date; the law doesn't go into effect until July 1, 2008.