Driven to Distraction

Published in IEEE Spectrum Magazine, January 2010

I used to feel that the ultimate aim of communications research was telepresence —creating the perfect illusion of being where you're not. However, now I'm thinking that we did too good a job of creating this illusion and that the law of unintended consequences is taking hold. Sometimes you need instead to enforce the perfect sense of being exactly where you are at the moment—like when you're behind the wheel of a car hurtling down the highway. Maybe instead of telepresence (distant presence) we need plesiopresence (near presence).

The problem is technological distraction. It's becoming serious, and there is no solution in sight. Laws against using phones in cars aren't working, because the problem isn't that you're putting a phone to your ear; it's that you're putting your brain somewhere else. So are we technologists helpless? Now that we've created technological distraction, can we create technological traction?

Why is talking to the passenger next to you different from talking to someone on the phone? We need to better understand the psychology of cellphone conversations. And not just in cars. There's an illusory cone of privacy that surrounds people's all-too-public conversations everywhere.

I often travel in the quiet car on Amtrak trains, where cellphone conversations are forbidden. Like my fellow occupants, I'm alert to anyone breaking the rules. I can nearly always tell when someone is talking to a "cellmate" rather than a seatmate. It simply sounds different. I can't explain how I know, I just know, and apparently so do the other passengers. Perhaps there are some useful clues there for the psychologists.

While this much-needed research in the cognitive area is going on, we might as well go ahead and try some things. First, we need to detect when someone is using a cellphone in a moving vehicle. There could be a monitor in the car, but this would be either relatively ineffective or hugely expensive, given the number of older vehicles still on the road. Instead, we could detect cellphone users through base-station handoffs, triangulation, or Doppler shift of the received signals. That's the easy problem. The hard problem is what to do when we do detect it.

We could, of course, just block calls to and from moving vehicles, but this probably wouldn't be allowed. Doctors and other people would cite emergencies or other urgent circumstances in which calls must be permitted. Maybe we could simply make the conversations unpleasant by adding static to simulate the old analog long-distance calls, randomly adding the screech of chalk on a blackboard, or inserting a simulated satellite delay. Most deviously, we could introduce an echo delay that makes it almost impossible to talk. But I'm not sure whether such annoyances would help or hurt the sense of telepresence. They might draw the brain further into the call instead of pushing it back out.

A milder alternative would be to query the user at the beginning of a call, the way software does when it thinks you're about to overwrite a file. "Are you sure you want to do this stupid and dangerous thing?" the cellphone might ask. Or maybe we should limit calls to some small amount of time, then prevent you from redialing until you stop the car. Whatever we do to inhibit calls from moving vehicles, it would have to be mandated by law, as it would surely be unpopular. Everyone believes that he or she alone can use a cellphone faultlessly; it's the other driver who's the problem.

So here's my big idea. Remember the early days of high-occupancy-vehicle lanes, which you can use only with a passenger? Some people bought dummies to put in the passenger seat. Suppose we took these dummies and instrumented them so that a cellphone voice came out of their mouths, lips moving in sync with the words. Every now and then the dummy would nod its head, look at the road, glare at the driver, and say, "Watch the road, dummy!"

Maybe you have a better idea. I sure hope so.