Cell phones and choice architecture

This summer, the New York Times ran a series of articles looking at the risks of distracted driving, especially as this relates to cell-phone use and text messaging while operating an automobile.1 The Times ran these pieces on their front page, generating considerable interest and spin-off coverage in other media.

While acknowledging the difficulties of studying risk associated with a transient exposure, the stories emphasized that, in the USA, cell-phone use is the most prevalent driver distraction and one that is especially common among the youngest (and, at baseline, riskiest) drivers. Citing data from simulator studies, the author reported that cell phone use increased the risk of crashes or near crashes anywhere from four- to eightfold, an impact comparable with driving with a blood alcohol level above the legal limit.2 In a salacious aside, the series revealed that in 2003, researchers at the US National Highway Traffic Safety Administration compiled research demonstrating the risks of distracted driving but withheld these findings when the agency came under pressure from Congress.3

The Times also included data collected by researchers at the Virginia Tech Transportation Institute. In an elegant design, their studies use longitudinal data from event recorders mounted in vehicle fleets. These naturalistic studies are …