Information Design and the Placebo Effect

Despite Enron and Martha Stewart, scandal in the Catholic Church, and the failure to uncover weapons of mass destruction in Iraq, I would describe myself as a trusting sort, one who fundamentally still believes in the institutions that govern our public life. But last week, that trust was shaken to the very core by a report in the New York Times about the buttons that are mounted on poles at over 3,000 street corners in New York City. Despite the fact that they bear official-looking signs that read "To Cross Street/Push Button/Wait for Walk Signal/Dept. of Transportation," it appears that at least 2,500 of them have not worked for the last fifteen years.

Like everyone else, I've trusted those instructions, pressed the buttons and waited dutifully, fearing - and, indeed, this is the literal interpretation of the sign - that the light would not change, ever, unless one pushed the button. Now I learn that I've been the dupe of what Times reporter Michael Luo calls mechanical placebos, where "any benefit from them is only imagined." And, my eyes newly opened, I wonder: can this possibly be an isolated case?


Now that I think about it, I've always wondered about those "Door Close" buttons on elevators. I mean, the door always eventually closes, but it's hard to tell if there's really any causation involved. Like the crosswalk buttons, all of these buttons may function simply as therapy for the over-anxious. And it's significant that even if they seldom work, they still work sometimes. Every behavioral scientist knows that if you reward the rats every time, they take it for granted; if you never reward them, they give up. The most effective approach is to reward them every once in a while. This principal of intermittent reward is well understood by casino owners.

I myself have deployed meaningless information to assuage my own anxiety. We bought our first house from a fairly paranoid owner who had outfitted the (modest) property with an elaborate security system. Its operation was well beyond the ken of my family, and after setting off various alarms at various hours of the early morning, we finally had the whole thing disabled. But we left up all those signs reading "This Home is Protected by the Neverrest Ultra Security System," reasoning that intruders would be as alarmed by the signs as by the (now disarmed) alarms.

In post 9/11 Manhattan, this exchange of meaningless information has become part of daily life. Visit any office building over four stories in height and you're likely to run a gauntlet of inquisitors. The truly diligent ones subject visitors to x-ray examination and require tenant escorts. It's an inconvenient procedure, but at least you can understand its efficacy. More often, you're merely asked to sign a log and, sometimes, present your driver's license. How this is supposed to deter cunning terrorists, who presumably can acquire cheap fake id's as easily as anthrax or dirty bombs, I've never understood.

And of course, to move from the personal to the political, no one is exploring the frontiers of information as placebo like our own Department of Homeland Security. What exactly are we expected to make of Tom Ridge's color-coded terrorism alert levels? When the level is raised, are we supposed to hide under the bed or go about our business? Are they trying to reduce anxiety or increase it? Do they mean anything at all? We don't know, and I'm not sure they really know either. But one way or another, they seem to be trying to press our buttons.