Catching Terrorists with “Behavior Detection”?

An editorial in yesterday’s New York Times questioned the nearly $1 billion the U.S. Transportation and Security Administration has invested in training and employing officers to identify high-risk airline passengers. In 2011 and 2012, T.S.A. behavior-detection officers at 49 airports “designated passengers for additional screening on 61,000 occasions.”

The number successfully detected and arrested for suspected terrorism? Zero.

But then again, the number of plane-destroying terrorists they failed to detect was also, I infer, zero. (Wonkish note:  A research psychologist might say the T.S.A. has made no Type II errors.)

Regardless, psychological science studies of intuitive lie detection, as the Times’ John Tierney noted in an earlier article, suggest that this has not been a wise billion-dollar investment. Despite our brain’s emotion-detecting skill, we find it difficult to detect deceiving expressions. Charles Bond and Bella DePaulo reviewed 206 studies of people discerning truth from lies. The bottom line: People were just 54 percent accurate—barely better than a coin toss. I have replicated this in classroom demonstrations—by having some students either tell a true or a made-up story from their lives. When seeking to identify the liars, my students have always been vastly more confident than correct.

Moreover, contrary to claims that some experts can spot lies, research indicates that few—save perhaps police professionals in high-stakes situations—beat chance. The behavioral differences between liars and truth-tellers are just too minute for most people to detect.

Before spending a billion dollars on any safety measure, risk experts advise doing a cost-benefit analysis. As I reported in Intuition: Its Powers and Perils, some people were outraged when the Clinton administration did not require General Motors to replace ill-designed fuel tanks on older model pickup trucks. The decision spared General Motors some $500 million, in exchange for which it contributed $51 million to traffic safety programs. “GM bought off the government for a pittance,” said some safety advocates, “at the expense of thirty more people expected to die in fiery explosions.” Actually, argued the Department of Transportation, after additional time for litigation there would only have been enough of the old trucks left to claim 6 to 9 more lives. Take that $500 million ($70 million per life)—or the $1 billion more recently spent on behavior detection—and apply it to screening children for preventable diseases (or more vigorous anti-smoking education programs or hunger relief) and one would likely save many more lives. By doing such cost-benefit analyses, say the risk experts, governments could simultaneously save us billions of dollars and thousands of lives.

Ergo, when considering how to spend money to spare injuries and save lives, critical thinkers seek not to be overly swayed by rare, dreaded catastrophes. The smart humanitarian says: “Show me the numbers.”  Big hearts can cohabit with cool heads.