My ex-wife made up a game to play at the airport. She called it “Make Up A Life.” It was a simple game: one person picks a fellow traveler out of the crowd and the other must spin a plausible background narrative.
My answers were notoriously fantastic (“Oh, that’s Princess Beatrix of the Netherlands; she’s headed to Miami to purchase a new yacht.”) but my wife could create credible sounding biographies effortlessly. We once spied a man with a slight paunch in a red flannel shirt. She declared, “High school shop teacher. Wood, not metal. His wife died unexpectedly last month. He’s trying to move on, so he’s decided to go to Florida on one of those free trips you get if you agree to listen to the pitch for buying a timeshare.”
I took another look at the man and suddenly I could see the sadness on his face from the loss of his wife. I could tell that he was putting up a brave front. And I could almost smell the sawdust on him. Her description was so apt and she delivered her verdict with such confidence that I instantly agreed.
A few minutes later we ran into the man at a newsstand on the concourse. I said hello and asked politely where he was headed. I was prepared to say, “I’m sorry for your loss.”
It turns out the man was a Canadian software engineer. His work visa had just expired and he was heading back to Toronto to renew it. Oops.
Since 2007 the TSA’s “Behavior Detection Officers” (BDOs) have been playing a version of our game. BDOs supposedly watch for signs of fear, anxiety, or deception among travelers, on the assumption that these are characteristics of Bad People. According to the TSA, BDOs can detect “micro-expressions” the way a card-sharp detects his opponent’s “tell” at the poker table.
For years the Government Accountability Office has been skeptical. The GAO first criticized the TSA for rolling out the program–known as SPOT–when there was no scientific proof that security staff could really be trained to see into your soul. Once SPOT was in operation, the GAO instructed the TSA to at least try to validate that it was working, the way empirical studies prove that aspirin relieves headaches, even if we don’t exactly know how it works.
That was two years ago. Last week the GAO said that TSA still hadn’t produced reliable data to support the program.
The TSA insists that the $900 million program is critical to our nation’s safety. I’d like to believe them. After all, SPOT may be the least invasive security protocol the TSA has in place. If there is truly a way to flag terrorists without hassling other members of the traveling public, I’d be all for it, but if you’re going to relying on human dowsing rods to find bad guys, you have to demonstrate that they work, at least a little. But even the TSA’s parent agency, DHS, has reported they don’t work.
So far, the TSA has not done so, at least not to the satisfaction of the non-partisan GAO. That should come as little shock to those who follow the agency. The TSA has had a long-standing grope-now-and-ask-questions-later policy. Their controversial nude body scanners were rolled out without the required public notice and comment period, and the TSA still has not allowed independent testers to ascertain whether the machines constitute a health risk. Other machines like the infamous “puffers” (explosives detection) were purchased, seemingly on a whim, and mothballed when they were found not to work.
Through it all, the TSA has assured Americans that everything is fine and we’re better off not asking questions. When we do, sometimes they simply don’t answer, even when the questioners are Congressional committees.
Here’s the thing about behavior detection. Even if a BDO could spot fear, anxiety, or deception, that’s not saying much. Airports are full of the stuff:
- Why isn’t this line moving? I’m gonna miss my plane. I can’t miss my plane. I’ll have to run to get to the gate on time. I hate running through airports. What’s the hold-up up there?
- I gotta nail tomorrow’s interview. I gotta. I need this job. OMG, what if I don’t nail this interview?
- I can’t believe I’m flying off for a weekend with a guy I only met online. Does that make me a slut? I must be out of my mind. But I’m not a slut. Really, I’m not. Okay, maybe a little. Oh this is a bad idea.
- I shouldn’t have smoked so much weed last night. Is that a police dog? Do they sniff for drugs or just bombs? I hope just bombs. Oh dear, here they come.
- So, here I am, flying home for the holidays again. Really looking forward to another season of family dysfunction. I need a life.
Hello, Traveler, do you have a moment?
That guy in the security line shifting nervously from one foot to the other and casting furtive glances down the concourse may have a bomb in his briefcase, or he may just have to pee. Assuming a BDO can sniff out actual anxiety at 30 paces, even proponents of the screening admit that determining its cause requires additional investigation.
To accomplish that, the BDOs will engage a suspicious traveler in casual conversation. That’s harder than it sounds. News reporters know that the presence of a camera often creates or changes a story; medical researchers know that they may create a placebo effect unless a study is double-blind; quantum theorists have discovered that the act of just observing a stream of electrons can influence the behavior of those electrons. The TSA needs to be similarly hip to the idea that when an agent of the federal government starts asking why you seem so nervous, it tends to add to the nervousness.
In other words, these interviews require some finesse. As with answers to Congressional questions or validation studies, finesse is frequently lacking at the TSA. It is are, after all, the no-exceptions, everyone-must-be-screened (even that disabled kid, even the 90-year-old in the wheelchair, even the US Senator) agency.
This might be where I should mention that there are 2,199 BDOs nationwide and that two thirds of these finessers (65.3%) have only a high school diploma or less.
It’s also worth nothing that, according to the GAO, an interaction with a BDO requires about 30 seconds. If the BDO determines that his initial suspicions about the traveler have merit, it can take (on average) 13 minutes more to determine if law enforcement should be called. Given the $900 million pricetag of the program so far, each BDO represents an investment of $409,276. That doesn’t seem particularly cost-effective to me, but what do I know?
One final fact for those who may find themselves mano-a-mano with a BDO. According to the TSA:
Information contained in the SPOT database is based primarily upon observations . . . and casual conversation which may be terminated at any time by the individual without consequence to the individual. If an individual is referred to a LEO, their ability to decline is subject to law enforcement protocols.
In other words, the BDO won’t tell you this, but you’re free to walk away without answering his questions. Of course, if you do so, you’ll probably be deemed to be acting suspiciously, and law enforcement may be called.
Try This at Home!
Naturally, the TSA won’t tell us exactly how their BDOs are able to make a determination that a traveler is suspect, but there are lots of places that provide “training.” The TSA favors former U.C. Davis professor Paul Ekman, about whom we wrote here (he now has his own behavioral analysis consulting company). I tried my luck on a website run by The Center for Body Language. Without any training, my accuracy was 46%, meaning I’d have a better chance of discerning someone’s motives by flipping a coin than by trying to read his expression. The Center says that with training I can raise my accuracy as high as 80%, meaning that I’ll only be wrong a fifth of the time, rather than half.
Regardless of who provides the training, all behavior detection is based loosely — very loosely — on the concept of neurolinguistic programming (NLP). If that term sounds familiar, you probably heard it on a commercial for Lumosity, the controversial “brain training” software. NLP posits that there is a link between neurological function and linguistic expression. Psychologists Richard Bandler and John Grinder, who introduced the concept in their 1979 book Frogs into Princes: Neuro Linguistic Programming, suggest that an NLP expert can interpret “Visual Accessing Cues” to divine the truth.
Here’s how it works: head out to the airport, find an average business traveler, and ask him to “imagine putting a bomb on an airplane.” Assuming (a) that he has no experience putting bombs on planes, he’ll have to construct a mental image of that act. Further assuming that (b) his brain is “normally organized” and (c) that he is right handed, his eyes will track up and to the left as he accesses the part of his cantaloupe that spins out ideas. In contrast, if he is a normally organized, right-handed terrorist who just planted a bomb on an aircraft, his eyes will track to the right as he accesses the part of the brain that remembers the act.
Terrorists look to the right. Check. Of course, if you phrase the question differently –“did you put a bomb on that airplane?”– his eyes will track up and to the left as he spins a lie. So terrorists look up and to the left. Check.
Simple, right? I mean, we all learned this from watching “Law & Order.”
Just remember that when you’re talking to someone, his right is your left, unless he’s part of the 10% of the population that is left-handed, in which case his left is your right. And if he’s a non-native speaker of English, he may look to the right (your left) anyway (to access memory) as he translates your question. After that he’ll look directly at you, a clear sign that he thinks you’ve lost your mind. Check and mate.
Got it? Okay, let’s test your skills. You’re in the airport and you approach a normally organized right-handed native speaker of English who had a concussion from playing high-school football and you ask “where are you headed today, sir?” Your newfound friend looks down at his boarding pass and mutters “Denver.”
Quick: For the grand prize of $409,276, should you tackle the lying sack and call law enforcement?
Wait, don’t answer yet.
You may be surprised to know that Bandler and Grinder never said that their idea would hold water. They postulated that visual accessing cues might work. Even that wasn’t ambiguous enough to keep most peer reviewers from trashing their idea. After reviewing over 400 studies from the last 60 years, the GAO concluded that when trying to decide if a traveler is lying, behavior detection works only slightly better than flipping a coin. Tails: you’re a terrorist.
But really, what does the GAO know with its empirical review and multivariate regression analysis? I’ll take the word of TSA spokesman Bob Burns, who wrote in 2010, “I was a trained BDO in my airport days and personally feel the training I received was extremely valuable to security. Also, my wife, kids, and used car salesmen are not able to get one over on me.”
That’s good enough for me.