Verified by 
[This is a special 9/11 post, unrelated to the "living single" theme of this blog. I'll return to single life in my next post.]
Since 9/11, a lot has changed at airports. You already know about the scans of passengers and their luggage, the show-your-liquids and the take-off-your-shoes rules. But did you know that as you are standing in line at airport security, before you even get to any of the machines, you are already being monitored?
It doesn't happen everywhere. As of the last publicly available U.S. government report (GAO, May 2010), the program was ongoing in 161 airports in the U.S. It is an observational program called SPOT: Screening of Passengers by Observational Techniques. While you are standing in line waiting to be screened, two "behavior detection officers" are unobtrusively observing you. What they are trying to figure out, just by looking at you, is whether you may be up to no good.
This is an even more challenging task than the already formidable task of trying to tell whether someone is lying or telling the truth at that moment just from their nonverbal (and perhaps verbal) behavior. In the airport security instance, people are trying to tell whether you are behaving in a way that might indicate that you are going to do something bad, or maybe even catastrophic, such as blowing up the plane.
You can imagine the potential objections on grounds such privacy and accuracy (which I'll turn to next), but imagine, for a moment, that the technique were successful. It would mean that useful information could be obtained without adding any other machines or scans or disrobing or (for the vast majority of passengers) extending the time spent in the security process.
SPOT is actually not designed to replace any of the other layers of security. It is an added layer, one more chance to catch the bad guys before they do harm. At the core of SPOT is a checklist of behaviors that the behavior detection officers are trained to recognize. Each behavior is assigned a certain number of points. The officers memorize the list and the point-values, and use that information to determine whether any given passenger should be flagged as suspicious.
So what are these behaviors? How many are there? How many points are associated with each? We don't know. That information is classified as "sensitive security information." That's one reason it is hard to evaluate the SPOT program.
There are a few hints about the behaviors in the 2010 report. The cues are ones believed to be indicative of stress, fear, or deception. Officers, we are told, are trained to look for and recognize "facial expressions, body language, and appearance that indicate the possibility that an individual is engaged in deception and fears discovery." The checklist does not include indicators that might suggest profiling, such as race, ethnicity, or religion.
If you do get flagged as suspicious, the first thing that will happen is that one of the officers will initiate an informal conversation with you. The officer might make some small talk, ask where you are headed, and so forth. The purpose is to see if there is some obvious reason why you may be behaving suspiciously that does not pose any threat to anyone else (for example, you are headed out of town to have an affair). If the officer is satisfied with your answers, nothing else happens - you are good to go. If you had not read this blog post or the government report, you might not even realize that the conversation was anything but casual.
If the officer is not satisfied with your answers, then you are referred for further screening to another Behavioral Detection Officer and a Transportation Security Officer. If your behavior exceeds some threshold, then you will be referred to a Law Enforcement Officer, who can tap into various data systems, do background checks, and decide whether to make an arrest or clear you for boarding. If you are cleared by that officer, the Transportation Security Officer can still decide that the threat is too great to allow you to board.
So does SPOT work? On April 6, 2011, a Congressional subcommittee held a hearing on that question. It was broadcast on C-SPAN and can be viewed here. According to the proceedings and the written statements submitted by the witnesses, a high-quality study has been conducted and described in a report, but that report has not yet been made publicly available. The yet-to-be-released results are from a base rate study, in which the number of people detained by the SPOT program (and the reasons for which they were detained) was compared to the results of choosing people for screening at random.
The results that were reported in the 2010 report were based on passenger boardings at SPOT airports between 2004 and 2008. The relevant passengers numbered about 2 billion. Of those, about 152,000 were referred for secondary screening. Of those 152,000, about 14,000 were referred to law enforcement officers, and then about 1,100 were arrested. We don't know how many should have been arrested but were missed.
What were the arrests for? The greatest number, 39%, were arrests of people who were undocumented. Next were those with outstanding warrants, 19%, then those who had fraudulent documents, 15%. Another 12% were arrested for a miscellaneous category of reasons that included intoxication and unruly behavior, and an additional 12% were arrested for drug possession. Two other categories each comprised, at most, 1% of the arrests: undeclared currency and documents that were suspect. For a final 1% of the arrests, no reason was recorded.
So where were the arrests of terrorists? There may not have been any. The report estimates that suspected terrorists account for 1 in every 173 million passengers. During the time of the 2004-2008 study, 23 suspected terrorists passed through SPOT airports. However, even in airports that are participating in the SPOT program, behavior detection officers are not stationed at every check-point. The report notes that there was no way of knowing whether the 23 suspected terrorists passed through checkpoints monitored by SPOT-trained officers.
If you want to listen to the subcommittee testimonies or read the written reports submitted by the witnesses, you may want to pay special attention to Professors Maria Hartwig and Paul Ekman, who provide starkly contrasting assessments of the performance and promise of SPOT. It will be interesting to see what we learn from the results of the newer base-rate study.
