TUCSON, AZ (Tucson News Now) - Border crossing involves a lot of waiting. So there is the trusted traveler program to help frequent crossed cut down on lost time.
But there aren't enough officers to handle all the applications for the program. That is where University of Arizona researchers come in.
Face-to-face interaction is especially important to law enforcement when they trying to evaluate any degree of deception. But what if that face-to-face interaction would take place between yourself and a machine?
Or in this case an avatar which you could soon be seeing at the Nogales Port of Entry.
It is a research tool right now, though we are probably only a couple of years away from seeing something like this at every port of entry in the United States.
The idea is to expedite the huge backlog of trusted traveler applicants who must now go through face-to-face interviews with customs agents before being cleared for travel.
"So they fill out an application online, they pay a fee and then they wait because there's not enough officers to do these interviews," Dr. Aaron Elkins, UA researcher said.
That is where the avatar comes in to ask people questions about their application and to try and make sure they are in fact telling the truth.
"What is your name? Do you use any other names? Do you use drugs? Have you been convicted of any crimes? Questions relevant to their application," Dr. Elkins said.
How does the avatar know if you are really telling the truth? Consider it an educated guess, an extremely educated guess at that.
"It relies very heavily on 30 plus years of communications research," Dr. Elkins said.
Dr. Elkins is a post-doctoral researcher at the University of Arizona and his area of expertise is vocalics, or the study of the human voice.
"Some people when they lie or they're upset -- it's all over their voice," he said. Combing his own research with three decades of communications data, Dr. Elkins has been able to fine tune the avatar so that it can detect inconsistencies in the human voice.
When that occurs, it notifies agents through a notepad that something's not quite right with an answer given.
"I will stress that is a very large leap to say that they're lying...or that what they're saying is untrue -- but what it does is draw attention that there is something going on," he said, precisely the intent behind any credibility assessment. What is more, with the avatar, the interviewer is always the same.
Meaning it doesn't run the risk of outside interference possible in other tests.
"I think it's critical that we can use technology for applications that may be more effective or more accurate than a human," Jeffrey Proudfoot, UA Ph.D. student said.
Proudfoot studies the viability of deception sensors compare to the gold standard in lie detection, the polygraph test.
From a control standpoint alone, he sees a lot of potential in the avatar.
"One of the problems with the polygraph is that the interviewer plays such a huge role. And you can't control for that interviewer across polygraph interviews. That's one of the benefits of the Avatar is that it's the same interview every time," he said.
Again, it is still a work in progress.
Dr. Elyse Golob, executive director for the border center of excellence says, "We take it out into the field, it gets dirty, people kick it, officers curse at it." But that is what makes this so rewarding for those who have put so much into it.
"I hope we keep moving closer and closer to artificial intelligence," Dr. Elkins said. Maybe not to the level of Star Trek or movies like I-Robot just yet, but the avatar just might be the first step.
"I'm sick to death of the mouse and keyboard. I would love for us to talk to our computers in a natural way -- and have them understand and respond in a way that I would expect a person to," Dr. Elkins said.