SCOTT SIMON, host:
This month, the U.S. Army will issue hand-held screening devices to soldiers in Afghanistan. Those devices will be used to question suspects in crisis situations, like roadside bomb sites, so that Army personnel might determine whose account to trust.
The technology used is not even as exact as those in polygraphs. Will using the machines endanger innocent people or let U.S. soldiers be fooled?
Donald Krapohl is special assistant to the director at the Defense Academy for Credibility Assessment at Fort Jackson army installation in Columbia, South Carolina. That's where the military trains personnel on how to conduct tests. Thanks very much for being with us.
Mr. DONALD KRAPOHL (Special Assistant to the Director, Defense Academy for Credibility Assessment, Fort Jackson, Columbia, South Carolina): Thank you for the invitation.
SIMON: And how do these things work?
Mr. KRAPOHL: They are essentially computerized devices that record physiology, and they have an algorithm to analyze the physiological data, which produces a screen indicating that the person has reacted consistently to the relevant questions, that they've not reacted consistently to the relevant questions or you have an inconclusive result.
SIMON: Now when you talk about that data, what is that exactly, breaking into a sweat, rolling your eyes?
Mr. KRAPOHL: The devices uses two separate channels. One is called photoplethysmography, which its function is to record changes in capillary dilation of the tips of the fingers. And the second one records very subtle changes that occur at perspiration on the surface of the skin.
SIMON: And so what, soldiers will ask what kind of question?
Mr. KRAPOHL: There would be technical questions for baseline material, and then there would be relevant questions. It would really depend on what the application might be.
SIMON: Give us an example of a baseline question, if you could, because we, you know, we've all seen lie detectors used in movies.
Mr. KRAPOHL: It's not a lie detector. In fact, a real lie detector doesn't even exist. The closest thing I ever knew to a lie detector was my mother.
There might be test questions such as: Is today Saturday? There might be an overall question such as: Do you intend to be truthful to all of the questions here today?
SIMON: I've read that the Pentagon estimates these devices are between 82- and 90-percent accurate. Is that high enough?
Mr. KRAPOHL: As a screening device, it's actually quite good, as long as one understands that it's not a stand-alone technology. In fact, regulations prohibit its use as the final determination for anything. It's designed to help decision-makers make wiser decisions about the credibility of the individuals that they encounter.
SIMON: Now is there a bias built in, which would make a suspicious finding, a red light, if you please, more common?
Mr. KRAPOHL: The algorithm is risk aversive, which means that it's going to be better at detecting liars than it does truth-tellers, and the best analogy I've ever heard for that is the standard TB test that I think most of us have taken. If it turns red, then you have failed the screening test.
Now, there are lots of reasons for having that little spot on your arm turn red, some of them unrelated to TB. We think of the PCASS, to use maybe a crude analogy, is like the TB test. If the red shows up on your arm, that would indicate that you need more scrutiny, but it's not sufficient in and of itself to take any direct action.
SIMON: Are these devices, and I want to note they're called PCASS, are they being used anywhere in this country now in peacetime conditions, highway patrol or something?
Mr. KRAPOHL: The device was developed by the Department of Defense for a specific application, which is right now perhaps Iraq and Afghanistan. It is not authorized for use on U.S. citizens.
SIMON: What do you say perhaps to an Afghan citizen who would say how come you're willing to use them on us but not your own people?
Mr. KRAPOHL: The challenge has been there are simply not enough polygraph examiners in all of the federal government to meet the needs of our commanders overseas. If there were enough polygraph examiners, then we would be using that technology because, as one might expect, it has better accuracy.
But resources being what they are, rather than to leave our troops with no assistance, we've taken the path of providing at least some technical support for them with the understanding that it's designed only for that application and only in that setting.
SIMON: But again, what's the answer for someone from Afghanistan?
Mr. KRAPOHL: I guess I could give that person the very long answer and after, leave it up to him to determine whether what I'm saying is true or not.
SIMON: Well Mr. Krapohl, thanks so much.
Mr. KRAPOHL: You're very welcome.
SIMON: Donald Krapohl, special assistant to the director of the Defense Academy for Credibility Assessment in Columbia, South Carolina.