In the 5 June 2008 issue (10 mb PDF) of the Fort Jackson Leader, Mike Glasch reports on current research at the Defense Academy for Credibility Assessment:
DACA researchers developing high-tech ways to detect deception
Mike A. Glasch
Leader StaffTubes strapped across the chest, sensors clipped to the fingertips and an inflated blood pressure cuff attached to the upper arm; all the while needles scratch on a roll of paper as an examiner grills his or her subject with a series of questions.
The needles and paper of the polygraph have been replaced by computer monitors, and soon the polygraph itself may be replaced.
Researchers at the Defense Academy for Credibility Assessment at Fort Jackson are studying new devices to replace the polygraph in a variety of situations — Soldiers in Baghdad investigating an IED explosion, Homeland Security officials screening potential terrorists at airports or FBI agents searching for a kidnap victim.
“There are no devices that measure deception. We are looking for devices that can measure physiological responses from which we can infer deception,” said DACA’s Director of Research, Dr. Troy Brown, who holds a doctorate degree in physiology. “We want to find a better way to tell if a subject is stressed by a certain question or image, which would lead to a more detailed investigation.”
Dr. Dean Pollina, Brown’s fellow researcher who holds a doctorate in experimental psychology, explained how researchers are focusing on ways to measure when a subject is stressed over what is presented to them.
“The changes in the body’s reaction (temperature, heart rate, muscle movement, etc.) doesn’t really indicate that they were deceptive,” Pollina said. “It’s much like a polygraph where all you can really say is that they were paying attention (or) they thought that was a salient thing that they just saw. It’s almost like a startled response, and that’s what we see.”
Devices they are developing include:
–A thermal facial imager that measures changes in heat in different areas of the face.
–An avatar to ask questions in a consistent manner; eliminating cultural differences and biases.
–A laser that can be pointed at the carotid artery to detect stress.
–Eye scanners that can tell where a person is looking on a computer screen, perhaps giving away guilty knowledge.“Right now they are all research grade instruments. We are trying to show that there is a very consistent physiological response,” Brown said. “Once we do that, we can use that in credibility assessment.”
Unlike the conventional polygraph, these devices incorporate remote tracking sensors so that the subject is never “hooked up” to the machine. Brown said that will allow for a more accurate results [sic] and a more natural exchange of information.”
“Whenever I hook up anything to your body, attach sensors to you, it immediately makes (you) much more uncomfortable,” he explained. “Also, it often limits the amount of time you can ask questions.
“In the case of a polygraph, where you have the blood pressure cuff inflated, there is a physical limitation there because it gets painful if left inflated. If we can get around using the cuff in the polygraph, or other sensors attached to the body, it allows us to ask more questions and have longer interviews. It makes the entire exchange more natural.”
Each of the new devices is designed to work individually. However, Brown is not ruling out combining them in the future.
“We’re pushing the envelope somewhat with the technology. Without any tools, what we have is chance, 50 percent, a coin flip. Anything above that helps us out,” he said. “We are still a long ways from putting any of these tools into the field. There is still a lot of testing to be done.”
Brown’s last-quoted remark seems to contradict reporter Mike Glasch’s opening statement that “soon the polygraph itself may be replaced.” The full article, at pp. 13-14 of the above-linked PDF, includes illustrations and additional information about the four devices mentioned.
Improving an existing technology is fine, so long as the underlying theory is valid.
The theory underlying “deception detection” is not valid. There are many reasons, besides deception, why a person might show a stress reaction to a question posed to them. INFERRING deception, due to a measurable stress response, is not the same as DETECTING deception.
Not all deception detection methods are based on detecting stress responses. For example, the eye scanner mentioned above does not detect stress. It detects specific patterns of eye movements that are related to visual content. If you recognize something you may look at it for an extra 100 ms. There are many ways one can look at differences without using the “stress” construct.