Deducing whether someone is lying or telling the truth is by no means a simple task.
“”On average, untrained humans can detect deceit 54 percent of the time, while trained interrogators can detect a liar around 70 percent of the time,”” said Kevin Moffitt, of the University of Arizona’s Center for the Management of Information at the Eller College of Management.
Moffitt and colleagues are currently analyzing the results of an experiment performed last fall, where researchers used data collected from electronic imaging devices to help investigators determine whether test subjects were lying or telling the truth.
The experiment, funded by a 16 million dollar grant to the UA from the Department of Homeland Security, involved a staged burglary.
Several test subjects, out of a total of 150, were instructed to steal a ring from an office room while the rest were used as a control. All of the subjects then underwent a Behavioral Analysis Interview conducted by a professional interrogator.
During the interview, an array of cameras and imaging devices constantly measured the subject’s blood pressure, pulse, blink-rate and retinal response.
“”We are trying to monitor the same readings that a traditional polygraph test gives investigators without any attachments,”” said Moffitt.
A high-speed camera slowed down the verbal and physical responses of the subjects while a thermal imaging camera showed increased levels of blood flow to the brain. “”Liars tend to blink more and have an increased flow of blood to the brain which is visible around the eyes using a thermal imaging device,”” said Moffitt.
Using Pupillometric Video Imaging Technology, Moffitt and his colleagues mapped the retinal responses that subjects had to objects they had seen before and those they had not. “”If you’ve seen something before you react to it differently,”” said Moffitt.
All of the subjects, regardless of whether they had taken the ring or not, were shown 30 images. Fifteen they were familiar with and fifteen they had never seen.
Five images from the crime scene were included that only the guilty subjects had viewed. “”The eye-tracking machine compared the eye movements of people who had seen the objects from the crime scene, to those who did not,”” said Moffitt. “”We then used this data to determine who had been in the office.””
By coupling the results from the imaging machines with human judgment, Moffitt and his fellow researchers hope to greatly improve the odds of detecting deceit.
“”Our experiments have ranged between 60-80 percent accurate,”” said Moffitt.
While the research is still in the analysis phase, the practical applications of using imaging devices to detect deceit are numerous.
“”If this technology is applicable, it could be used to increase border security and help prevent smuggling,”” said Larry Head, associate professor and Department Head of Systems and Industrial Engineering.
However, while the technology could potentially make our borders more secure, its application raises some morale questions as well.
“”Do we have the right to monitor heat coming off of someone or their blood pressure when they are crossing the border?”” said Moffitt. “”I don’t think so, I am big into privacy and think this technology should be used to aid investigators interviewing criminals and protecting the border.””
