Could technology detect workplace fibbers?

You’re on a TelePresence call with a potential supplier. They’re bidding on a vital contract. You need to know they can do the job on time. “Sure,” they say. Then you notice a signal flashing on your desktop. Suddenly you know it’s time to end the call… because they are lying.

This is fiction, for now. But the ability to detect lies with high-definition video already exists. And some experts believe it’s only a matter of time before it becomes integrated into everyday business communications tools.

In an article in IDG Connect last year, for example, DrPete Technology Experts CEO Dr. Peter Chadha said it would be relatively easy to embed lie detection into current technology.

See also: More co-bots work alongside human workers

“A simple smartphone recording, video conferencing footage, or Google Glass real-time video can be used to identify the physiological state of the CEO or IT director you’ve just had a business meeting with,” he said.

What’s more, it is possible such video-based methods could outperform the technology that has been used for lie detection since 1921. The polygraph has been the gold standard for lie detection for almost a century, but its efficacy is disputed.

Among the cases where a polygraph failed to yield accurate results is that of Aldrich Ames, a CIA analyst who was convicted of spying for the KGB, Russia’s former intelligence agency, in 1994. The CIA gave Ames a polygraph test in 1986 and again in 1991. He passed both times.

Concerned about the polygraph’s shortcomings, one of the leading experts in the technology, Dr. John Kircher, teamed up in 2003 with fellow University of Utah psychologist Dr. Douglas Hacker to investigate alternative approaches to lie detection, using audio-visual information.

One of these is computer voice stress analysis, which works by monitoring for changes in vocal pitch caused by anxiety. The technique works in up to 65 percent of cases. But the number of false positives is just as high.

See also: Cisco offices showcase the future of work

Kircher and Hacker decided to pursue a different line of investigation, based on eye movement. Perfected over nine years, this yielded much better results, correctly identifying 83 percent of deceptive individuals and 88 percent of truthful ones.

The duo, joined by a number of other eminent cognitive scientists, created a commercial ocular lie detection system called EyeDetect in 2014.

As well as being just as accurate as the highest polygraph accuracy of about 85 percent in screening tests, ocular-motor lie detection has a number of other advantages over traditional methods.

“EyeDetect tests are faster, 30 minutes compared to two hours; cost less than a polygraph; and do not require any sensors to be attached to the person tested,” says Todd Mickelsen, CEO and president of Converus, the company that owns and is commercializing EyeDetect.

EyeDetect can help companies uncover workplace problems such as theft, drug use, or bribery. The system has even been used in a commercial by the Spanish auto servicing group Midas, to show the firm’s car mechanics don’t lie to customers.

Currently, EyeDetect is delivered as a stand-alone system and is not integrated into third-party video technologies. And there are legal limits on how it can be used. U.S. law forbids the use of all lie detectors on private company employees, for instance.

Furthermore, the advent of easy-to-deploy lie detection systems raises social and ethical questions.

As a 2014 paper on lie detection using functional magnetic resonance imaging notes: “Lie detection … raises privacy issues that require societal control, much as we place limits on other practices that intrude on privacy, from DNAcollection to wire tapping.”

Being able to easily detect lies might avert “much crime and misbehavior,” say the authors. Clearly, society will have to decide if this is a benefit worth having in exchange for less privacy. But the technology to make it possible, at least, is just around the corner.

###

The contents or opinions in this feature are independent and may not necessarily represent the views of Cisco. They are offered in an effort to encourage continuing conversations on a broad range of innovative technology subjects. We welcome your comments and engagement.

We welcome the re-use, republication, and distribution of “The Network” content. Please credit us with the following information: Used with the permission of http://thenetwork.cisco.com/.