Researchers Develop New Lie-Detecting Software

A team of researchers from the University of Michigan said they have found an advanced and improved way of detecting whether a person is lying or not. This new method is very much different from the polygraph test, said its makers.

The team created the software using videos from some high-stakes court cases. Explaining the working of the software, the researchers said the software uses a person’s words and gestures to detect behavioral patterns that may be different from the normal.

The software is presently undergoing a series of tests and evaluation processes. So far, the software has been 75% accurate in identifying a deceptive subject during testing.

The researchers during the study noted that the individual who moved their hands more frequently, tried to sound more confident and those who looked their questioners in the eye were more likely to be lying.

Using the machine learning technique, the team taught and trained the software on a set of 120 video clips from actual trils.

Rada Mihalcea, professor of computer science and engineering, said in a news release that people are poor lie detectors.

A major difference between the new software and the other lie-detecting equipment is that subjects don’t have to be hooked up to the equipment like in polygraph test, which measures heart and breath rates to detect if a person is lying.

Prof. Mihalcea said, “In laboratory experiments, it’s difficult to create setting that motivates people to truly lie. We can offer reward if people can lie well - pay them to convince another person something false is true. But in real world there is true motivation to deceive”.