Ann Arbor, Michigan – A group of researchers from the University of Michigan have developed a new and unique lie-detecting software which can sense dishonesty by examining the words and gestures made by the speaker. The interesting part of the software is that the subject does not need to know about is being analyzed to run the test.
The database for the software took typical lying behavior like moving hands excessively, scowling or grimacing, saying “um” more frequently, trying to sound more certain and looking to the questioners directly in the eyes. Then, the researchers developed the software by using machine-learning techniques and train the software to pick up these typical lying behaviors from footage of real life trials.
The researchers used 120 video films from 120 real court cases to train the software. The footage was analyzed by in order to identify different signs of a lie by comparing behaviors previously analyzed from real human testing, the software, then learned how to pick up on similar patterns.
This means that the new software doesn’t actually need to measure a subject’s pulse, breathing rate, or even touch them at all. It doesn’t need the subjects in the same room in order to analyze if they are indeed lying.
The results showed up to 75% accuracy at identifying if a person is lying or telling the truth, while humans could only tell 50% of the time.
The leader of the project, Rada Mihalcea, explained that there are clues that humans give naturally when they are lying, but we’re not paying close enough attention to notice them. Basically, people are poor lie detectors.
Researchers believe the software works so precisely giving the real-life aspect of the experiments relying on actual media coverage from genuine settings, rather than trying to recreate an artificial environment.
“In laboratory experiments, it’s difficult to create a setting that motivates people to truly lie. The stakes are not high enough. We can offer a reward if people can lie well – pay them to convince another person that something false is true. But in the real world there is true motivation to deceive,” said Prof. Mihalcea.
The group hopes to carry the research forward by using thermal imaging to integrate physiological factors, including heart rate, respiration rate, and body temperature fluctuations for better accuracy, all without even touching the subject. The other big step they’re working on is letting the computer classify gestures itself.
Source: The Christian Science Monitor