Can a machine tell when you're lying? New research suggests the answer is yes.
In a study of 40 cases, a new computer-based system developed by University of Buffalo researchers correctly identified liars more than 80 percent of the time, by analyzing facial clues and eye movement that indicate someone is telling a fib.
The study, which was inspired by the work of psychologists who study the human face for clues that someone is telling a high-stakes lie, found the UB computer had a better accuracy rate than expert human interrogators typically achieve in lie-detection judgment tests (65 percent).
"What we wanted to understand was whether there are signal changes emitted by people when they are lying, and can machines detect them?” said Ifeoma Nwogu, a research assistant professor at UB's Center for Unified Biometrics and Sensors who helped develop the system. “The answer was yes, and yes."
The research was peer-reviewed, published and presented the recent Conference on Automatic Face and Gesture Recognition.
For the study, researchers used the automated UB system to track how 132 test subjects moved their eyes in two distinct situations: during regular conversation and while fielding a question designed to prompt a lie.
People whose eye movements changed between the first and second situations were assumed to be lying, while those who maintained consistent eye movement were assumed to be telling the truth.
Nwogu said the findings suggest that computers may be able to assist even experienced lie-detection interrogators.
She noted that the technology is not foolproof: A very small percentage of subjects were excellent liars, maintaining their usual eye movement patterns as they lied.