A University of Buffalo researcher says she's developed a lie detector based on human eye movements that's more accurate than expert human interrogators.
In a study of 40 videotaped conversations, the system correctly identified whether interview subjects were lying or telling the truth 82.5 percent of the time.
That's a better accuracy rate than human experts can deliver. And while some proponents of polygraphs claim they're near-infallible, independent tests have shown them to be much the same as humans, at about 65 percent accuracy.
"What we wanted to understand was whether there are signal changes emitted by people when they are lying, and can machines detect them? The answer was yes, and yes," says research assistant professor Ifeoma Nwogu.
The automated system uses a statistical technique to model how people moved their eyes in two distinct situations: during regular conversation, and while fielding a question designed to prompt a lie.
People whose pattern of eye movements changed between the first and second scenario were assumed to be lying, while those who maintained consistent eye movement were assumed to be telling the truth.
Previous experiments in which human judges coded facial movements have found clear differences in eye contact at times when subjects told a high-stakes lie.
Nwogu concedes that the technology isn't foolproof: a very small percentage of subjects studied were excellent liars, maintaining their usual eye movement patterns as they lied.
She says the next step will be to try and develop automated systems that analyze body language as well as eye contact.