"There's no such thing as 100 percent certainty," said Paul Ekman, a psychology professor from the University of California who pioneered the study of deceit-detection through fleeting facial expressions. "That's why I'm not eager to see my work be used in criminal court cases."
Advanced Lie-Detectors Within a Decade?
Kosslyn of Harvard shares these concerns. "More advanced lie-detectors could be out within the next decade, but the big question is, how will they be used?" he said.
Iowa-based neuroscientist Lawrence Farwell, however, is eager to see his "brain fingerprinting" work get into more courtrooms, convinced as he is that it has a near-perfect accuracy rate.
His method focuses on a specific electrical brain wave, called a P300, which activates when a person sees a familiar object. The subject wears a headband of electrodes and faces a computer screen, which flashes photos.
This technique provides a potential window into someone's past visual experience. If a person looks at random pictures of weapons, without activating a P300 wave, these objects are presumably unknown to him. But if the murder weapon is shown, and a P300 wave activates, then the person clearly has some experience with that weapon.
"This technique is used to see if they have the information stored in their brain or not," said Farwell, a Harvard graduate who now runs Brain Wave Science in Fairfield, Iowa. "All of this relates indirectly to lie detection."
Of course, for the P300 to be truly incriminating, the prosecutor would have to show that the tested person didn't see that murder weapon in some other innocent way, such as in media accounts or by being a bystander.
His "brain fingerprinting" helmet of electrodes is currently available within the CIA, Farwell said, though he doesn't know if or how often it's used.
However, Farwell knows some strategies for using P300 to detect moles. A US agent suspected of being a spy for Cuba, for instance, could be shown objects known only to Cuban undercover agents, something as simple as a job-related paper form or the "contact" person.
Farwell's lie-detection technique won a modest legal victory this past March, when an Iowa judge ruled there was enough scientific basis to admit "brain fingerprinting" results as evidence in the case of Terry Harrington, a convicted murderer trying to win a new trial.
Farwell showed that Harrington did not have a P300 wave when showed key parts of the crime scene, but did emit the P300 wave when shown scenes from his alibi, suggesting he was unfamiliar with the crime.
After reviewing evidence from all sides, however, the judge did not grant a new trial, though Harrington is appealing.
Harvard psychology professor Kosslyn also focuses on the brain in his study of deception, but he uses brain-scanning equipment to see what areas receive intense blood flow during questioning.
While his work has not yet been completed, preliminary results show that different regions of the brain light up when people tell the truth or lie. Further, he believes different regions are activated depending on the type of lie.
His data so far, he said, show the anterior cingulate, located near the front of the brain and associated with conflict resolution, is often activated during lies.
When lies are spontaneousmaking it up on the spot, as opposed to a fiction created over a period of timethe back part of the brain associated with visualization is also lit up.
Kosslyn said he believes that is related to the fact that the spontaneous liar has to visualize whether quick fibs make sense or not.
Focus on Smirks and Eye Flutters
Beyond the brain, tiny alterations in facial expressions, from smirks to eye flutters, are the focus of work at the Salk Institute for Biological Studies in La Jolla, California.
Piggy-backing on the work of psychologist Eckman, neuroscientist Terry Sejnowski is convinced that specific expressions suggest deception, though they are often imperceptible to the untrained eye. He is trying to link computers with video cameras, which would be able to record someone's face during testimony and see if it's truthful.
Not everyone is thrilled about this new generation of lie-detectors. Mark Zaid is a Washington attorney representing 11 people who say they were unfairly rejected for federal law-enforcement jobs when they failed a mandatory polygraph.
While some polygraph advocates put its accuracy rate at 90 percent, Zaid is among those who say the polygraph, in use since the 1920s, is hardly better than a coin-toss. He says future lie-detectors may be no different. "It's possible that new brainwave machines will say if someone is lying or not, but now, I don't buy it," he said.
Yet, even a Massachusetts plaintiff in Zaid's case, who works in local law-enforcement, said he keeps an open mind that someday a better lie-detector will come along.
This man, who asked to remain anonymous, said he feels today's polygraph is faulty and gave him a "false positive" when asked about past drug use. But he knows from his own job that the mere presence of a polygraph can be useful in questioning.
Back in 1995, for instance, police had no evidence linking Susan Smith to the drowning of her two young sons. While she claimed the boys were abducted by a carjacker, she had three polygraphs showing signs of deception. Police kept the focus on her. Later, she admitted she let the car roll into a South Carolina pond, and the bodies of her two boys were found in the water.
(c) 2001 The Boston Globe