Functional MRI sniffs out liars and cheats

September 28, 2005

TV cops are unlikely to visit their local MR center when extracting confessions from suspected criminals. But the appearance of 3T scanners on detective programs could simply be a matter of time, following research demonstrating the accuracy of fMRI-based lie detection.

TV cops are unlikely to visit their local MR center when extracting confessions from suspected criminals. But the appearance of 3T scanners on detective programs could simply be a matter of time, following research demonstrating the accuracy of fMRI-based lie detection.

Group-based studies have already shown that truth-telling and lying cause different areas of the brain to light up on fMRI. Now a team from the University of Pennsylvania has shown that this neuropsychological phenomenon can be used to reveal individual liars. Data analysis has revealed the power of fMRI to identify truthful or false statements correctly almost 90% of the time.

"For fMRI to be used in practice for lie detection, you need to be able to put a subject in a scanner, ask 20 questions, and distinguish between truth and lies without knowing which is which in advance," said Dr. Daniel Langleben, an assistant professor of psychiatry at Penn. "That is what we have done."

Each study participant was given two playing cards. They were told to lie about holding one of the cards and admit the identity of the other. They were further advised that successful deception would be rewarded with $20. Questioning then proceeded while participants underwent blood oxygen level-dependent imaging in a clinical 3T MRI unit. Full details of the investigation, which involved members of the radiology and psychiatry departments, are to be published in Human Brain Mapping and NeuroImage.

Researchers found that the brain has to work harder to lie than to tell the truth. While lying, participants' frontal lobe, a region that regulates thinking, showed increased BOLD activity. No corresponding region showed increased activity when the person told the truth.

Media suggestions that fMRI could be used to detect terrorists are wide of the mark, according to Langleben. The use of brain imaging to spot a lawbreaker or predict a tendency toward criminal activity is best left to science fiction moviemakers, he said.

"But if you are asking me whether fMRI now can be used with some degree of reliability to detect deception or characterize concealed information, the answer is 'yes'. So in an indirect way, fMRI could be used for criminal investigations," he said.

Team members acknowledge that further work is needed to validate the technique. Quizzing a collection of English-speaking, white, right-handed, male undergraduate volunteers about a deck of cards is likely to provide very different data from that generated in a genuine criminal interrogation, where lying leads to far higher rewards or penalties. The impact of gender and aging or illness on brain response should also be considered, Langleben said.

"Without these extra steps, we cannot go out and start using it in real cases," he said.

The technique could also be used to test the validity of psychotherapy and psychoanalysis. An fMR scan of subjects believed to be suppressing knowledge of past events could determine whether their denial was a form of conscious deception, Langleben said.

For more information from the Diagnostic Imaging archives:

Functional imaging leads hunt for 'buy' trigger

Functional MRI speeds psych drug development

Vivid imaginations trigger false memories

Functional MRI makes broader inroads into clinical practice