Your lying eyes

Margaret Talbot writes in The New Yorker about the controversial use of functional magnetic resonance imaging (fMRI) to determine when a person is lying. Brain scans, Talbot writes, represent the latest attempt by humans to find the perfect lie detector. In 2003, Carnegie Mellon Statistics Professor Stephen Fienberg chaired a National Research Council committee that determined the polygraph is too unreliable in distinguishing between those telling lies and those telling the truth to justify its continued use as a tool for screening federal employees. Nonetheless, the polygraph continues to be widely used, even though, as Talbot notes, no one has ever tested its accuracy against a "placebo"--a fake machine that might have the same intimidating effect on test-takers as the polygraph.

Unfortunately, other methods of lie detection suffer from many of the same problems, according to Fienberg. "These alternatives are not only still unproven scientifically, they raise major issues about privacy rights that no one has addressed," Fienberg says.

If you are a member of the media who is interested in talking to Fienberg about this issue, contact me at jpotts@andrew.cmu.edu. 

Jonathan Potts