The Search for Truth
Can Brain Imaging
Be the Perfect
In June 2008, a judge in India accepted a brain scan as evidence in a murder trial. The scan was not performed on the victim to show or prove injury; instead, the scan was of the defendant, a woman accused of killing her fiancé. Although the woman denied involvement in the murder, according to the prosecution her brain activity during interrogation indicated a level of knowledge about the crime that only the killer could possess, and she was sentenced to life in prison. These types of tests have not yet made their way into U.S. courts, but the idea of using brain scans in lie detection is becoming increasingly common. Private companies, like No Lie MRI in California, offer functional Magnetic Resonance Imaging (fMRI) tests for corporations screening new employees, lawyers hoping to validate statements made in court, and individuals looking to reduce their risk in dating. But according to a recent report from the National Research Council, the rush to use brain scans in lie detection, particularly in legal cases with significant consequences, could be premature.
The report identified current and emerging fields of cognitive neuroscience research with practical applications, including the use of neuropharmacology in regulating cognition, prostheses in enhancing performance, and computer modeling of the human brain. Neuroimaging, a technology regularly used by doctors to scan the brain and monitor function in real-time, was discussed in the report primarily in terms of its non-health-related applications, such as determining what an individual is thinking or intending. Although brain scans have been a boon to physicians in detecting and treating neural abnormalities, their utility in measuring psychological states is still a matter of debate.
According to the committee that wrote the report, brain scans can only measure correlates of mental states like anger, fear, or deception. They can not identify the mental state itself. The fMRI measures blood flow for instance, which is an indicator of increased neural activity. Based upon the parts of the brain that show increased activity, researchers assume the mental state of the individual.
But brain scans encounter the same problem as polygraphs: no physiological indicator, or neural activity pattern, exists that has a one-to-one correspondence with mental state. A true lie detector requires an indicator that appears every time a person is being deceptive, and which appears only during deception. Researchers find that although there is some consistency in neural activity during deception, the same neurons fire during other mental and emotional states as well.
Furthermore, research on the use of brain scans in lie detection has mainly been done in the controlled environment of the lab, and researchers don't know how these findings relate to the complex and emotional lies that people tell in real life. For example, a spouse who was tempted to cheat but never acted could have enough guilt over the temptation that their brain activity patterns would be the same as someone who had cheated.
Despite the technical hurdles, some experts feel that tests such as fMRI and functional near-infrared spectroscopy hold promise in the field of lie detection. "The committee was split," said Kit Green, chair of the committee. "Those focusing on the physiological aspects [of lie detection] tend to believe that lie detectors will eventually work. Those focusing on the psychological factors were less likely to believe this."
Because truth can be subjective, influenced by experiences like religion and culture, the most fMRI could reliably show is that someone believes what they are saying is true or false. How that belief relates to the facts of a court case, or the suspicions of a spouse, is harder to determine. "If we can't yet define a lie," explained Green, "then we can't define a lie detector." -- Rebecca Alvania
Emerging Cognitive Neuroscience and Related Technologies. Committee on Military and Intelligence Methodology for Emergent Neurophysiological and Cognitive/Neural Science Research in the Next Two Decades, Division on Engineering and Physical Sciences and Division of Behavioral and Social Sciences and Education (2008, 214 pp.; ISBN 0-309-11894-8; available from the National Academies Press, tel. 1-800-624-6242; $49.00 plus $4.50 shipping for single copies).
The committee was chaired by Christopher (Kit) Green, assistant dean for Asia Pacific of the Wayne State School of Medicine in Beijing. The study was funded by the U.S. Defense Intelligence Agency.