Can brain mapping scans detect if we are cheating? There are scientists who think so and even want to integrate the scans into police investigations and to "reduce risks in romantic meetings"
By Gary Stix
Sean A. Spence, a professor at the University of Sheffield Medical School in England, performed brain scans in 2007 that allegedly showed that a woman convicted of poisoning a child in her care was telling the truth when she denied committing the crime. This study and two other studies on cheating conducted by the Sheffield research group were funded by Quickfire Media, a British Channel 4 television company. Videos about the work of the researchers were broadcast on the channel as part of the TV program "Laboratory of Lies". The study of the suspected woman's brain was subsequently published in the European Scientific Journal of Psychiatry.
Brain function imaging using magnetic resonance imaging (fMRI) claims to detect lies based on what is happening inside the brain itself and not based on measures of anxiety, such as changes in heart rate, blood pressure or breathing, used in the polygraph machine. Apart from the hundreds of thousands of viewers of the program, the fMRI method also aroused curiosity among entrepreneurs. Two companies, Sepos from Pepperell, Massachusetts, and No Lie MRI from Tarzana, California, claim they can determine with at least 90% accuracy if we are telling the truth. No Lie MRI, whose name evokes a feeling of casualness, like a dental clinic in a shopping mall, claims that the method can be used even "to reduce risks during romantic meetings".
Many neurobiologists and jurists doubt these claims, and some even wonder if brain imaging for the purpose of detecting lies is even intended for any use other than research on the nature of lying and the brain.
An fMRI device monitors blood flow to active areas of the brain. The assumption behind lie detection is that the brain must exert more effort when we lie, and the areas that perform the strenuous action therefore consume more blood. Such areas are "illuminated" during the scan. During the study of lies, mainly areas involved in decision-making are illuminated.
To assess how the fMRI method and other discoveries in neuroscience affect the law, the MacArthur Foundation allocated ten million dollars in 2007 for three-year funding of a new project: "Law and Neurobiology". Part of the funding will be allocated to establish benchmarks that will define the accuracy and reliability of lie detectors that rely on fMRI and other imaging methods. "I believe that it is impossible to trust the results obtained from the technology used today," says Marcus Reichel, a neurobiologist at the Washington University School of Medicine in St. Louis, who heads the project's research group dealing with lie detection. "But it is not impossible to launch a research program that will determine the feasibility of this."
An important review article published in 2007 by Henry T. Greely from Stanford University, and Judy Eales, now at the University of British Columbia, in the American Journal of Law and Medicine, examines the shortcomings of the studies conducted so far, and what needs to be done to advance the technology. The two researchers found that the studies on lie detection (fewer than 20 studies in total) failed to demonstrate that fMRI "is effective as a real-world lie detector at any level of accuracy."
Most studies examined groups, not individuals. Others were conducted only once and were not repeated. The subjects were young and healthy adults, so it is not known what results would be obtained if they tested people taking medications that affect blood pressure or one of their arteries is blocked. Also, the two reviewers doubted the uniqueness of the reaction of the areas that "lit up". They noticed that these areas are active in a wide variety of cognitive tasks, such as memory, self-control and self-awareness.
The biggest challenge of all, which the "Law and Neurobiology" project has already allocated funding for its investigation, is how to reduce the artificiality of the tests. It is not clear if it is a lie in the answer to the question "Are you holding in your hand a 7-face card?" Activates the same cortical areas that would be activated in a lie about robbing a corner store. In fact, it is possible that the most realistic studies so far are precisely the studies conducted for the TV show The Lab of Lies.
The two companies marketing the technology are not waiting for more data. In an attempt to gain recognition in the courts, the Sepos company provides free scans to people who claim innocence if they meet certain conditions. The approval of the scans as legally admissible evidence may open the way to a huge and profitable market. "We may have many failures before we get to see a court," says Stephen Leiken, CEO of Sepos. He claims that the accuracy level of the technology is 97%, and that more than 100 people whose brains were scanned by Sepos provided data that solved a significant part of the problems raised by Greeley and Eales.
But until formal clinical trials prove that the method meets acceptable standards of safety and efficacy, Greeley and Eales call for non-research uses to be banned. The tests planned for the approval of the technology by the authorities hint at the technical challenges. Actors, professional poker players and sociopaths will be compared to ordinary people. Religious people will be scanned after secular people. The tests will take into account social conditions. It would be necessary to compare white lies, such as "no, really, dinner was wonderful", to lies about sexual flutters, to make sure that the brain reacts in the same way.
The possibility of abusing the technology dictates the need for caution. "The danger is that we will negatively affect people's lives because of mistakes in technology," says Greeley. "The fear is that science will be snookered because this well-publicized use of brain imaging will go wrong." Reviewing the long and controversial history of the polygraph, moderation seems to be the wisest way to use this new diagnostic tool that examines an essential quality that affects our social actions.
