http://blog.wired.com/wiredscience/2009/03/noliemri.htmlMRI Lie Detection to Get First Day in Court
By Alexis Madrigal March 16, 2009 | 7:41:39 PM
Defense attorneys are for the first time submitting a controversial neurological lie-detection test as evidence in U.S. court.In an upcoming juvenile-sex-abuse case in San Diego, the defense is hoping to get an fMRI scan, which shows brain activity based on oxygen levels, admitted to prove the abuse didn't happen.
The technology is used widely in brain research, but hasn't been fully tested as a lie-detection method. To be admitted into California court, any technique has to be generally accepted within the scientific community.
The company that did the brain scan, No Lie MRI, claims their test is over 90 percent accurate, but some scientists and lawyers are skeptical.
"The studies so far have been very interesting. I think they deserve further research. But the technology is very new, with very little research support, and no studies done in realistic situations," Hank Greely, the head of the Center for Law and the Biosciences at Stanford, wrote in an e-mail to Wired.com.
Lie detection has tantalized lawyers since before the polygraph was invented in 1921, but the accuracy of the tests has always been in question. Greely noted that American courts and scientists have "85 years of experience with the polygraph" and a wealth of papers that have tried to describe its accuracy. Yet they aren't generally admissible in court, except in New Mexico.
Other attempts to spot deception using different brain signals continue, such as the EEG-based techniquedeveloped in India, where it has been used as evidence in court. And last year, attorneys tried to use fMRI evidence for chronic pain in a worker's compensation claim, but the case was settled out of court. The San Diego case will be the first time fMRI lie-detection evidence, if admitted, is used in a U.S. court.
According to Emily Murphy, a behavioral neuroscientist at the Stanford Center for Law and the Bioscienceswho first reported on the fMRI evidence, the case is a child protection hearing to determine if the minor should stay in the home of the custodial parent accused of sexual abuse.
Apparently, the accused parent hired No Lie MRI, headquartered in San Diego with a testing facility in Tarzana, California, to do a brain scan. The company's report says fMRI tests show the defendant's claim of innocence is not a lie.
The company declined to be interviewed for this story, but its founder and CEO, Joel Huizenga, spoke to Wired.com in September about the technology.
"This is the first time in human history that anybody has been able to tell if someone else is lying," he said.
Though the company's scientific board includes fMRI experts such as Christos Davatzikos, a radiologist at the University of Pennsylvania, some outside scientists and bioethicists question the reliability of the tests.
"Having studied all the published papers on fMRI-based lie detection, I personally wouldn't put any weight on it in any individual case. We just don't know enough about its accuracy in realistic situations," Greely said.
Laboratory studies using fMRI, which measures blood-oxygen levels in the brain, have suggested that when someone lies, the brain sends more blood to the ventrolateral area of the prefrontal cortex. In a very small number of studies, researchers have identified lying in study subjects with accuracy ranging from 76 percent to over 90 percent (pdf). But some scientists and lawyers like Greely doubt that those results will prove replicable outside the lab setting, and others say it just isn't ready yet.
"It's certainly something that is going to evolve and continue to get better and at some point, it will be ready for prime time. I'm just not sure it's really there right now," said John Vanmeter, a neurologist at Georgetown's Center for Functional and Molecular Imaging. "On the other hand, maybe it's good that it's going to start getting tested in the court system. It's really been just a theoretical thing until now."
No Lie MRI licensed its technology from psychiatrist Daniel Langleben of the University of Pennsylvania. Langleben, like the company, declined to be interviewed for this article, but offered a recent editorial he co-authored in the Journal of the American Academy of Psychiatry and the Law on the "future of forensic functional brain imaging."
From the editorial, it's clear that Langleben is a bit uneasy that his work has been commercially applied. He draws a clear distinction between "deception researchers" like himself and "the merchants of fMRI-based lie detection" and describes the "uneasy alliances between this industry and academia, brokered by university technology-commercialization departments."
Langleben has pushed for large-scale trials to determine the efficacy of fMRI-based deception-spotting. But in an interview conducted in late 2007, he doubted whether No Lie MRI and its competitor, Cephalos, had the resources to conduct the type of trials he wants.
"We need to run clinical trials with 200 to 300 people, so we can say, 'This is the accuracy of this test,'" Langleben told Wired.com. "But only two or three companies are trying to develop the technology. Do those companies have deep pockets? No. Do clinical trials cost a lot? Yes."
In September, Huizenga said the company was trying to get a grant for a study on a large group of people. "To date there really has been no study that has tried to optimize fMRI for lie detection," he said.
But even if the science behind a technology isn't fully established, Brooklyn Law School's Edward Cheng, who studies scientific evidence in legal proceedings, said it might still be appropriate to use it in the courtroom.
"Technology doesn't necessarily have to be bulletproof before it can come in, in court," Cheng.
He questioned whether society's traditional methods of lie detection, that is to say, inspection by human beings, is any more reliable than the new technology.
"It's not clear whether or not a somewhat reliable but foolproof fMRI machine is any worse than having a jury look at a witness," Cheng said. "It's always important to think about what the baseline is. If you want the status quo, fine, but in this case, the status quo might not be all that good."
But the question of whether Cheng's fMRI can be "somewhat reliable but foolproof" remains open.
Ed Vul, an fMRI researcher at the Kanwisher Lab at MIT, said that it was simply too easy for a suspect to make fMRI data of any type unusable.
"I don't think it can be either reliable or practical. It is very easy to corrupt fMRI data," Vul said. "The biggest difficulty is that it's very easy to make fMRI data unusable by moving a little, holding your breath, or even thinking about a bunch of random stuff."
A trained defendant might even be able to introduce bias into the fMRI data. In comparison with traditional lie-detection methods, fMRI appears more susceptible to gaming.
"So far as I can tell, there are many more reliable ways to corrupt data from an MRI machine than a classic polygraph machine," Vul said.
Elizabeth Phelps, a neuroscientist at New York University, agreed there is little evidence that fMRI is more reliable than previous lie-detection methods.
"When you build a model based on people in the laboratory, it may or may not be that applicable to someone who has practiced their lie over and over, or someone who has been accused of something," Phelps said. "I don't think that we have any standard of evidence that this data is going to be reliable in the way that the courts should be admitting."
All these theoretical considerations will be put to the test for the first time in a San Diego courtroom soon. Stanford's Murphy reported that the admissibility of the evidence in this particular case could rest on which scientific experts are allowed to comment on the evidence.
"The defense plans to claim fMRI-based lie detection (or “truth verification”) is accurate and generally accepted within the relevant scientific community in part by narrowly defining the relevant community as only those who research and develop fMRI-based lie detection," she wrote.
Murphy says that the relevant scientific community should be much larger, including a broader swath of neuroscientists, statisticians, and memory experts.
If the broader scientific community is included in the fact-finding, Greely doesn't expect the evidence to be admitted.
"In a case where the issues were fully explored with good expert witnesses on both sides, it is very hard for me to believe that a judge would admit the results of fMRI-based lie detection today," Greely said.
But that's not to say that lie-detection won't eventually find a place in the courts, as the science and ethics of brain scanning solidify.
Wired.com editor Betsy Mason contributed to this report.
· Woman Convicted of Child Abuse Hopes fMRI Can Prove Her Innocence
· Don't Even Think About Lying
· Brain Scanners Know Where You've Been
· This Is Your Brain on Hillary: Political Neuroscience Hits New Low
· Religion: Biological Accident, Adaptation — or Both
· India's Judges Overrule Scientists on 'Guilty Brain' Tech