Your spouse promises to "forsake all others." An investor says she'll share the profits with her partners. A parolee maintains he's done with his life of crime.
Can you trust them?
Scientists from the University of Zurich say a specialized form of MRI may be able to answer that question.
Thomas Baumgartner and his team have used functional magnetic resonance imaging – fMRI—to tell whether someone intends to keep a promise.
In their experiments, volunteers pretended to be investors and financiers, and the scientists were able to predict which subjects would keep their promises to hand over money and which wouldn't. Those who intended not to keep their word registered increased activity in the areas of the brain associated with emotion and control.
Mr. Baumgartner, who holds a doctorate in neuroscience, says he thinks the discovery could help psychiatrists decide whether criminals should be paroled by measuring the likelihood they will break a promise not to re-offend. The technology also could be used in the investment market or in the detection of an unfaithful spouse.
He says he doesn't think it will be used that way anytime soon, however.
"For purposes like that, the technology is too expensive and the analysis much too complicated," he says.
In addition, questions have been raised about the reliability of fMRI as a measure of neural activity. FMRI determines which brain cells are active by responding to changes in oxygenated blood. This usually offers a good indication of neural activity, but it isn't always successful. Studies have shown that not all blood changes in the brain are related to neural activity.
A Dartmouth College team led by neuroscientist Craig Bennett used fMRI on a salmon and found that its brain registered humanlike emotional responses when shown a variety of stimulatory pictures. There was just one problem, however: The salmon was dead.
In light of results like this, Mr. Bennett says he thinks fMRI research should be regulated carefully.
"Due to the sheer amount of data we are working with, it is possible to find relationships completely by chance," he says. "While the possibilities are intriguing, these methods aren't ready yet for widespread application."
Mr. Bennett's observations have particular relevance in the United States, where fMRI already has been used in criminal cases. FMRI evidence was used in the sentencing phase of the trial of an Illinois man convicted of raping and killing a 10-year-old girl. Brian Dugan was sentenced to death, and it is thought the fMRI evidence played a key part in the jury's deliberations.
In a sexual abuse case earlier this year in San Diego, the defense wanted to use fMRI evidence to prove the defendant was telling the truth. Prominent scientists spoke out against it, however, as they said they didn't think the technology had reached a sufficient level of accuracy. As a result, the request to use fMRI was withdrawn.
Some fear that fMRI use could lead to a situation like that explored in the 2002 science-fiction film "Minority Report," in which people are arrested for crimes they haven't yet committed.
Peter Schaber, professor of applied ethics at the University of Zurich, is wary of placing too much faith in these kinds of techniques.
"I think that all technologies can be abused, and the same holds for fMRI," he says. He says he is doubtful whether it will ever be used for predictive purposes, but if it is, he says a clear line should be drawn between what can be established with certainty and what is merely probable.
"Someone's intentions can't be measured in the way you see a cell structure," he points out. "If 80 percent of people with a certain type of brain activity committed a crime later on in their life, that's still only a probability. You can't say, 'This person will act as most of the others did.' "
Dr. Daniel D. Langleben, a psychiatry professor at the University of Pennsylvania who has written extensively about fMRI, says he thinks it isn't the technology itself that needs refinement, but rather the way it is used.
"FMRI is just a tool to measure brain activity," he says. "Of the many such available tools, fMRI is one of the easiest to use and one of the hardest to properly analyze."
He says he thinks fMRI research has become disconnected from clinical practice and that if more results were tested on a practical level, the wheat quickly would be separated from the chaff.
"What is needed are properly controlled experiments and honest scientists who are more critical about themselves than anyone else would be," Dr. Langleben maintains.
© 2009
Copyright 2009 
All Rights Reserved
