Is there a way to predict who is most likely to commit a crime in the future, and if so, what should the authorities do about it? In work that seems like science fiction, researchers have found that brain scans could help detect which convicts might get arrested for new crimes once they leave prison.
Forecasting crimes before they happen might seem to echo the film "Minority Report," where sci-fi advances enable police to capture would-be criminals before they carry out the acts they get arrested for. However, "I wouldn’t draw the parallel at all with our work," cautions neuroscientist Kent Kiehl at the nonprofit Mind Research Network and the University of New Mexico in Albuquerque. "We’re not saying anything is predetermined. This is just a slightly different way of judging risks."
Convicts often reoffend after leaving prison — one study found that more than four in 10 inmates nationwide return to state prison within three years of their release, a massive problem the Pew Center on the States called "the revolving door of America’s prisons."
"The best economic analyses suggests that about the same amount of money is spent on crime as on all healthcare in the United States, so about $2.3 trillion. That’s the annual burden of crime — police, prosecution, prison and so on — and recidivism is a huge part of that," Kiehl says.
Given the enormity of this problem, the authorities would like to find better ways to figure out which convicts are most likely to commit more crime when deciding whether they receive bail, jail, probation, and parole or involuntary commitment to a facility like a psychiatric institution. Such attempts to predict future antisocial behavior by offenders are already done using a range of factors, such as the age when they committed the crime and the kind of offense.
Searching for a better measuring stick
One of the strongest risk factors for recidivism is impulsivity. This is currently measured via personality exams and other means, but Kiehl and his colleagues wondered if they could directly test brains to judge how impulsive a person is and try to predict future antisocial behavior.
"When you’re trying to understand and make judicial decisions, you want the best science available in front of you," Kiehl says.
The researchers investigated 96 adult male convicts at two New Mexico state prisons. The inmates voluntarily completed a battery of psychological and behavioral tests. Then they got their brains scanned with a functional magnetic resonance imaging (fMRI) machine as they took a test measuring impulse control — their aim was to push a button when the letter "X" flashed on a video screen and resist pressing the button when the letter "K" was displayed.
After the convicts were released from prison, the researchers tracked them for four years from 2007 to 2010. Excluding minor parole and probation violations, about half were rearrested at least once.
The scientists focused on a brain region known as the anterior cingulate cortex, which is linked with impulse control. They found that offenders with relatively low activity in this area were 2.5 times as likely to get rearrested than offenders with high activity in this region, holding constant other known risk factors like alcohol or drug abuse or dependence. Their research, which was funded by the National Institute on Drug Abuse and the MacArthur Foundation’s Law and Neuroscience Project, was detailed online in the Proceedings of the National Academy of Sciences.
"We have a system of the brain linked with risk we can study now, and we can identify changes in that system that promote recidivism. That’s really exciting," Kiehl says.
Neuroscience is playing an increasingly large role in courtrooms. A study in the journal Science in 2012 even found that presenting scientific evidence about biological factors behind a defendant’s mental illness caused judges to be slightly more lenient in their sentencing decisions.
Kiehl says a lot more research is needed before brain scans for impulsivity find use in a legal setting. Others not associated with his work agree.
"This is a preliminary study," says lawyer and psychologist Stephen Morse at the University of Pennsylvania, who did not take part in the study. "More research is now needed to see if expensive scanning technology can do sufficiently better than conventional behavioral predictive measures."
Future use and questions
Morse says broader questions would also come into play should this technology start proving accurate. Though authorities already make predictions about the future behavior of prisoners, "if people were involuntarily scanned, constitutional and civil liberties issues could be raised," he says. "And if this was used not just in the context of prisoners, but for more general screening of people, there would be enormous constitutional and civil liberties issues."
Future research might also find ways convicts could change for the better. For instance, it might be possible to increase activity in this brain region and see if doing so reduces poor decision-making and recidivism.
"My hope would be to leverage this data to help develop treatments that can reduce violence and crime in society," Kiehl says. "Our goal is to make society a better place."
Questions remain over whether such differences in the brain actually reflect whether a person is more or less likely to commit a crime. Perhaps they are only signs of how likely a person will get caught.
"It could be that higher anterior cingulate cortex activity is indicative of people who are better at avoiding detection," Morse says. "Or people may be capable of countermeasures against these scans that might make it difficult to obtain valid predictive results. That’s why we need more studies."