By: Kamala Kelkar
Judges in the future could tap straight into criminal brains and nip second offenders before they’ve had a chance to do it again, says the Obama administration.
Incriminating biomarkers could eventually be used by courts to predict recidivism and influence decisions about parole, bail and sentencing, finds the second volume of a report called Gray Matters released late last month by the president’s commission on bioethics.
The judicial system already uses questionable methods to adjust sentences based on the defendant’s criminal, psychological and social background, so there’s an allure to using brain scans for possibly more efficient and objective risk assessment profiles.
But the prospect of neuroprediction in the courtroom leads to a slew of ethical and moral questions. Should we assign longer sentences to a criminal functioning with what scientists say is a brain ripe for a second offense? Why not just let brain scans identify the most dangerous people and send them straight to jail before they’ve committed a crime?
“There’s a lot of motivation to literally get inside the heads [of criminals],” said Lisa Lee, the executive director of the commission. “What the commission was really concerned about was the careful and accurate use of neuroscience in the courtroom—given what’s on the line.”
If a judge uses unsettled science as evidence to determine the fate of the person on trial, it could send someone who is innocent to jail, free a dangerous criminal, or otherwise unfairly bias proceedings against a defendant.
“I think that the overwhelming concern is that we don’t rush to use things that will impede justice,” Lee said.
But she also said lawyers and judges are hoping neuroscience can increase the accuracy of behavioral assessments. The Bureau of Justice Statistics last year reported that 77 percent of 405,000 prisoners released in 2005 had been arrested again by 2010. It used age, race and gender to break down who among them were the most likely recidivists.
When a defendant is facing a judge, these statistical constructs are coupled with psychological evaluations to form a risk profile along with criminal and drug history and other factors such as affiliations with gangs.
Simin Shamji, a deputy public defender in San Francisco, said the city uses a survey of 135 questions to establish whether defendants are a low, medium or high risk for reoffending. Questions vary from “Do any current offenses involve family violence?” to “Do you live alone?” and “Is there much crime in your neighborhood?” Then an algorithm calculates the ranks and out pops the defendant’s risk category.
“Judges are getting these reports and relying on them. Do we know these are even accurate?” said Shamji, who also directs programs to deter reoffense. She said that judges interpret the results however they want, but she hopes they use them to help with early interventions or rehabilitation.
She said she was open to the idea of using brain scans or other biological data to help humanize risk assessments. “In the legal field, we’ve just sort of kept science out of it,” Shamji said.
Neuroprediction is premature, but it’s hard to find anyone in the neurolaw community who disagrees with funding research to help understand criminal behavior. When Gray Matters highlighted the potential legal uses of neuroprediction, it cited a handful of scientific papers written by some of the country’s leading researchers connecting neuroscience and criminology.
One of the papers, called “Neuroprediction of future rearrest,” was published in Proceedings of the National Academy of Sciences in 2013, and focuses on impulsivity because it is so closely related criminal behavior. Scientists used functional MRI scans, or fMRI, to measure the blood flow in a limbic region of the brain that is most often associated with impulse control, called the anterior cingulate cortex (ACC).
In the study, the research team took 96 inmates who were just about to be released and did fMRI scans while conducting impulse control tests designed to activate the ACC. The researchers did this by telling the inmates to push a button as quickly as possible when there was an “X” on the screen, but only when there was an “X.” Sixteen percent of the time, the scientists would flash a “K,” which would force inmates to control the impulse to hit the button again.
Roughly three years after their release, researchers checked in again with study subjects to see if they had returned to jail. Inmates who had low blood flow in the ACC during the study exercise were much more likely to reoffend.
Neuroethicists said that the ACC is just one of many regions in the brain and there’s no way one study can say for sure it is predicting recidivism. High blood flow in the ACC, for example, could instead predict a criminal’s ability to be clever and avoid getting caught, or it could be a marker of malnutrition or poverty, they said.
Eyal Aharoni at RAND Corporation, who was one of the authors of the study, agreed that there needs to be a lot more research before these methods are put to use. But he said his prediction rates were slightly better than the current methods of risk evaluations, and that there’s sufficient evidence that the judicial system needs to improve.
“By identifying who is at greatest risk of reoffending, it is possible to deliver necessary treatments and resources to those who need them most—before they return to crime,” he wrote in an email.
The bioethics report also stated that the courts could use similar brain imaging tools to identify psychopaths.
This idea piggybacks on the work of neuroscientist Kent Kiehl, who directs the Mind Research Network in New Mexico and co-authored the impulse paper with Aharoni. Kiehl wrote the book The Psychopath Whisperer based on his research of profiling psychopathic brains with fMRI and MRI scans. His nonprofit has scanned more than 4,000 inmates and written dozens of federally funded papers identifying neuropsychological trends.
Kiehl is perhaps most known in this community for identifying correlations between the interconnecting regions of the brain’s paralimbic system and psychopaths. This system involves a long list of players including the orbital frontal cortex, which may give insight about a person’s aggression levels, ability to plan, and ability to feel empathy. Since psychopaths are often violent and unfeeling, the theory goes, they are potentially the most dangerous. They’re also the most likely to commit a crime again, according to Kiehl.
He said his nonprofit is extending its work on neuroprediction and also replicating the original study. He also said from an ethics perspective brain scans are “no different” than any other method the judicial system uses to sketch risk assessments; they may even be better.