In 2000, a 40-year-old schoolteacher from Charlottesville, Virginia (we’ll call him Alex) began feeling something he never had before: attraction to children. He began viewing online child pornography and quickly developed an obsession, downloading hundreds of files. When Alex eventually made sexual advances toward his prepubescent stepdaughter, his wife had him legally evicted from their home. Found guilty of child molestation, he was medicated for pedophilia and committed to a 12-step Sexaholics Anonymous rehabilitation program, which he was promptly expelled from for inappropriately asking women there for sex. He then faced jail time.

As Alex’s day in court approached, he began experiencing intense headaches. The evening before his sentencing, he checked himself into the emergency room, unable to stand the pain. Tests showed he was having balance problems and was unable to write. When doctors performed an MRI, the scan revealed an egg-sized tumor in the orbitofrontal cortex—a brain region linked to judgment, impulse control, and social behavior. Once surgeons removed the tumor, Alex’s troubling sexual behavior vanished as suddenly as it had appeared.

Alex’s story illustrates a truth of criminal activity that routinely goes overlooked: malicious behavior can be as much a product of our anatomical as it is a product of our own volition. In Alex’s case, for instance, the tumor itself was demonstrably the source of his behavior. Should that affect how the law treats him?

To most people, the answer would likely be “yes.” Alex later said that though he tried to resist his urges, the tumor impaired his self-restraint and compelled him to do reprehensible things. It was, in a sense, the tumor’s fault, not Alex’s.

In other cases, the blamelessness of the perpetrator is even more apparent. In the summer of 2008, police arrived at the home of Brian Thomas in Aberporth, Wales, after Thomas called the station in the middle of the night. Horrified, he told the operator that his wife, who had been sleeping next to him, was inexplicably dead. It soon became apparent that Thomas suffered from night terrors and had unknowingly strangled his wife in his sleep, leading the prosecution dropped the murder charges.

Not only did Thomas, like Alex, suffer from a powerful biological impairment, but he was not even conscious when he committed the crime. Most people would likely agree that Thomas, more than just being free of blame, was actually deserving of sympathy for the tragic death of his wife—a death that was not his own fault.

In both cases, the perpetrators clearly should not bear full legal or moral responsibility for their actions. Yet what is it, exactly, that separates them from other violent criminals? We intuitively place them in a different category than “regular” criminals because the biological roots of their behavior are clear-cut and easily labeled. It’s easy to understand how a tumor—not to mention sleep—could inhibit a person’s self-control and thereby relieve them of guilt. But as David Eagleman, a neuroscientist at Baylor College of Medicine, argues in his essay “The Brain on Trial,” developments in genetics, neuroscience, and cognitive psychology are uncovering the biological underpinnings of an ever-widening array of aberrant behaviors. He writes that currently, “we can detect only large-scale problems, but within the coming decades, we will be able to detect patterns at unimaginably small levels of the microcircuitry that correlate with behavioral problems.” Is it possible that the only reason we think Alex is uniquely guiltless is that a tumor is a “large-scale problem”—that it happens to be identifiable with the relatively crude technology we have?

The research demonstrating the neuroanatomical roots of criminal behavior is extensive. In 2009, researchers for the Canadian Journal of Psychiatry Adrian Raine, Yaling Yang, and Robert Schug published research on the brain scans of 27 psychopaths and found that their amygdalae—the region associated with emotional capacity that’s critical to moral conduct—were, on average, 18 percent smaller than those of normal people. “The amygdala is the seat of emotion,” said Raine at a 2011 conference in Washington, D.C. “Psychopaths lack emotion. They lack empathy, remorse and guilt.” By contrast, studies on “hot-blooded” criminals—criminals whose violent acts were more spontaneous—revealed their subjects to have less activity in the prefrontal cortex, the higher-order brain region responsible for problem-solving and deliberate thought.

Another study by the Mayo Clinic took brain scans of 21 people with antisocial personality disorder, a condition common among convicted criminals. The scans showed, on average, an 18 percent reduction compared to healthy brains in the size of the middle frontal gyrus, a section of the brain’s frontal lobe related to executive functions.

Research shows such neuroanatomical causes of criminal behavior have a genetic basis. In 1984, psychopathology researcher Sarnoff Mednick found that children in Denmark who had been adopted from parents with a criminal record were statistically more likely to commit crimes in adulthood than were other adopted kids. In other words, they inherited a propensity for crime that had nothing to do with their environment. Other studies have shown that identical twins, people who have the exact same genes, are much more similar in terms of their proclivity to crime than are fraternal twins, who share only half of their genes. Environmental factors like neglect and abuse in childhood can also predict an individual’s criminality.

So is all of this research making any headway in the judicial system? Joel Fleming, an attorney at Block & Leviton in Boston, said: “It’s hard to imagine, politically speaking, that many politicians are going to be eager to start putting laws in place that reduce the sentences for violent crimes.” Yet other areas of law—notably narcotics—are beginning to reflect “a more nuanced understanding of neuroscience,” he added, and the various factors that influence people’s behavior. Indeed, the number of judicial opinions mentioning neuroscience evidence tripled between 2005 and 2011, from roughly 100 to more than 300, according to research conducted at Duke University.

And examples of this “more nuanced understanding” have, in fact, begun to appear in cases of violent crime in the last few decades with fascinating results. Take the case of Donta Page. In 1999, Page broke into the home of a young woman named Peyton Tuthill in Denver, Colorado. He tied Tuthill up, raped her and stabbed her to death with a kitchen knife. Page was charged with first-degree murder and considered for the death penalty. Yet Raine, who was serving as an expert witness on the case, conducted brain scans on Page and found an unusual lack of activity in the ventral prefrontal cortex, the brain region associated with emotional regulation and impulse control. Page’s file showed a glimpse at what had clearly been a nightmarish life story: as a child growing up in abject poverty he had suffered poor nutrition, severe parental neglect, numerous instances of physical and sexual abuse, and major head trauma. By the age of 18, Page had been referred for psychological treatment no less than 19 times and had never received treatment. A panel of three judges agreed these conditions mitigated his responsibility enough to avoid execution.
Page’s mental development was stunted in measurable ways—that much is clear—but was the judiciary panel right to consider these grounds for decreased blameworthiness? Revolutionary as the Page decision seems, the justice system actually includes a long-standing precedent for mitigated responsibility in the case of criminals with inferior psychological maturity: juvenile law. In the 2012 case Miller v. Alabama, the Supreme Court ruled that sentences of life in prison without the possibility of parole are unconstitutional for juvenile offenders. “Our decision rests not only on common sense but on science and social science,” they wrote in the closing statement. “Developments in psychology and brain science continue to show fundamental differences between juvenile and adult minds.” At the highest level, the law cites “fundamental differences” in mental maturity as the reason to scale down the punishment for a young person. Is it possible this principle should apply to adult criminals with cognitive deficits as well? If so, where do we draw the line?
Many reformers would like to see a more nuanced system with a greater focus on individualized sentencing and treatment. Eagleman takes it a step further. He believes no difference exists at all in terms of blameworthiness among criminals like Page, Alex, Thomas, and “regular” convicts with no conventional mental health defense. Eagleman argues that sooner or later, neuroscience will make the usefulness of blame as a legal paradigm obsolete entirely. Everything has a biological reason—we just haven’t discovered it yet. He writes, “As we become more skilled at specifying how behavior results from the microscopic details of the brain, more defense lawyers will point to biological mitigators of guilt.” The way we make choices is inseparable from the neural architecture of our brains—systems we have no control over. “The more we learn, the more the seemingly simple concept of blameworthiness becomes complicated,” he adds.
Others aren’t so sure. Sam Sommers, a social psychology professor at Tufts, said: “Human behavior is complicated and multiply determined, we don’t really have the ability to reduce all human behaviors to a set of specific biological predictors.” Megan Krench, a PhD candidate in neuroscience at MIT, believes it’s unlikely that technology will advance to the point where it can be used reliably as legal evidence in the foreseeable future. “We can always find something that makes your brain a little different,” she told me, “but I don’t think we’re ever going to be able to point to that one thing and say, ‘This absolves you of guilt.’ ”
Steve Fleming, a cognitive neuroscientist at NYU, makes an interesting case that knowing more about our brains will actually make us more responsible for our behavior. Like a diabetic who knows to take a daily insulin shot, we’ll be obliged to take the necessary measures to prevent disaster. “A link between brain and behavior is not enough to push responsibility out of the courtroom,” he writes in an article for AEON magazine. Better technology and treatment infrastructure will eventually enable this sort of potentially life-saving self-knowledge.
Clearly neuroscience research is in the beginning stages of making inroads into the criminal justice system. That brings us to another question—how should the law incorporate such research?

The American justice system is predicated on retributive justice, or the theory that criminals should suffer for their actions and the suffering should be proportionate to the crime. Sentencing is largely punitive—punishment for punishment’s sake—and rehabilitation remains a secondary concern. Whatever its merits, such a system does not leave room for a nuanced approach to the question of an individual’s culpability, and very few cases involving violent crime focus on the defendant’s cognitive to the extent the Donta Page case did.
A healthier system, said Joel Fleming, would not draw such a bright line between the tiny number of legally insane convicts and everyone else. “If you’re not so crazy that you don’t meet the test for ‘not guilty by reason of insanity,’ there’s no middle ground where it can recognize less severe forms of mental illness or addiction,” he told me. “A lot of people are struggling with mental illness or addiction or both, and that’s not really something the system’s built to handle.” A stronger focus on rehabilitation would benefit not only criminals, but society as a whole—both of which are considerations more important than mere punishment, he argues. One place to start? Removing mandatory minimum sentences for many crimes. “I think the evidence bears out [that] people are equally deterred by a 20 year sentence as they are by a 40 year sentence,” he said. In many cases, holding an inmate indefinitely serves no one, and actually costs the state thousands.

Raine imagines using neuroscientific data to predict which criminals will offend and reoffend, allowing the law to prevent crime at the source. With neuroscience, he writes in the Wall Street Journal, “We can in theory assess any individual in society for his or her criminal propensity—making it possible to get ahead of the problem by stopping crime before it starts.” Eagleman makes a similar case, arguing that if blame is off the table, we should turn the focus to a convict’s probability of future recidivism based on their neurological makeup. “Some people will need to be taken off the streets for a longer time (even a lifetime), because their likelihood of re-offense is high,” he writes. “Others, because of differences in neural constitution, are less likely to recidivate, and so can be released sooner.”
Such a system might benefit society most directly, assuming such methods for predicting re-offense will be accurate enough to justify them. But would such an approach be ethical? Krench argues that predicting recidivism is important, but we should be wary of over-relying on neuroscientific research in the courtroom. Brain scans are excellent tools for the purpose of laboratory research, allowing us to examine patterns among large groups of subjects, but are much less meaningful on an individual level—especially when the stakes are this high. “If you say, ‘I have this brain area that can predict 99.99 percent whether or not you’re going to commit a murder in the next ten years,’ that’s good for lab research,” she said, “But scientific evidence isn’t fact. You’re never going to be able to completely eliminate that shadow of doubt, that shadow of human error, which brings in real problems if you’re using that to convict someone.” Using someone’s brain scans against them could violate their constitutional right against self-incrimination.
The law already uses predictors of recidivism (mainly the subjective analyses of prison psychologists) against one group of criminals: sex offenders. The results have been morally disastrous. In 2006, Congress passed the Adam Walsh Child Protection and Safety Act, which allows the federal Bureau of Prisons to hold inmates in prison past their release date if a psychologist believes they will have “serious difficulty in refraining from sexually violent conduct or child molestation if released.” Yet sex offenders actually reoffend at lower rates than nearly all other criminals (the exception being murderers). According to the American Psychiatric Association, such interment is unethical and legally and medically baseless. “Confinement without a reasonable prospect of beneficial treatment of the underlying disorder is nothing more than preventative detention,” their 1999 report concluded.
Not only would such a radical shift in focus from culpability to recidivism for all violent crimes be ethically dubious, but it would perhaps not even be socially beneficial. The idea rests on the assumption that, if our criminal behavior is determined by a combination of genetic and environmental factors, we cannot be held to blame for it. But how would such a change itself influence behavior?
In his essay “What Neuroscience Will Tell Us About Moral Responsibility,” philosopher and cognitive scientist Daniel Dennett compares the function of legal accountability (i.e. punishment) to that of red cards in soccer. “One will seek far and wide for a football player or fan who thinks that the whole practice of issuing yellow cards and red cards and calling fouls should be abandoned, because it is too ‘punitive’; because it deals with human beings who could never really deserve anything,” he writes. “It is quite clear . . . that strict rules don’t just improve a game; they make it possible.” Legal accountability serves a similar function: its presence changes the way anyone with any degree of self-control will act. Hence the question of free will is actually irrelevant. If criminal behavior results from the confluence of nature and nurture, of genetics and environment, than the knowledge that one will be held accountable for a crime is itself effectively part of the “environment.”
The purpose neuroscience should now serve is not to remove the question of culpability, but to identify with more precision who should be subject to it. Moral responsibility is not mutually incompatible with the determinative chain of causation leading up to a decision that stems from whatever genetic and circumstantial factors neuroscience research is revealing. Most would agree that a sleepwalker should not be held accountable for his crimes. But that does not mean all criminals should be similarly guilt-free, because some types of crime are, in fact, preventable. As crucial as a greater emphasis on treatment may be, “my brain made me do it” cannot be a catch-all excuse.
“The way the system operates is to set up categories that at some level are essentially always going to be arbitrary,” Myles Jacobson, a criminal defense attorney, told me. The justice system draws a strict line between those eligible for mitigated legal accountability (whether under juvenile law or an insanity defense) and everyone else, and such immutable categories poorly reflect a reality neuroscience is proving to be much more complex. This is why the goal of neuroscientific research should be to make those categories less arbitrary—to determine which types of criminal will respond to the threat of punishment. A greater understanding of mental states does not, and will not, eliminate the value of blame, but it may change the way we implement it.