Neuroscience and Sentencing: Diminished culpability and capacity for change

I originally wrote this post on the Legal Aggregate on December 1, 2009:

The latest edition of the Stanford Lawyer magazine features an article on neuroscience and law, focusing on the likelihood and implications of admitting evidence about a defendant’s brain function. Although it is uncertain whether brain imaging data will ever be admitted as a tool in assessing the guilt of a defendant, such data has already found its way into the sentencing phase of death penalty cases. As Stanford Law School Professor Robert Weisberg says in the Stanford Lawyer article, the defendant has “a constitutional right to offer just about anything that could be characterized as mitigating evidence.”

For example, neuroscientist Kent Kiehl recently testified as an expert witness for the defense in the sentencing hearing for Brian Dugan, a man who pled guilty to the 1983 rape and murder of a ten-year old girl. The defense argued that Dugan suffers from psychopathy, a psychiatric disorder typified by antisocial behavior, impulsivity, and lack of remorse, which made it difficult for him to control his behavior. As a result, the defense argued, Dugan is less culpable for his criminal behavior, and his disorder should be considered a mitigating factor. Kiehl testified that based on functional magnetic resonance imaging (fMRI) tests and a diagnostic checklist, Dugan showed abnormal brain functioning and responses similar to other psychopaths Kiehl has tested. Although data from other brain imaging techniques, such as PET and SPECT, have been used in court previously, Kiehl’s testimony may be the first instance in which fMRI data was admitted in a criminal case.

Prosecution witness, psychiatrist Jonathan Brodie, criticized the defense’s use of brain imaging. Brodie argued that it was not possible to draw conclusions about psychopathic behavior based on fMRI data, much less about behavior that took place 26 years ago. Brodie and Kiehl’s different interpretations of the fMRI data reflect widespread disagreement about the degree to which one can draw conclusions about behavior based on neuroimaging, which is complicated for a number of reasons. Behaviors performed at the behest of a researcher during fMRI tests are necessarily artificial, and it is a challenge to show that a functional abnormality measured during the scan is related to complex behavioral traits, like psychopathy. Furthermore, it is difficult to determine whether any supposed brain abnormalities preceded behavior, or whether the reverse is true. For example, living a life congruent with the hallmarks of psychopathy and being incarcerated for much of the last 26 years might have produced the altered brain function. After the jury deliberated for ten hours, Dugan was sentenced to death. For Dugan, evidence of abnormal brain functioning was not sufficient to reduce his sentence.

In this case, the defense admitted fMRI evidence of psychopathy to show diminished culpability, but the same evidence could cast doubt on a defendant’s capacity to change his behavior. There is no known treatment for psychopathy, and psychopaths are highly likely to recidivate. According to Kiehl in a 2008 Science Magazine article, “psychopathy is the single best predictor of violent recidivism.” The article cites the statistic that “more than 80% of incarcerated psychopaths who are released from prison commit another crime, usually a violent one, within three years, compared with 50% for the overall prison population.” Notably, this statistic is less relevant to jurors deciding between the death penalty and life without parole, where either way the defendant will never be released and future dangerousness should be assessed in the context of a prison environment. Evidence that a defendant is a psychopath and has difficulty resisting his antisocial impulses may help the jury understand that he is ill and his crime should be evaluated differently; however, it may also convince the jury that he is still dangerous. Presented with evidence that a defendant is a psychopath, a juror could reasonably conclude that the evidence calls for a more severe penalty, rather than a lesser one.

Evidence of some neurological conditions, like psychopathy, may weaken one justification for sentencing – culpability – only to bolster another – future dangerousness – but there need not always be a tradeoff. In fact, the existence of data on abnormal brain functioning does not imply the associated behaviors are permanent. For example, consider Graham v. Florida and Sullivan v. Florida, in which the U.S. Supreme Court will determine whether juveniles can be sentenced to life imprisonment without parole for committing non-homicide offenses. In addition to petitioners’ claims that juveniles have diminished culpability, they argue that research on brain development indicates juveniles, as a class, are more likely than adults to change their bad behavior. As medical science develops better treatments for conditions that often underlie criminal behavior, such as addiction or mental illness, a greater number of defendants may seek to use evidence of their neurological condition to argue they can be rehabilitated.

Although scientists disagree about what behavioral conclusions can be drawn from brain imaging tests, neuroimaging evidence will continue to be admitted in cases like Brian Dugan’s, because the bar for admissibility is lower for mitigating evidence during the penalty phase of capital cases. Additionally, neuroscience research is being used to challenge the constitutionality of existing sentencing practices, demonstrated most recently by Graham and Sullivan. As a result, neuroscience evidence may have a greater impact on how a defendant is sentenced than on whether a defendant is convicted. How large that impact will be remains to be seen, but as Justice O’Connor said to the Stanford Lawyer magazine, “If we can better evaluate what the problem is and what the chances are of controlling the defendant’s behavior in the future, we’re going to be better off.”

– Kelly Lowenberg