« "What is the fairest way for Pa. to deal with juvenile lifers petitioning for resentencing?" | Main | California Supreme Court unanimously applies Graham to lengthy term-of-years sentence »

August 16, 2012

New (curious? useful? suspect?) study on hypotheticals sentencing results and neurobiological evidence

This new New York Times article, headlined "Brain Evidence Sways Sentencing in Study of Judges," provides a lengthy discussion of this newly published research in Science exploring the (hypothetical) sentencing impact of brain science evidence when presented to judges.  Here are excerpts from the Times article, with one particular aspect of the study highlighted for dicussion:

Judges who learned that a convicted assailant was genetically predisposed to violence imposed lighter sentences in a hypothetical case than they otherwise would have, researchers reported on Thursday, in the most rigorous study to date of how behavioral biology can sway judicial decisions.

The findings, published in the journal Science, are likely to accelerate the use of brain science in legal proceedings, experts said, and to intensify a long-running debate about its relevance. Courts have increasingly admitted such evidence — brain scans, mostly, as well as genetic analyses — though many experts say the science is still too primitive to inform legal decisions.  Defense lawyers now commonly introduce brain scans of convicted clients as mitigating evidence in appeals of death sentences, experts said.

Previous studies of how such evidence affects legal decisions are scarce and their results mixed....  The new experiment focused on sentencing by judges, not jury verdicts.  It found that neurobiological evidence reduced judges’ sentences by an average of about 7 percent for a fictional defendant convicted of battery and identified as a psychopath.

“What’s path-breaking about this paper is that it both isolates, as well as one can, the effects of biological testimony on outcomes and it also does this within a sample of the real-world decision makers, the judges themselves,” said Owen D. Jones, a professor of law at Vanderbilt University and director of the MacArthur Foundation Research Network on Law and Neuroscience.  “This moves our understanding forward considerably.”

Dr. Jones and other experts cautioned that the effect might not apply broadly to other kinds of criminal defendants.  It may also play out differently in jury trials.  But they said that the findings were convincing and plausible.

In the study, three researchers at the University of Utah tracked down 181 state judges from 19 states who agreed to read a fictional case file and assign a sentence to an offender, “Jonathan Donahue,” convicted of beating a restaurant manager senseless with the butt of a gun.  All of the judges learned in their files that Mr. Donahue had been identified as a psychopath based on a standard interview — that is, he had a history of aggressive acts without showing empathy.

The case files distributed to the judges were identical, except that half included testimony from a scientist described as “a neurobiologist and renowned expert on the causes of psychopathy,” who said that the defendant had inherited a gene linked to violent, aggressive behavior.  This testimony described how the gene variant altered the development of brain areas that generate and manage emotion....

The judges who read this testimony gave Mr. Donahue sentences that ranged from one to 41 years in prison, a number that varied with state guidelines.  But the average was 13 years — a full year less than the average sentence issued by the judges who had not seen the testimony about genetics and the brain.

In interviews about their decisions, the judges said that a crime of aggravated battery like this one normally carried a sentence of nine years, on average, and 15 years if the defendant was identified as a psychopath, the researchers found.  “But then those who read about the biological mechanism subtracted a year, as if to say, ‘This guy is really dangerous and scary, and we should treat him as such, but the biological evidence suggests that we can’t hold him as responsible for the behavior,’ ” said James Tabery, an assistant professor of philosophy at Utah. He wrote the study with Lisa G. Aspinwall, a psychologist, and Teneille R. Brown, an associate professor in the university’s school of law....

This mixed result — added punishment for the defendant’s being identified as a psychopath, tempered by empathy for his having a possible genetic predisposition — provides a good illustration of what legal researchers call the double-edged sword of biobehavioral evidence.  On one hand, a biological predisposition suggests that a person is likely to be dangerous in the future and should get a longer sentence; on the other, it implies a lower threshold of responsibility. The evidence could cut either way, depending on the judge....

The Supreme Court has been fairly inclusive in what it considers relevant to sentencing, said Dr. Brown, of the Utah law school, so the likelihood is that courts will see much more neurobiological evidence in the future, not less. And the interpretation of that evidence is, like much else, subject to cultural trends as well as to the law. “Our study found that this evidence went in favor of the defense,” Dr. Brown said, “but you can imagine the exact same evidence used for different purposes, shifting back to the prosecution.”

The researchers involved in this study (as well as the reports now about it) are obviously focused on the impact that case-file testimony about genetics and the brain had on the average sentencing outcome.  But I am much more intrigued, and much more troubled, by some other findings in this study based on what is highlighted above.  Specifically, I find most telling and most notable that the proposed sentences for this hypothetical case of aggravated battery ranged from 1 to 41 years and that simply adding the (unscientific?) label "psychopath" apparently increased sentences over 60%. 

Most importantly, I think it is critical to consider this study and all its findings with a large grain of salt.  According to this discussion of the research in Science, the studied "sentencings" were not only of a hypothetical denfendant on a paper record, but the entire study was "web-based."  Though we may be able to learn something about judges' sentencing instincts from these kinds of faux, web-based inquiries, the absence of a real defendant and real lawyers making real arguments in an effort to impact a real sentence makes me cautious about drawing many real-world conclusions from this study.

August 16, 2012 at 06:45 PM | Permalink


TrackBack URL for this entry:

Listed below are links to weblogs that reference New (curious? useful? suspect?) study on hypotheticals sentencing results and neurobiological evidence:


"I think it is critical to consider this study and all its findings with a large grain of salt."

Bingo. In addition, it's unclear, to say the least, why a psychopath should be put away for less time than a person better able to control his behavior. Punishment is important, but so is the safety of potential future victims.

Posted by: Bill Otis | Aug 17, 2012 8:54:30 AM

Thanks for posting this. As one of the authors of the study, I can tell you we certainly recognized the many reasons results like this cannot be directly extrapolated for legal use. But sometimes you can't let the perfect (live, simulated trial with actors) be the enemy of the good (cheaper online study) It was just a first step to try to do better than read the tea-leaves of what judges were thinking, based on reported appellate cases. In order to isolate the impact of biological evidence, we had to remove lots of other real world variables a judge would hear (evidence of the defendant's upbringing, environment, etc). And obviously when this evidence was introduced at trial (we said it all had been) there would have been arguments by both sides. Even more, the presence of the firearm, whether the parole board ultimately has the final say, whether there is a statutory min/max, these all would significantly impact the sentences between states. We did not have enough judges in any one state to make within state analyses, but that of course would be the ideal. Even so, the variance among recommended sentences was striking, as well as the fact that so few judges knew what the average sentence for aggravated assault/battery even was. But what we were primarily interested in was whether biological evidence would change the way judges reasoned through sentencing. And it did. Quite a bit. Whether it *should* influence sentencing is a totally different question, and not one we addressed.

Posted by: Teneille | Aug 17, 2012 12:39:17 PM

Quite to the contrary, Doug. This study deserves very serious consideration. Interestingly, the sample case "aggregated" different aspects of the problem in two different ways, one with biological evidence and one without. What would the result have been if the sample case disaggregated all aspects of the problem, presenting biological evidence only as to risk. I expect the result would have been quite different. The current approach to sentencing greatly oversimplifies the problem.

Posted by: Tom McGee | Aug 17, 2012 2:53:53 PM

ARe going to add this finding to the list of mitigating factors that are really aggravating factors making the defendant super-dangerous? A genetic test for antisocial personality disorder, a valid disorder, that resulted in the abortion of all such babies, both male and female would send our economy soaring, and would end most crime.

Posted by: Supremacy Claus | Aug 18, 2012 9:46:29 AM

Published in the journal Science, are likely to go faster the use of brain science in officially permitted measures,

Posted by: hr dissertation examples | Oct 17, 2012 9:31:26 AM

Post a comment

In the body of your email, please indicate if you are a professor, student, prosecutor, defense attorney, etc. so I can gain a sense of who is reading my blog. Thank you, DAB