Psychopathy Research Ripe for Discovery With High-Throughput Computing

Modern day portrayals characterize psychopaths as one in a million – a bad apple in the bunch of humanity, but incident rates prove otherwise. Psychopathy affects nearly 20% of adult prison inmates in the US. For society at large, the costs of this single issue are between $200 and $400 billion dollars annually.

“When you consider the total societal costs of all of the crimes, police work, trials, and the additional expenses of incarceration, it’s a serious problem,” says Mike Koenigs, assistant professor of psychiatry at the University of Wisconsin-Madison, US. “But this is not necessarily an untreatable or unsolvable problem. One issue is that we haven't yet dedicated adequate resources to understanding this disorder at a more basic level.”

Koenigs’ lab is working to identify and characterize the brain circuits underlying emotion and social behavior. “We're trying to understand how the brain characteristics relate to specific personality characteristics,” Koenigs says. “When we identify an area of abnormal activity, we’re asking which facet of psychopathy it’s related to. For example, is it related to the fact that the individual is impulsive, or that they're remorseless, or both?”

Using a mobile MRI machine, the researchers are able to work with prison populations throughout the state of Wisconsin to create digital models of the psychopathic brain. “We could, for example, show an image or video of someone in physical pain, and we could see that the non-psychopath’s brain is sensitive to that and the psychopath’s brain is not,” says Koenigs.

The inmates are also asked to lie at rest for five minutes. Even when you’re at rest, your brain doesn't go off – there are still spontaneous, low frequency fluctuations and activity. “The areas of the brain that are most often working together day-to-day also tend to work together in the period of unconstrained rest. We can use these correlations as a very basic measure of how functionally interconnected two areas of the brain are.”

Koenigs’ team is able to work in prison facilities for limited periods of time. They may have only a single month where they are able to work six or seven days a week, scanning six to eight subjects a day. In a regular hospital facility, they would normally schedule scans every other day to keep up with data analysis.

“We may have a batch of data where we’ve scanned 275 inmates or more. Using Open Science Grid, we’re able to send all 275 sets at the same time and complete analysis in an average of two days.” says Koenigs. “Without access to these resources, we would be looking at years to analyze this amount of data.”

Working with the Center for High Throughput Computing, the researchers are now able to use a Dropbox-type model to streamline repeated data processing requests. Koenigs’ team has logged nearly 200,000 CPU hours since beginning the project in 2011.

In the case of one psychopath – already serving two life sentences and facing possible execution, trial lawyers successfully admitted neuroimaging data (similar to Koenigs’) into evidence. They claimed the defendant should not be put to death based on a condition he had at birth. The case marked the beginning of a new era in the courtroom: one in which neuroscience and neuroimaging could be used in informing sentencing decisions.

However, judges generally block admission of the data, and for good reason. “There are still plenty of unknowns,” Koenigs says. “In particular, one thing we don't have a good handle on is how these brain abnormalities relate to an individual. Most of the research we do is at a group level, looking for general trends and characteristics across a group.”

Koenigs receives a lot of questions about culpability. Law schools and departments of corrections and justice are all asking – out of both academic and practical interest. Koenigs routinely speaks at other institutions, answering inquiries and sharing insights into his research.  “As our understanding of the neurobiology of psychopathy continues to improve, and our brain imaging measures become more sensitive, the culpability question will become ever more relevant.” -- by Amber Harmon, © i SGTW