The independent news organization of Duke University

Jurors more likely to convict on same level of evidence for more serious crimes, says new Duke study

Duke researchers have found that the type of alleged crime may increase a jury’s belief in guilt in a recent study published in Nature Human Behavior. 

“Human beings really value coherence in their explanations, and we’re likely to ignore or diminish pieces of evidence that disagree with our preferred conclusions," said John Pearson, lead study author and assistant professor of biostatistics and bioinformatics. "There are a number of cases of wrongful convictions in which crucial pieces of exculpatory evidence were ignored because they went against the incriminating narrative that was already in place. It’s important for society that we get this right.”

Pearson said conversations with his colleague from Duke, Pate Skene, associate research professor of neurobiology, inspired the study.  

“He and I would run into each other over lunch and talk about how we could borrow tools from the study of economic decision-making to think about legal decision-making," Pearson said. "Pate always had a hunch that the sorts of emotional processes that govern our reactions to heinous crimes could exert subtle effects on the way we process evidence.” 

To investigate Pate's idea, the researchers designed 33 fictional legal cases about different types of crimes—from  murder to less serious crimes—but with pieces of evidence that could be swapped out. They surveyed hundreds of people, including some law students and trained lawyers, asking them to rate how strong they found each case on a scale of 0 to 100. Each respondent only saw one variant of each case, and Bayesian statistical methods were used to estimate the average effect of each type of evidence on the population at large.

The researchers observed that the severity of the crime introduced bias, making respondents more likely to deem the defendant guilty before even seeing evidence. 

For respondents who were not legally trained, the researchers found that the more severe a crime, the more likely they were to be convinced of the defendant’s guilt, given the same level of evidence.

However, the legally-trained respondents reliably separated their emotional reactions to the case from their assessment of the evidence, Pearson noted. 

Unsurprisingly, researchers also noted the “CSI effect” across the board: people putting too much stock in fingerprints and fiber evidence, treating them as comparable to DNA. 

Researchers are currently debating a couple of explanations for why more severe crimes may sway individuals to deem an individual guilt despite the same level of evidence for a less severe crime. A sociological hypothesis is that individuals have a strong desire to “put things right” when a crime is committed, Pearson said. Alternately, the psychological hypothesis suggests it be that humans' emotional state affects how they process information. When worked up, people often make decisions more quickly and with less information, which can lead them to shakier conclusions, Pearson said. 

Pearson hopes to introduce more quantitative work into the study of psychology behind legal decisions. 

“These sorts of statistical approaches offer a complementary method of studying one of the most complex things we humans do as a species: make collective decisions about culpability and punishment.” 


Share and discuss “Jurors more likely to convict on same level of evidence for more serious crimes, says new Duke study” on social media.