Death Sentences Are Doled Out Based on Looks

Though there are guidelines for when a convicted criminal merits the death penalty (in states that still have capital punishment), ultimately, the jury makes the decision. A new study finds that the facts of the case are not the sole determinant of whether or not a jury will issue a death sentence—based on the research, certain “untrustworthy” facial features appear to play a significant role in capital-punishment sentencing. 

According to the study, published Dec. 14 in the journal Psychology Science, people associate certain facial features such as down-turned lips and heavy brows with being untrustworthy. It’s one of the earliest forms of stereotype bias humans learn—even babies prefer those without these traits—and scientists have found it affects outcomes such as who we select as leaders, who gets paid more, and criminal-sentencing outcomes.

“There's longstanding knowledge among practicing attorneys that jurors form impressions of defendants, oftentimes based on arbitrary unreliable characteristics,” says Craig Haney, a professor of psychology at the University of California, Santa Cruz. For example, decades of evidence suggest that Black defendants generally and defendants of any skin color accused of killing white females are more likely to be sentenced to death. 

“Researchers have for decades used what's called counter-stereotype interventions to reduce things like racial bias, gender bias, etc.,” says Jon Freeman, an associate professor of psychology at Columbia University, and an author of the new study. “We’ve been wanting to apply those same kinds of principles and take a very different approach to understanding facial stereotype biases as learned and malleable.” In Freeman’s study, he shows for the first time that facial bias can be accounted for with a short training when the death penalty is at stake.

To test this, Freeman conducted a series of four experiments using images of 400 inmates convicted of murder in Florida, all white males, some of whom received a death sentence and some of whom received life in prison. In the first experiment, more than 450 volunteers were shown the images and asked to score each on trustworthiness and attractiveness. Before the exercise, a portion of the participants were put through a short training module designed to break the association between facial features and trustworthiness, in which traditionally “untrustworthy” faces were shown with descriptions of positive behaviors, and vice versa. Across the board, men who were sentenced to death were more likely to be labeled as untrustworthy by participants in the control group, with attractiveness scores closely related as well. In the trained group, however, trustworthiness didn’t predict real-world sentencing outcomes. 

More From TIME

The other three experiments included similar trainings with slightly different tests afterwards, including one where participants were asked to make sentencing recommendations assuming full guilt and another where they were asked to do the same after being given the full details of a case. In each experiment, participants who received the training were less likely to fall into the same associative patterns.

Read More: What My Week of Jury Duty Taught Me About Race

That facial bias can be corrected so easily in the short term is really telling of just how unprepared jurors in the real world are, says Haney, who was not involved with the study. Jury selection is “a fairly crude process,” he says. “We really put jurors in the position of making profoundly important decisions, including the decision between life or death. And it's a role for which they receive no training whatsoever.”

A fundamental philosophical shift happens when capital punishment enters a courtroom, says Haney. Rather than looking at evidence to determine how an event occurred, when a jury is considering the death penalty, their analysis becomes about a person. “At that stage, they've been convicted,” Haney says, “Now, the question is, do they deserve the ultimate penalty or the next worst penalty? And that is very much a decision based on who [a jury] thinks the defendant is.” Any biases that jurors feel are more likely to bubble to the surface when making this more subjective moral evaluation.

Still, training jurors before they sit in on actual cases just isn’t realistic yet, say Freeman and Haney. First, experts need to know more about how these different types of biases interact—trustworthiness, race, gender, and more all tend to be associated with one another in different ways that Freeman hopes to uncover by replicating his study with other populations of inmates. 

Even with all the information in the world, says Haney, it’s unlikely that widespread anti-bias training for juries would ever be supported across the political spectrum. “I can imagine differences of opinion about what the content [of such training] should be,” he says. Second, and perhaps the biggest non-political barrier than anyone attempting to design an anti-bias jury training would encounter, is that short-term trainings like Freeman’s don’t tend to correct biases for much longer than it takes to run an experiment. Trials for capital offenses are often weekslong, and in research settings, lasting changes in implicit bias require repeated, regular interventions. But learning that attitudes towards facial features can be changed at all is “quite striking,” says Freeman. 

“I think the larger point is that there are these biases, and there are things that can be done about them. And this is just one more way in which we don't really prepare jurors for the all important role that they're going to be asked to play,” says Haney.