Toronto police officers respond to a call about a black man with a history of mental illness threatening a woman with a hammer in the hallway of their apartment building. The man approaches them with the hammer raised. One officer shoots him. The man, Andrew Loku, is killed.
Montreal police officers respond to a call about a black man with a history of mental illness destroying his apartment. The man approaches them with a screwdriver in each hand. Officers fire plastic bullets, tase him, then shoot him with their pistols. The man, Pierre Coriolan, is killed.
In both cases, enraged protesters accuse police of using unnecessary force because these men were black. If that's true, were the officers explicitly prejudiced against the men they shot? Or did they have a less obvious implicit bias against them – an unconscious attitude based on stereotypes?
New research suggests the way our brains make associations between black people and the physical threat we think they pose is the greatest predictor of police using lethal force against a black person. These biases are held not just by the officers in question, but by the wider communities in which black people are killed by police.
This correlation is reported by a team of researchers led by Eric Hehman, an assistant professor of psychology at Ryerson University, in the journal Social Psychological and Personality Science. Dr. Hehman's study adds to a growing body of research on implicit bias and how it can influence how police interact with black people.
For their study, Dr. Hehman's team looked at the results of 2,156,053 U.S. residents who completed Harvard University's famous Implicit Association Test, an online tool that measures the strength of the associations one makes between white people, black people and good and bad traits. They geolocated the results and analyzed them alongside data on people killed by police in the U.S. during a nine-month period in 2015.
They found that in places where implicit bias against black people and an association between black people and weapons were stronger, there was a disproportionate use of lethal force by police against black residents. Canadian data on fatal police shootings of black people was not available to include in the study, but Dr. Hehman said the principles they were researching could extend to Canada, too.
"We're measuring the lady down the street who lives on the corner, the person who's selling you some oranges. Just regular, average community members," Dr. Hehman said. "But we're still predicting these extremely potent and important consequences that are by police."
It may be even more difficult to defeat the implicit biases police officers hold because of the nature of their work. In training simulations where individuals must decide whether or not to shoot armed or unarmed individuals, police who deal with non-white individuals in routinely dangerous situations – such as those on a drug force or SWAT team – have been found to be more likely than beat cops or civilians to shoot unarmed black men.
"In a moment where they're under extreme stress and duress, they're not really able to think consciously about what they're saying, what they're doing and so on. They're going to revert back to their instincts," says Nicholas Rule, a Canada Research Chair in social perception and cognition.
In June, Dr. Rule, who teaches psychology at the University of Toronto, testified at the coroner's inquest into the death of Andrew Loku. He shared results of one study he did, in which participants consistently guessed that black men, just based on photos of their faces, were larger and stronger than white men of similar build. With those misperceptions came the assumption that more force would be needed to subdue them compared with white men.
In the verdict following the Loku inquest, the jury made several recommendations, one of which Dr. Rule had pushed for: to require all new officers and those requalifying to take the Implicit Association Test – the same one that was used in Dr. Hehman's research. The jury also suggested officers receive implicit-bias and anti-blackness training.
But there's little evidence to support implicit-bias training across various sectors. Several analyses found that after 24 hours, the bias-reducing effects of the training had vaporized, usually as a result of the individual returning to their regular life and exposure to the very stereotypes they were trying to stamp out.
Based on decades of research, many social scientists believe the best treatment for bias is what was first described by American psychologist Gordon Allport in 1954 as the "intergroup contact hypothesis" – a theory that the more contact members of a majority group having with a minority group, the less prejudice they feel towards them. But Dr. Allport emphasized that not just any contact would work: the quality was important and required equal status between all individuals.
For this reason, Emilie Nicolas is skeptical of whether anything can change implicit bias in police because of the immutable power dynamics between officers and the people they serve. Ms. Nicolas is the president of the NGO Québec Inclusif, which has been pressing the Quebec government to launch a commission into systemic racism in the province. She says there is a hierarchy between black people and white people that is naturalized through policing. Even if a beat cop spends all his time in a black neighbourhood and hosts community events, the nature of his interactions with residents isn't the sort of quality contact Dr. Allport's theory requires.
"Community barbecues are based on the assumption that if you don't do them, these people may be impolite or whatever," Ms. Nicolas says. "You don't have these community barbecues in [wealthy white neighbourhoods] so the very fact that they have them speaks of prejudice that exists."