In the popular imagination and in conventional discourse — especially in the context of highly charged news events such as the shooting of Trayvon Martin — prejudice is all about hatred and animosity.

Scientists agree there's little doubt that hate-filled racism is real, but a growing body of social science research suggests that racial disparities and other biased outcomes in the criminal justice system, in medicine and in professional settings can be explained by unconscious attitudes and stereotypes.

Subtle biases are linked to police cadets being more likely to shoot unarmed black men than they are unarmed white men. (Some academics have also linked the research into unconscious bias to the Trayvon Martin case.)

Calvin Lai and Brian Nosek at the University of Virginia recently challenged scientists to come up with ways to ameliorate such biases. The idea, said Harvard University psychologist Mahzarin Banaji, one of the researchers, was to evaluate whether there were rapid-fire ways to disable stereotypes. Groups of scientists "raced" one another to see if their favorite techniques worked. All the scientists focused on reducing unconscious racial bias against blacks.

"Within five minutes, you have to do something to somebody's mind so that at the end of those five minutes you will now show a lower association of black with bad. And so this was run really like a competition to see which ones of them might work to reduce race bias and which ones don't," Banaji said.

The results were as surprising for what they didn't find as for what they did. Teaching people about the injustice of discrimination or asking them to be empathetic toward others was ineffective. What worked, at least temporarily, Banaji said, was providing volunteers with "counterstereotypical" messages.

"People were shown images or words or phrases that in some way bucked the trend of what we end up seeing in our culture," she said. "So if black and bad have been repeatedly associated in our society, then in this intervention, the opposite association was made."

Banaji, who has been a pioneer in studying unconscious biases, said she has taken such results to heart and tried to find ways to expose herself to counterstereotypical messages, as a way to limit her own unconscious biases.

One image in particular, she said, has had an especially powerful effect: "My favorite example is a picture of a woman who is clearly a construction worker wearing a hard hat, but she is breast-feeding her baby at lunchtime, and that image pulls my expectations in so many different directions that it was my feeling that seeing something like that would also allow me in other contexts to perhaps have an open mind about new ideas that might come from people who are not traditionally the ones I hear from."

Copyright NPR 2018.