UW News

September 13, 2018

UW psychology professor honored for founding research on implicit bias

UW News

When Tony Greenwald and his colleagues developed the online Implicit Association Test two decades ago, it enjoyed quick success in the pre-laptop, pre-smartphone, nascent Internet world, with some 45,000 participants in the first month.

The test, which requires classifying words and images rapidly according to their meanings, captures unconscious biases toward — depending on the test version — race, gender, age and dozens of other traits and preferences. Since its debut in 1998, the test has been taken online more than 25 million times and has been used in over 2,000 peer-reviewed research articles. Its concepts have been the subject of classroom and workplace debates, policies and programs.

Now its creators are being honored for their contributions to science and society. Greenwald, a University of Washington professor of psychology, along with Mahzarin Banaji of Harvard University and Brian Nosek of the University of Virginia, will receive the Golden Goose Award from the American Association for the Advancement of Science. The award honors federally funded work that, in the words of AAAS, “may have been considered silly, odd or obscure when first conducted but has resulted in significant benefits to society.” The award was established in 2012 to counter criticisms of wasteful government spending, such as the late U.S. Sen. William Proxmire’s Golden Fleece Award.

Tony Greenwald

Tony Greenwald

“Our work has had much more impact than any scientist has reason to expect for their efforts,” Greenwald said. “We have been blessed by the attention we got both inside the profession of psychology and outside.”

For Greenwald, who joined the UW faculty in 1986, the award is a recognition of work that started 30 years ago at the UW, as he and colleagues developed the field of implicit social cognition. Creating the Implicit Association Test, or IAT,  and placing it online gained it instant attention, he said. But as federal agencies shifted their funding emphasis to applied rather than basic research about 10 years later, the test eventually lost its federal sources of research support.

“As it turned out, the judgment that our research lacked practical application was totally wrong,” Greenwald said.

He points to the myriad uses not only of the online tests — now supported by the nonprofit Project Implicit, which the researchers established in 2005 — but also of the very idea of implicit bias and how it makes its way into our everyday decisions and behaviors. Unconscious beliefs, stemming from childhood, can influence how teachers evaluate students, how employers hire job candidates or promote workers, how police officers treat suspects and how judges mete out punishment. In many cases, data demonstrate those influences, as with bias in the courtroom: Bail for Black or Hispanic people is often set higher than for White people for the same offenses.

Greenwald points out that many corporations, schools and other organizations provide presentations, workshops and lessons in recognizing and combating bias. But the science needs more development before these efforts will become effective.

A reliable finding: Most of those who take the test have a moderate-to-strong preference for racially White people over racially Black people, except for African Americans, whose results are distributed widely from White preference to Black preference. Another: People of all ages favor young over old.

Giving in to implicit bias is like “being the carrier of a disease you don’t know you have,” Greenwald said. Learning that you have it is not enough to prevent its spread.

“For some people it’s totally easy to accept that there are things in their head that they don’t want to be there,” he said. “A leader needs to know that implicit bias can afflict an organization’s decision makers, and that there are policies that will avoid those undesired effects. A leader needs to examine the organization, determine where the disparities are occurring, and have the fortitude to implement policies that will fix the problem.”

There are methods of evaluating employees and colleagues that can reduce the influence of unconscious bias: the blind audition in music, for instance, in which a performer plays behind a screen; or with job candidates, pre-determining the criteria used for evaluation.

Greenwald hopes that scientists can soon come together to take a position, not unlike that for climate change, on the importance of implicit bias and how to address its consequences.

“In a way, it’s surprising that we’re talking about the IAT being 20 years old,” Greenwald said. “One expects a method like this to be replaced before long by a superior method. I was among many who have tried to improve on the IAT, but no one has yet succeeded.”

Greenwald, Banaji and Nosek will be honored in a ceremony Sept. 13 at the Library of Congress in Washington, D.C.