Researcher makes two face breakthroughs

Tuesday, February 3, 2009

Michael Tarr, professor of cognitive and linguistic sciences, had a vision ­- or two. Conducting two research projects simultaneously regarding visual perception, Tarr and his fellow researchers concluded not only that visual training can lead to a reduction of racial bias, but also that it is possible to distinguish men from women using only color patterns shown on a person’s skin.

In the first project, Tarr, co-director of the Center for Vision Research, and Sophie Lebrecht GS used research conducted at the University of Victoria to analyze the “unconscious biases” that Caucasians have toward other races and performed visual training exercises on subjects to help them identify members of the opposite race, according to the study released in the journal PLoS One on Jan. 21.

Jim Tanaka, professor of psychology at the University of Victoria, and McGill University graduate student Lara Pierce conducted the experiments on 20 Caucasian volunteers.

“Our emphasis is the idea of face recognition as a form of expertise,” Tanaka said.

The experiment used the Affective Lexical Priming Score, a test designed to measure unconscious social biases, Pierce wrote in an e-mail to The Herald. The test subjects were presented with a series of African-American and Caucasian faces, which were followed by a string of letters from the alphabet that formed either non-words or words with a positive or negative connotation.

“If racial bias is present, Caucasian participants are faster at responding to a negative word when it is preceded by an African American face,” Pierce wrote.

The participants then underwent training during which they learned to identify and name the different African-American faces, Tanaka said. The team concluded that after such training, participants were less likely to identify African-American faces with negative words.

“I think it’s a very optimistic result,” Tanaka said. “Hopefully, it’s a way for us to break racial stereotypes.”

Lebrecht and Tarr said this training might be used to decrease racial profiling by police and immigration officers in the future.

“By learning how to tell (races) apart, they are less likely to stereotype them,” Tarr said.

Lebrecht has already begun using the knowledge gained from her and Tarr’s research in a “real world” study. As part of a collaborative project led by Leslie Roos ’09, Lebrecht visits Amos House, a soup kitchen and shelter in Providence, to improve the staff’s “social interactions,” Roos said.

Amos House is a very “multicultural environment,” Lebrecht said, adding that she and Roos “are going monitor how the training affects interactions of the staff.”

In addition to his work on racial biases, Tarr also collaborated on a research project with Adrian Nestor GS that examined the differing skin tones of men and women.

Tarr and Nestor analyzed over 200 images of Caucasian male and female faces to determine that, contrary to popular belief, Tarr said, men have more reddish skin while women have more greenish skin.

The researchers’ findings, which were published in the journal “Psychological Science,” could help determine why women wear make-up, he said.

“Women may be putting on red make-up to highlight green areas,” Tarr said.

The information could also be used in advertising and facial recognition technology, according to a Brown press release.

Leave a Reply

Your email address will not be published. Required fields are marked *