It is often assumed that processes such as visual perception work in the same ways in all people, but research now suggests that how we see things may be influenced by our expectations and opinions.

Yair Pinto, a cognitive neuroscientist at the University of Amsterdam, and his colleagues asked 45 white Dutch people to perform a binocular rivalry task — a standard tool in visual perception studies. The researchers presented low-contrast images of white, Moroccan and black faces to one eye and high-contrast changing patterns to the other.

At first, the study participants were aware of seeing only the patterns. But when the contrast of the patterns was reduced and that of the faces was increased, the patterns became invisible and the faces broke through into the participants' awareness.

The researchers asked the participants to indicate when they became aware of seeing the faces by pressing a button on a computer keyboard. It took the participants an average of one-hundredth of a second longer to become aware of the Moroccan and black faces than the white ones.

The team also measured participants’ racial biases using the implicit association test, in which participants pair concepts such as 'black' and 'white' with qualities such as 'good' and 'bad'. The participants who exhibited greatest implicit bias in the association test took longest to become aware of the black and Moroccan faces.

“We expected the Moroccan and black faces to break through more quickly so we were very surprised to find the opposite,” says Pinto.

Pinto and his colleagues presented their results this week at the annual meeting of the Association for the Scientific Study of Consciousness in Brighton, UK.

Subjective experience

Earlier work1 with binocular rivalry tasks had shown that photographs of faces with emotional expressions enter conscious awareness more quickly than those with neutral expressions. Pinto and his colleagues found, however, that the emotional content of the faces had no effect on their participants’ reaction times.

The results suggest that high-level mental processes such as racial stereotyping can exert a direct ‘top-down’ effect on lower-level processes such as visual perception.

Axel Cleeremans, a cognitive neuroscientist at the Free University of Brussels, says that the results could be unrelated to racial bias. “The simpler explanation is that it is difficult to generalize what we know about faces of our own ethnic groups to other groups. This is known as the ‘other race’ effect and has nothing to do with racial prejudice.”

To test this, Pinto wants to repeat the experiments with black and Moroccan participants. “We know that Moroccans and black people in Holland associate more with their own group, and also that they have negative stereotypes of their own group,” he says. “If performance on the task was due to familiarity with your own race, the effect would be reversed, but if it was driven by negative stereotypes it would be the same as with the white people.”

He would also like to probe the effects of likes and dislikes using pictures of animals instead of faces. “Most people won’t readily admit their racial prejudices, but they are comfortable saying they don’t like insects or rats,” says Pinto.