Despite that egalitarian attitude, according to new Northwestern University research, subconscious -- or implicit -- bias can emerge subtly but quickly from its hiding places in the psyche and cause even well-meaning whites to look at identical facial expressions of African Americans and European Americans and see greater hostility in the African American faces.
Or take whites' perceptions of racially ambiguous faces that combine both African American and European American features. If the expression on the racially ambiguous face is hostile, European Americans are more likely to identify it as African American.
The unusual research, by Kurt Hugenberg, assistant professor, Miami University, and Galen Bodenhausen, professor of psychology at Northwestern, strongly suggests implicit bias distorts perception of facial cues so important to effective communication and perpetuates stereotypes. A self-fulfilling prophecy may be among the most troubling consequences.
"If stereotypes color something as basic as face perception, then the downstream consequences may be considerable," said Bodenhausen. "Perceived hostility will at best promote avoidance -- or worse, may foster reciprocation."
Further research is needed to determine how perceptual bias in the first moments of contact might play themselves out over the course of social interaction, but current results suggest that negative dynamics will follow.
While implicit bias has been the focus of a variety of research, Bodenhausen's work is rare in that it uses changing computer -generated facial expressions to tease out how deeply-rooted prejudice distorts perceptions. Most experiments on face perception use still photographs as stimuli, despite the dynamic nature of facial displays in real-life interactions.
The two companion studies are titled "Facing Prejudice: Implicit Prejudice and the Perception of a Facial Threat" (Psychological Science, 2003) and "Ambiguity in Social Categorization: the Role of Prejudice and Facial Affect in Race Categorization" (Psychological Science, in press).
In the first experiment, when identical facial expressions of African Americans and European Americans changed from hostile to positive expressions, white viewers often perceived the hostility to linger longer in black faces.
The experiment included 24 European Americans who observed four movies with computer-based faces of both whites and African Americans that morphed from unambiguous hostility to unambiguous happiness. The target's expression was ambiguous, somewhere between hostile and friendly, for a substantial period of time in each movie; and black and white faces were matched precisely for both facial structure and expression. Participants, who watched the movies on computers, were instructed to press the space bar when the target face no longer expressed its initial emotion.
They were then presented with a "feeling thermometer" about the five different social groups, including Caucasians and African Americans. The thermometer indicated how warmly or coldly they felt about each group on a one –to-100 score, with higher responses indicating warmth.
Finally, participants took an implicit association test, a simple word categorization test designed to measure implicit racial bias. Although explicit attitudes toward blacks did not relate to face perception, individuals who tested as high in implicit bias, compared to those who measured low, saw hostility as lingering longer and appearing more quickly in the faces of African Americans.
Twenty-four European Americans also participated in the experiment on ambiguity in social categorization that is in press. Fifteen computer-generated faces were morphed to contain racially ambiguous facial structures, skin tone and hairstyles. Each face was further manipulated to create two expressions, one clearly happy and the other clearly hostile. This time the task was to identify the race of the racially ambiguous faces. A feeling thermometer and the word categorization test also were used in these experiments to measure prejudice.
"When faces were seen to display relatively hostile expressions, individuals high in implicit prejudice tended to categorize them as African American," said Bodenhausen.
The relationship between prejudice and categorization was not evident for happy faces. Thus, the results strongly suggest that in cases of uncertain categorization, the stereotypical biases of individuals relatively high in implicit prejudice take hold and drive categorization.
"The susceptibility of biracial individuals to societal prejudice depends on whether or not they are categorized as members of a culturally stigmatized group," said Bodenhausen.
"Of course, angry displays are likely to result in problematic social interactions in general, but biracial individuals may be saddled with an additional burden when dormant negative racial stereotypes are activated."
Both experiments show how stereotypes can color our early and most immediate perceptions of other people, he said. "We are constructing the meaning of other people's behavior on the basis of our attitudes and assigning stereotypical characteristics readily."
Journal
Psychological Science