An Eye-tracking Research On The Development Of Other Race Effect During Face Processing

Processing faces are important for our adaptation in human society because perceive others’thoughts and emotions through their faces are necessary for social communication and cooperation. Indeed, the item we look at most in everyday life is probably human face, especially faces of our own race. Exploration into this issue provides a better understanding of the role of experience in the formation of visual processing expertise and the origin and establishment of racial prejudice and stereotypes. Thus, this topic has attracted researchers on psychology and neuroscience.In this field, there is an effect confirmed by many research studies:Other Race Effect (ORE), that is, individuals generally recognize own-race faces more accurately and faster than other-race faces. It has been observed among infants, children and adults. Its underlying neural mechanisms was investigated by Event-related-potential and functional MRI methodologies. The existing behavioral and neural imaging studies have provided a great deal of insights into ORE, yet researchers still do not agree with each other on the reasons why ORE exist. Some researchers supposed that people’s experience with own-race faces were much more than that with other-race faces, thus they get expertise on recognize own-race faces while that do not exist for other-race faces. However, other researchers found evidences against this "experience-expertise" hypothesis. They argue that social-categorization model is enough to account ORE. People focus more on group feature information when viewing out-group or other-race faces while focus more on individual feature information when viewing in-group or own-race faces.This debate may be settled by exploration into the visual strategies used by observers for encoding and recognizing own-race and other-race faces. Eye-tracking is one of the ideal methodologies for such purpose because it allows for recording the observer’s fixation on various parts of the face in real time with relatively high temporal and spatial resolution. There have been several recent studies on ORE that have used the eye-tracking methodology. However, there exists a marked difference between the findings. Caldera, Kelly and their colleagues found that although Caucasian and Chinese adults use differently scanning strategies (the eye centric strategy by Caucasians and the nose centric strategy by Chinese), their scanning strategies for both own-and other race faces were the same. However, Kang Lee and his colleagues found that the scanning strategies used for own-and other race faces were different. It should be noted, participants used in the research studies of Caldera, Kelly and their colleagues, the Chinese participants are all studying abroad, they may have enough visual experience on other race faces to affect their scanning strategies before they attend the experiments.In order to clarify this issue and explore into the development of Chinese’s face processing strategies for own and other race faces, we did a face-recognition research on native Chinese children aged from4-7years and native Chinese adults aged from19-25years. They had no or very little direct contact with other-race individuals. We first showed the participants a set of Chinese and Western Caucasian female faces fixed randomly, instructing them to remember these "acquaintance" of experimenter. Then these photos will be mixed randomly with "no-acquaintance" photos of the same amount in the following testing period. All photos were shown to the participants one by one. The participants should judge whether the person in the photo was an acquaintance of experimenter.A Tobii1750eye tracker recorded the participants’fixations on the faces during face encoding (learn and review) and recognizing periods. Then the participants’ fixation data were analyzed by Excel VBA program, Spss, and Matlab2009software.We hypothesized that Chinese participant would be more inclined to scan the nose region of Chinese faces than that of Caucasian faces, and more inclined to scan the eye region of Caucasian faces than that of Chinese faces, because scanning on the nose of own-race faces is the most effective strategy to get the individual feature information, while scanning on the eyes is a better way to get the group feature information, according to the social-categorization model and the existed research results.(Hypothesis1). We also hypothesized that Chinese participants’holistic processing strategies would develop with age, visual scanning becoming more and more nose-centric (Hypothesis2).The experiment results confirmed our Hypotheses to some degree. No matter it’s during face encoding period or recognizing period, Chinese participants’fixation duration on the eyes of Caucasian face was significantly longer than that of Chinese face. Their saccades between two eyes or between eyes and nose of Caucasian faces were significantly more than that of Chinese faces. However, the effect of age group was only significant when participants recognized Chinese faces:younger children’s fixation duration on eyes was significantly longer than that on nose; elder children’s fixation duration on eyes was not significantly different with that on nose; adults’ fixation duration on eyes was significantly longer than that on nose.The experiment results seems to indicate that "social-categorization" model is valid in explaining the difference of participants’visual processing strategies for different race faces, while "experience-expertise" hypothesis is more effective in predicting the development of participants’visual processing strategy when recognizing own-race faces instead of that during encoding faces.