Reconnaissance des visages d’un autre groupe ethnique : éclairage d’une approche développementale

Résumé En Fr

We are reviewing the recent findings on the Other Race Effect (ORE), which is the difficulty to recognize faces from other race compared to own race faces. Several hypotheses were proposed: negative attitude (Brigham and Malpass, 1985), attentional mechanisms (Chance and Goldstein, 1981), and the contact hypothesis (Brigham et al., 1982) which postulates that the ORE is due to a lack of experience with other race's faces. The contact hypothesis being, of course, modulated by attention and attitude. Valentine (1991 ; Valentine and Endo, 1992) proposes a norm-based coding model in which faces are encoded as vectors according to their deviation from a prototypical average. The resulting dimensions will differ according to the input received with certain salient, individuating dimensions carrying more “weight” than others. Predominant exposure to faces of a specific species, gender, or race early in life will cause the dimensions of one's prototype to become “tuned” towards such faces. The evidence presented support Valentine's model but we are showing that the quality of the contact is crucial in the experience. We are arguing that a developmental approach is necessary to better understand the ORE. Infants are able to process and recognize faces from an early age and several recent studies have found that the ORE can be observed from 9–10 months of age. Experimental results indicate that 6- to 9-months of age represents an important time of transition in the face processing system. If a certain type of face (other races) is not experienced prior to this period, then we appear to lose our ability to discriminate between individual faces within those groups.

Dans cet article, nous nous attachons à explorer les déterminants de l’other race effect (ORE). Valentine (1991) a proposé un modèle de codage dans lequel chaque visage est encodé en fonction de sa déviation par rapport à une norme qui varie selon l’environnement des individus. Différents travaux chez les adultes montrent qu’en général les résultats soutiennent les prédictions issues du modèle. Nous nous intéressons également au développement de l’ORE au cours de l’enfance. Les travaux de Pascalis et de Schonen (1994) et de Pascalis et al. (1995, 1998) montrent que les nourrissons, de la naissance à l’âge de trois mois, sont capables de reconnaître des visages familiers. Par ailleurs, on note que des difficultés à reconnaître des visages peu fréquents dans l’environnement existent dès la première année.