Face recognition and perception by young children

Résumé du document

This study tested four and five-year-olds for their ability to use differences in the spacing among features when asked to pairing faces. The task was presented as a drag and drop features game which provided double reinforcement to children.

Cognitive variables were taken in consideration during the test's elaboration and spatial selective attention was measured with a neuropsychological test from the NEPSY. Because stimuli were either modified on the mouth or on the eyes according to the three levels of difficulty, the measure of perspective's variables on children's performances was possible.

A lot of children were excluded from the final sample because their performances were influenced by cognitive variables. Other children performed above chance level but were poorer than the adults. They showed the same pattern of performances for the three levels of difficulty than adults for modifications on the mouth but not on the eyes. Those results showed the necessity to take cognitive variables into consideration when interpreting young children performances. They brought us to the conclusion that both qualitative and quantitative changes are responsible for the gap between children and adults in configural processing skills.

Extraits

[...] Pascalis & A. Slater (Eds.), The development of face processing in infancy and early childhood (pp.79-97). New York: Nova. Pascalis, O., de Haan, M., Nelson, C.A., & de Schonen, S. (1998). Long-term recognition memory for faces assessed by visual paired comparison in and month-old infant. Journal of Experimental Psychology: Learning, Memory, and Cognition 249-260. Pascalis, O. Demont, E., de Haan, M., & Campbell, R. (2001). Recognition of faces of different species: a developmental study between 5 and 8 years of age. [...]

[...] This last assumption is particularly true when young children are tested. However, analysis of perseverations is only possible when a sufficient number of trials are administered, which is a limitation of some previous studies on 4 year-olds. In Mondloch et al. (2008) for example, about half of the children failed the task and authors concluded that they are not sensitive to configural processing. It is possible that performances of those children were affected by cognitive variables. However, it is more difficult to measure perseverations when only 4 or 6 trials are administered. [...]

[...] It was equipped with an EMR pen, which was presented as a “magic to children. Participants were tested individually and were seated 30 cm away from the monitor. The task was presented as a drag and drop features task: two pictures were displayed on the left side of the screen (original picture and one of its altered versions) and the same pictures were displayed on the right side of the screen. The task consisted of pairing same pictures by dragging and dropping each of the right- sided pictures next to the corresponding left- sided picture (see Figure 2). [...]

[...] (1991). Newborn's preferential tracking of face-like stimuli and its subsequent decline. Cognition 1-19. Jones, L.B., Rothbart, M.K., & Posner, M.I. (2003). Development of executive attention In preschool children. Developmental Science, 498-504. Korkman, M., Kirk, U., & Kemp, S. (2003). NEPSY, bilan neuropsychologique de l'enfant. Les Editions du Centre de Psychologie Appliquée Levy, F. (1980). The development of sustained attention (vigilance) in children: some Normative data. [...]

[...] We modified faces of the original set by altering the distance between facial features. This alteration modified the second-order relationships in the face, specifically the vertical metric distances in this study, which underline configural faces' processing more than horizontal alterations (Goffaux & Rossion, 2007). Six new versions were created from each of the original faces by moving eyes down 8.88 mm mm or 3.18 mm (respectively 0.92 and 0.31 from the testing distance of 30 cm) or by moving mouth up 5.3 mm mm or 2 mm (respectively and 0.21 from the testing distance of 30 cm). [...]