Abstract
The capacity to use configural information, along with local featural cues, is theorized to underlie adults' expertise in processing faces in an upright orientation (Diamond, Carey & Woods, 1980). That is, immature face processing is restricted to local featural information (i.e., eye shape and color), while sophisticated face processing also utilizes configural cues (i.e., distance between the eyes). Mondloch et al. (2004) presented a discrimination task to eight year-olds and adults in which faces, both upright and inverted, differed in either the spacing among local features or the appearance of the local features themselves. They found that the children were as sensitive as adults to featural differences among faces, but their ability to detect spacing differences was not as strong as adults'.
The aim of the current study was to determine whether adding rotational motion to the stimuli during encoding would facilitate configural processing in children. Because the use of motion cues is thought to be fundamental to early three-dimensional representation (Kellman, 1984), we predicted that motion would benefit children in the Mondloch et al. (2004) task, specifically in detecting spacing differences. The encoding condition varied among participants: while some saw still views of the faces during the learning phase, others saw laterally rotating faces. In both age groups, a greater inversion effect was obtained for pairs that differed spatially than those that differed featurally, suggesting that children at least as young as eight years old are able to use configural processing for faces. Contrary to predictions, while motion was beneficial to adults' performance, it was detrimental to children's performance. We are currently investigating whether these results reflect a general inability of children to extract facial form from motion efficiently, or if aspects of the specific task constrained their performance.
Support from NICHD HD38315-R01 to P.G.