Abstract
As adults we know that a 3D object seen from one viewpoint is likely to have surfaces on its far side occluded by the visible portions of that object. Do infants, likewise, represent objects seen from a limited perspective as coherent volumes, or instead as consisting solely of the surfaces that are visible from the current viewpoint? We investigated age differences in 3D object completion by presenting infants with a computer-generated 3D wedge that rotated back and forth through 15 deg, providing opportunity to see only two faces of the object. Two displays were then shown in alternation, both rotating through 360 deg. One display depicted a complete, solid wedge, and the other depicted an incomplete, hollow wedge composed of only two sides. We reasoned that consistently longer looking at one test display would reflect a novelty preference.
We tested 20 infants in two age groups (M age = 114.6 days and 234.6 days). There were no reliable preferences for either test display by younger infants. Older infants looked longer at the incomplete object in the first test trial pair, p < .05; there were no reliable preferences in the other two trial pairs. Older infants, therefore, may represent objects as complete in 3D space despite having only a limited perspective. One possible developmental mechanism is rooted in manual exploration of objects afforded by upright sitting—which arises between four and six months.