Abstract
As adults, we know that objects seen from a single viewpoint are most likely to have unseen surfaces that are occluded by the visible portions of that object. Previously, we found that infants 6 months and older achieved 3D perceptual completion of objects seen from limited vantage points in 2D displays. We hypothesized that infants with more sophisticated object manipulation skills would be more sensitive to the complete 3D form of objects, relative to infants with less experience handling objects.
Infants were habituated to a computer-generated 3D wedge that rotated back and forth through 15°, allowing only two faces of the object to be seen. Two displays were then shown in alternation, both rotating through 360°, one depicting a complete, solid wedge, and the other an incomplete, hollow wedge composed only of the two sides seen during habituation. A novelty preference for the hollow wedge test display was our measure of 3D object completion.
In further testing on the same day, infants were offered twelve graspable toys. Half of the toys had surfaces with congruent patterns on both sides, and half had different patterns on each side, to motivate infants to turn the objects over. We predicted that infants with more sophisticated exploration skills would examine the incongruent toys more thoroughly.
Infants younger than 6 months displayed more drops of the objects, p[[lt]].05, initiated contact with fewer objects, p[[lt]].01, and took longer to expose the other side of the objects in all postures, p[[lt]].05, compared to infants older than 6 months. Critically, only infants who were better at object manipulation were found to display a strong preference for the novel, incomplete 3D object following habituation, regardless of age. These results imply that emergence of sensitivity to the 3D form of a limited-view object may reside within action development.