The nature and development of nonverbal implicit memory.

The study investigated the course of developmental changes in performance on nonverbal implicit and explicit memory tests and examined the degree to which implicit memory performance is dependent upon the storage of specific perceptual information. Four-, 5-, and 10-year-old children were required to name fragmented pictures of common objects or to name and answer general knowledge questions about complete versions of the same pictures. After a 48-h retention interval, all subjects were presented with a fragmented picture identification task containing pictures identical to those present during encoding (old), pictures which were from the same basic category as the study items but which varied in their perceptual similarity to those items (same), and novel pictures which were visually and semantically unrelated to the study items (new). The amount of visual information needed to name each item (picture identification threshold) was recorded. Following identification, subjects were asked whether or not they had been shown the picture previously. All age groups showed significant priming such that the picture identification threshold for the old items was lower than that of the new pictures. A smaller but significant priming effect was obtained for the same-name items. This effect was maximized when the same-name items were perceptually similar to the study items. The magnitude of these priming effects did not vary as a function of age, but greater priming was found for those children who identified picture fragments during the study phase. In contrast, the sensitivity of recognition memory performance increased from 4 to 10 years of age. These results suggest that the processes that subserve pictorial repetition priming and recognition memory develop at different rates and that such priming is dependent upon access to specific perceptual representations of studied objects.