ABSTRACT
Underlying Craik and Lockhart's original levels-of-processing paper was an approach to perception that was perhaps represented most clearly by the work of Anne Treisman. Bransford, Franks, Morris, and Stein are clearly sympathetic to some of the later developments in levels of processing and also provide evidence for the importance of elaboration, particularly for learning the relations between material. However, they point out very clearly that quantity of elaboration is less crucial than the uniqueness of the specified relationship. Their concept of transfer-appropriate processing has some similarities with Tulving's encoding specificity principle in emphasizing the importance of matching the conditions at retrieval to those operating during initial learning. Bransford et al'. s paper manages to combine a series of ingenious and cogent experiments on current theoretical issues with a broad approach to human learning and memory that seems to offer some promise of developing into a usable theory of human learning having applications outside the narrow confines of the laboratory.
