Levels of Processing Model

The levels of processing model, created by Fergus I. M. Craik and Robert S. Lockhart in 1972, describes memory recall as a function of how deeply a stimulus is mentally processed, with deeper processing producing a more elaborate and durable memory than shallow processing does. Shallow processing, such as attending only to how a word is spelled or sounds, produces a fragile memory trace prone to rapid decay, while deep, semantic processing, in which a person encodes a word's meaning, produces a more lasting trace; the model distinguishes structural or visual processing, phonemic processing, and semantic processing as three such levels. This account stands against the earlier multi-store Atkinson-Shiffrin memory model, which treated memory strength as continuously improved by rehearsal; Craik and Lockhart argued that maintenance rehearsal, which merely repeats a prior analysis, does not itself strengthen long-term memory. In a 1975 study, Craik and Tulving gave participants a list of sixty words, each paired with one of three kinds of question, about the word's visual appearance, its sound, or its meaning in a sentence context, and found that words processed for meaning were remembered markedly better than words processed only for how they looked or sounded.