The depth with which incoming information is analysed and encoded, ranging from superficial processing of sensory features (1) to semantic and conceptual processing, with deeper levels of processing leading to longer-lasting memories. According to this interpretation, incoming verbal information may be analysed in terms of its sensory and surface features but may also be processed at progressively deeper levels in terms of its phonemic, semantic, and conceptual properties, the early sensory analyses being relatively automatic and effortless and the later deeper analyses requiring attention and effort. The concept was introduced in 1972 by the Canadian-based Scottish psychologist Fergus I. M. Craik (born 1935) and the Canadian-based Australian psychologist Robert S. Lockhart (born 1939). In 1975 Craik and the Estonian-born Canadian psychologist Endel Tulving (born 1927) published a classic experiment in which people first answered yes-no questions about a list of words and then tried to recall as many of them as possible. Some of the words were accompanied by questions asked about their visual appearance (Is the word in lower case? or Is the word in upper case?), others their sound (Does the word rhyme with dog? and so on), and still others their meaning (Is it the name of an animal? and so on). Recall was very poor for words that had been processed according to visual appearance, slightly better for words processed according to sound, and best by far for words processed according to meaning. This approach is sometimes seen as an alternative to the conventional model of three separate memory stores, namely sensory memory, short-term memory, and long-term memory. See also attenuation theory, bottleneck theory, depth of processing, domains of processing theory, elaboration, flanker compatibility effect, proof-reader's illusion (1, 2). LOP abbrev.