The levels of processing effect is a robust phenomenon whereby 'deep' processing of studied material (e.g. making a decision about the meaning of a word) results in better performance on subsequent tests of recall, relative to 'shallow' processing (e.g. counting the number of letters in a word).

My question is this: has any work to date compared deep and/or shallow processing to mere exposure (i.e. simply viewing a to-be-remembered word but not making any decisions about its orthographic/semantic content)?