Nonrecollective memory: The effects of context shifts and study tasks
Brooks, John Oliver, III
Watkins, Michael J.
Doctor of Philosophy
An experience can influence performance on subsequent tasks whether they require conscious recollection (e.g., the judgment of whether something has been previously encountered) or not (e.g., completing word puzzles). What kind of study activities influence performance on nonrecollective memory tests? A proposed resolution to this issue is the task-demand principle, which states that performance on a task is determined by the degree to which the demands of the task match the demands of the original experience. According to the principle, tasks can be categorized along a continuum ranging from data-driven tasks, which require thought about the physical aspects of an item, through conceptually-driven tasks, which require thought about the meaning of an item. Although findings with several types of test have been cited in support of the task-demand principle, the present focus is on two tests that have figured prominently: Perceptual identification, a data-driven task that involves rapid identification of visually degraded words, and word stem completion, a largely data-driven task that involves completing word stems (e.g., WIN for WINDOW) with the first word that comes to mind. The experiments investigated two effects germane to the task-demand principle: (a) the effect of altering, between study and test, the context in which an item is presented and (b) the effect of conceptually-driven study tasks. Contrary to the task-demand principle, context effects were obtained with perceptual identification and word stem completion after subjects engaged in conceptually-driven tasks: Performance was better when the study context was preserved for both perceptual identification and stem completion. Moreover, such context effects for perceptual identification were sensitive to the difficulty of a conceptually-driven task. Finally, perceptual identification performance benefited from a conceptually-driven study task even in the absence of any context manipulations while remaining virtually unaffected by a concurrent data-driven manipulation of typography. These findings limit the generality of the task-demand principle as an account of nonrecollective memory.