Attention and implicit memory tests: The effects of varying attentional load on conceptual priming

Abstract

The role of attention during encoding is important to many current accounts of the implicit/explicit memory distinction. Some accounts suggest that implicit memory tests reflect automatic (non-attentiondemanding) encoding processes, whereas other accounts (such as the transfer-appropriate-processing view) suggest that performance on conceptual implicit tests requires attention during encoding. The present study manipulates attention at encoding over several levels (by varying short-term memory load) and examines the effects on the category-exemplar production task (a conceptual implicit memory test) and its explicit counterpart, category-cued recall. Dividing attention decreased performance on both tests, but in different ways. Mild divisions of attention reduced recall but not conceptual priming. Strong divisions of attention reduced performance on both tests and, in addition, eliminated conceptual priming entirely. These findings resolve apparently conflicting results in the literature and help to clarify the relationship between attention and performance on implicit memory tests.

Divided attention, aging, and priming in exemplar generation and category verification

“…By this logic, dividing attention should limit semantic processing during encoding and should have deleterious effects on conceptual priming. Indeed, that is what we have found for exemplar generation, and what has been found by previous investigators ofthis task who have varied level of encoding or who have used sufficiently demanding distractor activities during encoding (e.g., Gabrieli et aI., 1995;Hamann, 1990;Maki & Knopman, 1996;Monti et aI., 1996;Mulligan, 1997;Mulligan & Hartman, 1996). By contrast, in the present experiments, neither reduced level ofprocessing nor division ofattention has produced reliable deficits in priming in category verification.…”

“…Here too, the results have been inconsistent. Some studies report that dividing attention reduces priming in exemplar generation (Gabrieli et aI., 1995;Gabrieli et aI., 1999;Mulligan, 1997;Mulligan & Hartman, 1996), whereas others do not (Isingrini, Vazou, & Leroy, 1995;Schmitter-Edgecombe, 1996). Although Koriat and Feuerstein (\ 976) found a null effect of dividing attention on priming in word association, another task that involves access to semantic information, Mulligan (1998) reported that digit monitoring reduced priming in this task.…”

“…Several studies bear out this conjecture. Hard, but not easy, divided attention tasks reduce priming in exemplar generation (Mulligan, 1997), word-stem completion (Wolters & Prinsen, 1997), and perceptual identification (Mulligan & Hornstein, 2000). This strongly suggests that conclusions about the role of attention in any given indirect measure of memory can be drawn only after a number of manipulations of attention have been tried.…”

“…Some studies have found that dividing attention at study does not reduce the amount ofrepetition priming observed in tasks thought to depend primarily on perceptual processing, such as perceptual identifica-tion, lexical decision, word-fragment completion, wordstem completion, or picture-fragment completion (Kellogg, Newcombe, Kammer, & Schmitt, 1996;Mulligan, 1997;Mulligan & Hartman, 1996;Parkin et aI., 1990;Schmitter-Edgecombe, 1996). Other studies have found reductions in repetition priming on such tasks following study in divided attention conditions (Hawley & Johnston, 1991;Weldon & JacksonBarrett, 1993).…”

Evidence for an implicit influence of memory on future thinking

“…For instance, participants are more likely to correctly answer general knowledge questions if they were exposed to the answer in an unrelated learning phase (e.g., Blaxton, 1989;Challis & Sidhu, 1993;Hamilton & Rajaram, 2001). Similarly, participants are more likely to incorporate a critical exemplar (typically of low frequency; e.g., cheetah) during a category instance generation task (e.g., list the first eight animals that come to mind; see Geraci & Rajaram, 2004;McDermott & Roediger, 1996;Mulligan, 1997;Rappold & Hashtroudi, 1991;Srinivas & Roediger, 1990) if they have been previously exposed to that item.…”