Academic journal article Memory & Cognition

Automatic Retrieval in Directed Forgetting

Academic journal article Memory & Cognition

Automatic Retrieval in Directed Forgetting

Article excerpt

In two experiments, we investigated the role of automatic retrieval in directed forgetting, using a stem completion test. In each experiment, we compared the performance of an implicit group, a process dissociation procedure (PDP) group, and a group given a speeded response task. The response times of the speeded response group on the stem completion task replicated earlier data in suggesting that this group adopted a purely automatic retrieval strategy. Experiment 1 revealed a directed-forgetting effect on automatic retrieval with the item method, but Experiment 2 revealed no directed-forgetting effect on automatic retrieval with the list method, consistent with the lack of implicit effects with the list method throughout the literature. Both experiments showed lower automatic estimates for the PDP group than for the implicit and speeded response groups, which did not differ. The data are consistent with either a selective rehearsal or an inhibition account of directed forgetting with the list method. The comparison of methods for assessing automatic retrieval is consistent with earlier evidence suggesting that subjects may adopt a generate/recognize strategy when given direct retrieval instructions.

In the directed-forgetting paradigm (see Golding & MacLeod, 1998, for an extensive review), subjects study a series of stimuli one at a time. With the item method, each stimulus is immediately followed by a cue to either remember or forget that item for a subsequent memory test, whereas with the list method, subjects are given an instruction halfway through the list to forget the preceding items and to remember the next set of items. With both methods, the subjects later complete a memory test under instructions to retrieve all the items, regardless of whether they were cued to be remembered or forgotten.

A key empirical dissociation in the literature is that free recall of remember (R) items is better than that of forget (F) items with both the item and the list methods; however, this directed-forgetting effect is evident on recognition tests only with the item method (Basden, Basden, & Gargano, 1993; E. L. Bjork, R. A. Bjork, & Anderson, 1998; but see Golding, Long, & MacLeod, 1994). This pattern of data has contributed to the general consensus that different processes operate in the two paradigms (see, e.g., Basden & Basden, 1998; E. L. Bjork et al., 1998; MacLeod, 1999). With the item method, differential processing of R and F items is generally acknowledged as the source of performance differences: Subjects withhold processing of a word until they are cued to remember it. Thus, R words receive more extensive processing than do F words at study and are, accordingly, better remembered on both recall and recognition tests. Variables that minimize processing differences tend to reduce the directedforgetting effect (e.g., deep processing, Wetzel, 1975; relations among items, Horton & Petruk, 1980; subjectperformed tasks, Earles & Kersten, 2002).

By contrast, the common interpretation of findings from the list method is that the F cue results in retrieval inhibition of the F items (Basden & Basden, 1998; E. L. Bjork et al., 1998; but see MacLeod, Dodd, Sheard, Wilson, & Bibi, 2003). Retrieval inhibition can be caused by subjects' establishing a set or context for the items from the first half of the list and then changing to a new context as items from the second half are processed (Basden & Basden, 1998; Basden, Basden, & Morales, 2003; Sahakyan & Delaney, 2003).' This retrieval inhibition builds as subsequent R items are learned (E. L. Bjork et al., 1998; Conway, Harries, Noyes, Racsma'ny, & Prankish, 2000). The lack of directed forgetting on a recognition test with the list method indicates that processing of F items on the test releases retrieval inhibition (E. L. Bjork et al., 1998; R. A. Bjork, 1989).

According to some authors, conscious retrieval of the stored representation (Basden & Basden, 1998) or conscious contact with the stored representation at retrieval (E. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed

Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.