|
Article Excerpt
When requested to state our current address, we typically do not state a former address. Accurate responding to such a request rests on our ability to select from competing related memories the correct memory for recall. As the number of related memories in storage rises, the temporary forgetting of competing memories can facilitate the selection of the correct memory. The retrieval-induced forgetting literature (e.g. Anderson, Bjork, & Bjork, 1994; MacLeod & Macrae, 2001) and the directed-forgetting literature (e.g. Bjork, 1989; MacLeod, 1998) both present explanations for the causes of temporary forgetting.
Both retrieval-induced forgetting and directed forgetting explain temporary forgetting in terms of suppression or preventing the retrieval of memories into consciousness awareness (Anderson et al., 1994; Bjork, 1989; MacLeod & Macrae, 2001; Saunders & MacLeod, 2002). Retrieval-induced forgetting produces inhibition of memories as a consequence of the repeated remembering of other related memories. In this instance, inhibition is advantageous as it reduces retrieval competition on the recall of other more goal-relevant memories. Moreover, the inhibition produced with retrieval-induced forgetting is not permanent and its influence dissipates over time (MacLeod & Macrae, 2001).
Directed forgetting also produces temporary forgetting. The paradigm commonly involves two variations: the item method and the list method. In the item method, each item is individually followed by either an instruction to remember or to forget, with the remember and the forget instruction items presented in a random order. The poorer recall of forget instruction items is thought to reflect the poorer processing of those items relative to the more extensive processing of the remember instruction items (Basden, Basden, & Gargano, 1993). As the mental mechanisms of retrieval inhibition and suppression are not associated with the item method, the list method is the major focus of the present paper.
The directed forgetting produced with the list method is thought to reflect the suppression or the preventing of retrieval of information at which an instruction to forget is directed (Bjork, 1989; Geiselman, Bjork, & Fishman, 1983). The suppressed or inhibited retrieval produced with the list method is advantageous to overall memory functioning because it reduces interference from memories at which the instruction to forget is directed on the recall of other memories. The list method of directed forgetting commonly involves presenting participants with two lists of items to study. Following the presentation of List 1, participants are instructed to either remember or to forget that list and are then presented with List 2 with instructions to remember this list. Two directed-forgetting effects are produced with the list method. Recall of List 1 items is greater for RR participants (those instructed to remember both lists) than for FR participants (those instructed to forget List 1 and to remember List 2), while FR participants recall more List 2 items than do RR participants.
Inhibition impairs access to List 1 items by blocking the retrieval routes to those items in episodic memory and is responsible for both directed-forgetting effects (Geiselman et al., 1983). Compared with RR participants who experience no inhibition impairing their recall of List 1 items, FR participants recall fewer List 1 items as a result of inhibition impairing access to those items. However, impaired access to List 1 items does reduce the proactive interference from those items. Reduced proactive interference reduces retrieval competition and enhances the recall of List 2 items for FR participants. RR participants suffer from retrieval competition from List 1 items and report fewer List 2 items.
Recently, Anderson and Green (2001) have employed a think/no-think task to study the effects of suppression on recall. After training participants to recall the second word of a pair when prompted with the first word, participants were repeatedly presented the first word in a pair and instructed to not think about the second word. Recall of the second word in a pair decreased as the number of no-think trials increased. Anderson and Green interpreted their results to mean that deliberately keeping material out of consciousness increased the later suppression of that material.
The inhibition produced with the list method is also temporary and its effects are released with the presentation of retrieval cues. For example, following a presentation of items from the inhibited list on recognition tests, no differences in the recognition of List 1 or List 2 items were observed between RR and FR participants (Basden et al., 1993; Geiselman et al., 1983). Retrieval inhibition has also been released following a second study opportunity (Geiselman & Bagheri, 1985). In the Geiselman and Bagheri experiment, recall of former to-be-forgotten items was higher following a second study opportunity than the recall of former to-be-remembered items. However, the Geiselman and Bagheri conclusion is suspect owing to their use of the item method of instruction thought to lead to inferior encoding of to-be-forgotten information, rather than the inhibition of that information (Basden et al., 1993). Nonetheless, the absence of directed forgetting reported in these studies suggests that to-be-forgotten items remain in memory at full strength but that their retrieval strength is impaired.
Further evidence that inhibition produced with the list method is temporary is the reinstatement of proactive interference from List 1 items in recall for List 2 items. Bjork and Bjork (1996) reported a reinstatement of proactive interference in the recall of List 2 items following participants completing a recognition test containing List 1 items as foils. Specifically, FR participants failed to show enhanced recall for List 2 items following an interpolated recognition test containing List 1 items as foils, but they did show enhanced recall of List 2 items when the items on the interpolated recognition test were not former List 1 items.
The presentation of experimenter-provided retrieval cues at or before a critical test is common to all the directed-forgetting research listed above. Relatively little directed-forgetting research has examined the impact on retrieval following exposure to participant-generated retrieval cues. One method for studying the impact of participant-generated retrieval cues is to test participants repeatedly and to minimize the influence of experimenter-provided retrieval cues. The hypermnesia literature (for reviews see Erdelyi, 1996; Payne, 1987) reports improvement in recall across repeated recall tests owing only to the experience of taking those tests (Roediger & Payne, 1982).
Ballard (1913) was first to study hypermnesia by presenting participants with a set of to-be-remembered information and then, after various retention intervals, administering several recall tests. His results showed that recall improved across the repeated tests. Payne, Hembrooke, and Anastasi's (1993) retrieval dynamics account provides a theoretical explanation for increased recall across tests. They believe that retrieving an item marks the location of that item in memory. Marking an item's location increases that item's accessibility for future tests. In subsequent tests, participants can quickly locate previously marked items, leaving additional sampling time (Marsh, Landau, & Hicks, 1996) to locate previously non-retrieved items. The combination of marking and additional sampling time to search for non-retrieved items leads to increases in recall across tests.
Following the lead of Bjork and Bjork (1996), Experiment 1 examined the pattern of recall across tests for List 2 items. Requiring the recall of only List 2 items allows for an examination of how retrieval inhibition of List 1 items impacts the recall of List 2 items. FR participants who inhibit access to List 1 items do not suffer from retrieval competition from those items while searching memory for List 2 items. In contrast, RR participants maintain access to items from both lists and suffer retrieval competition from List 1 items in their recall of List 2 items. Retrieval competition from List 1 items detracts from the determination of list status and facilitates the intrusion of List 1 items into the recall of List 2 items.
The source-monitoring framework of Johnson and colleagues offers a model...
|
