Memory Editing Mechanisms

A Special Issue of Memory

Susan E. Gathercole, Martin A. Conway

• Publisher: Psychology Press
• Year: 2020
• ISBN:
  9781000149883
• Edition: 1
• Topic: Psychology
• Subtopic: History & Theory in Psychology
• Index: Psychology

This special issue of Memory is devoted to an investigation of those mechanisms by which memory is edited for inaccuracies and inconsistencies. In the past 20 years false memories have been investigated from a variety of different angles. Substantial evidence indicates that false memories can be created in a number of different situations including word learning, sentence and story memory, eyewitness memory, memory for faces, and memory for naturalistic scenes. In each of these cases, it has been found that memory is subject to a range of distortions. But there has also been an increasing recognition that this is only half the story. For although memory is subject to distortion, there are also quality control mechanisms that are utilized that allow our memories to be relied on as reasonably accurate under most circumstances. These mechanisms include recollection rejection, distinctiveness, and source memory. The focus of this special issue then, is on the interplay between

Recollection rejection of false narrative statements

C. J. Brainerd, V. F. Reyna, and S. Estrada
Cornell University, Ithaca, NY, USA

Our research was focused on a false-memory editing operation that is posited in fuzzy-trace theory—recollection rejection. The main objectives were (a) to extend model-based measurement of this operation to a narrative task that ought to ensure high levels of recollection rejection and (b) to study five manipulations that ought to influence recollection rejection by affecting the accessibility of verbatim traces of narrative statements: recency of narrative presentation, narrative repetition, type of false-memory item, testing delay, and repeated testing. The results showed that the narrative task did indeed yield high levels of recollection, with an estimated 49% of gist-consistent distractors being rejected in this way on initial memory tests. Consistent with current theoretical conceptions of false-memory editing, the results also showed that recollection rejection increased as a function of manipulations that should enhance the accessibility of verbatim traces of narrative statements, with repeated testing delivering especially large increases in verbatim accessibility.

We report some research that revolved around a false-memory editing operation that is posited in fuzzy-trace theory (FTT)—recollection rejection. The crux of the operation is that participants rely on verbatim traces of actual experience (e.g., “I ate a hamburger and drank a Pepsi at the baseball game”) to perform rejections of false events that preserve the gist of their experience (“ate a hot dog”, “drank a Coke”). Importantly, recollection rejection involves positive, principled suppression, a feature that contrasts with the passive rejections that result when false but gist-consistent events produce such weak memory evidence that they are not deemed to be viable candidates for acceptance. With recollection rejection, the evidence that accumulates from memory is strong, but it is disconfirmatory. According to the current theoretical model of recollection rejection, this process occurs because verbatim traces neutralise the seductively high familiarity of false but gist-consistent events (Brainerd, Reyna, Wright, & Mojardin, 2003). It follows that, other things being equal, recollection rejection is most likely to occur in circumstances in which verbatim traces are readily accessible in memory (e.g., see Gallo, 2004; Lampinen, Odegard, & Neuschatz, 2004). A situation of this sort was explored in our research.

Recollection rejection differs in two key respects from the older and more extensively studied notion of recollection (e.g., Jacoby, 1991). The first difference is concerned with the type of memory response that each supports (see Rotello, 2001). Recollection rejection, as just noted, produces principled suppression of false but gist-consistent events, whereas recollection produces principled acceptance of true events. The other difference is concerned with the types of memory representations that the two operations tap (see Brainerd, Reyna, & Mojardin, 1999). Recollection rejection is assumed to involve retrieval of verbatim traces of the surface form of experience, whereas recollection is traditionally assumed to involve retrieval of semantic memories.

Prior studies of recollection rejection have dealt almost entirely with memory for word lists, usually long lists of unrelated words (for a review, see Brainerd & Reyna, 2005). In contrast, the present research was focused on a narrative paradigm that was developed by Reyna and Kiernan (1994, 1995). The paradigm has the following features. First, participants are exposed to one or more three-sentence narratives that follow the pattern exemplified by the two narratives in Table 1. The first two sentences specify a familiar relation (spatial or magnitude in Table 1) that holds among a trio of everyday objects (between a bird, a cage, and a table, or between coffee, tea, and cocoa in Table 1). These two sentences are called premises. The third sentence in each narrative is a filler target that refers to one of the three objects but not to the relation that is specified in the premises. Second, following such narratives, sentence-recognition tests are administered that consist of three types of probes: (a) the relation-specifying targets (the two premises; TPO1 and TPO 2 in Table 1); (b) three unpresented but gist-consistent distractors (the false-memory probes in Table 1); and (c) four distractors that do not preserve the gist of experience and serve as response-bias measures (the control distractors in Table 1). Concerning the three false-memory probes, note that the first (TPN in Table 1) is a true paraphrase of a premise that is generated by inserting a new word (e.g., “The coffee is hotter than the tea” becomes “The tea is cooler than the coffee”); the second (TIO in Table 1) is a valid inference that uses only old words to connect two objects that were not paired in either premise (e.g., “The coffee is hotter than the cocoa”); and the third (TIN in Table 1) is a valid interference that uses a new word in conjunction with old words to connect two objects that were not paired in either premise (e.g., “The cocoa is cooler than the coffee”). Finally, concerning the response-bias items, note that there is one control probe that matches each of the gist-consistent probes (targets and false-memory probes) in degree of surface overlap with presented sentences, thereby controlling simultaneously for hits or false alarms that are due to nonmemorial factors, such as guessing, as well as for hits or false alarms that are due to the overall familiarity of the words and word orderings that comprise these probes. Thus, FPO probes (false paraphrases of premises that contain only old words) are control items for premises; FPN probes (false paraphrases of premises that contain new words) are control items for TPN distractors; FIO probes (invalid inferences that contain only old words) are control items for TIO distractors; and FIN probes (invalid inferences that contain new words) are control items for TIN distractors.¹

¹ Because it was necessary to administer eight distinct types of recognition probes to unconfound surface familiarity from meaning resemblance, it is necessary to use three-letter acronyms for these probes to avoid numerous repetitions of phrases such as “true premises with original wording”. As these acronyms appear, their meanings can be regenerated merely by remembering that they follow a 2 × 2 × 2 factorial structure: The first letter indicates whether or not a probe is true, in the sense of being consistent with the gist of a narrative (T = true or F = false); the second letter indicates whether a probe mentions two objects that were connected in one of the premise sentences or whether it is an inference that involves two objects that were not connected in the premises (P = premise or I = inference); and the third letter indicates whether the probe contains only old words from the narrative or contains a new word (O =old or N = new).

TABLE 1 Examples of Reyna and Kiernan’...

Table of contents

• Cover
• Copyright Page
• Contents
• Notes to Authors
• Editorial: Memory editing mechanisms
• Phantom ROC: Recollection rejection in a hybrid conjoint recognition signal detection model
• Recollection rejection of false narrative statements
• Variations in processing resources and resistance to false memories in younger and older adults
• Editing (out) generated study words in a recognition exclusion task: Effects of response signal delay and generation procedure
• Two types of recollection-based monitoring in younger and older adults: Recall-to-reject and the distinctiveness heuristic
• Repetition effects in associative false recognition: Theme-based criterion shifts are the exception, not the rule
• Developmental differences in false-event rejection: Effects of memorability-based warnings
• Memory editing: Knowledge, criteria, and alignment