The effect of warnings on false memories in young and older adults

Abstract

In the present experiments , we examined adult age differences in the ability to suppress false memories, using the Deese-Roediger-McDermott (DRM) paradigm (Deese, 1959;Roediger & McDermott, 1995). Participants studied lists of words (e.g., bed, rest, awake, etc.), each related to a nonpresented critical lure word (e.g., sleep). Typically, recognition tests reveal false alarms to critical lures at rates comparable to those for hits for studied words. In two experiments, separate groups of young and older adults were unwarned about the false memory effect, warned before studying the lists, or warned after study and before test. Lists were presented at either a slow rate (4 sec/word) or a faster rate (2 sec/word). Young adults were better able to discriminate between studied words and critical lures when warned about the DRM effect either before study or after study but before retrieval, and their performance improved with a slower presentation rate. Older adults were able to discriminate between studied words and critical lures when given warnings before study, but not when given warnings after study but before retrieval. Performance on a working memory capacity measure predicted false recognition following study and retrieval warnings. The results suggest that effective use of warnings to reduce false memories is contingent on the quality and type of encoded information, as well as on whether that information is accessed at retrieval. Furthermore, discriminating between similar sources of activation is dependent on working memory capacity, which declines with advancing age.

Can corrective feedback improve recognition memory?

“…Although substantial false alarm rates generally persist following warnings, they are usually significantly lower than those of unwarned controls (e.g., Gallo, Roberts, & Seamon, 1997;McDermott & Roediger, 1998). Although some evidence indicates that the warning must be delivered before encoding of the study lists for a reduction in false alarms to occur (e.g., Anastasi, Rhodes, & Burns, 2000;Neuschatz, Payne, Lampinen, & Toglia, 2001), other evidence suggests that a warning delivered between study and test can have a beneficial effect as well (e.g., Lane, Roussel, Starns, Villa, & Alonzo, 2008;McCabe & Smith, 2002;Starns et al, 2007). If, as the latter finding suggests, participants benefit from insight into the test design without having such knowledge at study, the use of this information, and the consequent decrease in critical false alarms, must be achievable via processes operating at test.…”

Digitally manipulating memory: Effects of doctored videos and imagination in distorting beliefs and memories

“…Chambers and Zaragoza (2001), however, found that warnings helped subjects regardless of when they were delivered. In studies of the DRM effect, pre-encoding warnings robustly reduced the rate of false memories, whereas the efficacy of postencoding warnings depended on other variables (e.g., Gallo, Roediger, & McDermott, 2001;McCabe & Smith, 2002;Watson, McDermott, & Balota, 2004). Two studies of greatest relevance to the present research have also found mixed results: Landau and von Glahn (2004) found that subjects who received warnings about imagination exhibited a smaller inflation effect than those who received no warning; whereas a study from the communications literature (Kelly & Nace, 1994) revealed that warnings about the capabilities of digital editing software failed to reduce subjects' belief in news articles and photos from disreputable sources.…”

The Adaptive and Strategic Use of Memory By Older Adults: Evaluative Processing and Value-Directed Remembering

“…Tun et al (1998) observed that age-related diVerences decreased when all participants were encouraged, through task demands, to rely on a gist representation of the study list. McCabe and Smith (2002) have also shown that older adults can reduce false alarms to the critical lure if warned prior to the encoding session about the nature of the task and materials. This suggests that although older adults may typically rely on gist-based representations, under certain conditions they are able to access and use more specific information.…”