Imagination and False Memories

Abstract

Since the publication of the famous Roediger and McDermott article (1995), researchers have focused on the Deese-Roediger-McDermott (DRM) paradigm and studied many factors involved in memory illusions. The aim of the current study has been to investigate the effect of imaginal encoding on memory confusion in the DRM paradigm. Overall, results indicated that imaginal encoding improves true recall and recognition and significantly reduces false recall, whereas no reduction effect occurs on false recognition. These results are interesting because they are inconsistent with what we would expect based on the available findings in the literature. Otherwise, the present results call into question the underlying processes involved in the creation of veridical and false memories during recall and recognition in DRM tasks.

Effects of Image and Verbal Generation on False Memory

“…The reduction in false recalls with imagery encoding (.17) resembles previous findings using concrete words; imagery instruction in that kind of imagery encoding results in relatively lower rates of false memory than those reported following the standard presentation of abstract words or mixed words (i.e., in which the imagery value varies among words) in the DRM paradigm (Burns et al., 2007; Pérez-Mata et al., 2002; Robin, 2011). Moreover, the incidence of false recognition performances observed in the image condition (.37) is comparable to the findings of Schacter and his colleagues when using pictures (.35) and to those recently reported by Foley (2012) with specific imagery instructions (.24–.18).…”

“…Considering the amount of research that has been conducted with the DRM paradigm, it is surprising that few studies have explored the influence of imagery encoding on the likelihood of false memories. In fact, imagery encoding effects on false memory are virtually absent in the literature about the DRM (see Gallo, 2010), while the number of studies that focus on this cognitive process in the DRM procedure is relatively low (Foley, 2012; Foley, Hughes, Librot, & Paysnick, 2009; Foley, Wozniak, & Gillum, 2006; Gunter, Bodner, & Azad, 2007; Marmurek & Hamilton, 2000; Newstead & Newstead, 1998; Robin, 2011).…”

“…In a previous study (Robin, 2011), we examined the effects of imagery on both recall and recognition tasks, given that works in this area have generally focused on one kind of retrieval test . In the image condition, participants were explicitly told to imagine each object named and to assess the vividness of their image on a graduated scale.…”

“…However, there are now more arguments available in favor of a beneficial effect of the imagery reducing illusions in the DRM paradigm (Burns et al., 2007; Foley, 2012; Foley et al. 2009, 2006; Gunter et al., 2007; Robin, 2011). Consistent with these findings, one could predict that self-generated imagery should reduce the creation of false memories.…”

Reducing False Recognition in the Deese-Roediger/McDermott Paradigm: Related Lures Reveal How Distinctive Encoding Improves Encoding and Monitoring Processes

“…With the goal of improving overall memory accuracy, researchers have identified several methods for reducing the DRM illusion, including study list repetitions ( Benjamin, 2001 ), warnings ( Gallo et al, 2001 ; McCabe and Smith, 2002 ), and requiring participants to specify the source of their retrievals at test ( Multhaup and Conner, 2002 ). Relevant to the present study, study tasks that encourage distinctive processing have been very fruitful, including perceptual manipulations, such as presenting study list words in unique fonts ( Arndt and Reder, 2003 ) or paired with pictures ( Israel and Schacter, 1997 ; Schacter et al, 1999 ; but see Smith and Hunt, 2020 ), and distinctive encoding tasks, such as mental images ( Foley et al, 2006 ; Gunter et al, 2007 ; Robin, 2010 ; Oliver et al, 2016 ; Bodner et al, 2017 ), pleasantness ratings ( Gunter et al, 2007 ; Huff and Bodner, 2013 ), and generation from anagram cues ( McCabe and Smith, 2006 ; Huff et al, in press ). Anagram generation, explored in our study, often yields an increase in correct recognition and a decrease in false recognition relative to a non-distinctive control task, a pattern termed a mirror effect ( Glanzer and Adams, 1990 ; see Huff et al, 2015b for a review).…”