Mood and the DRM paradigm: An investigation of the effects of valence and arousal on false memory

Abstract

Recent studies regarding the effect of mood on the DRM (Deese-Roediger-McDermott) illusion have not been able to clearly establish yet whether valence or arousal is most critical in determining susceptibility to false memories, nor what the underlying processes are. In three experiments, both the valence and the level of arousal of participants' mood were manipulated. Six conditions were used: positive mood with high/low arousal, negative mood with high/low arousal, neutral mood, and a control condition. Memory was tested by means of immediate and delayed recognition and immediate free recall. The mood induction procedure was effective. For recognition memory, there was an effect of arousal on the endorsement of critical lures. Low-arousal moods elicited more false recognition than high-arousal moods, regardless of valence. Based on signal detection analyses, the effect was attributed to more liberal response criteria with low arousal, in combination with a tendency towards improved item-specific memory with high arousal.

An exploratory high-density EEG investigation of the misinformation effect: Attentional and recollective differences between true and false perceptual memories

“…The less positive P3b response in misinformation based endorsements relative to true memory endorsements are in line with theoretical proposals that false memory reports are associated with a weaker context-item memory match (Molinaro & Carreiras, 2010) as well as more liberal responding (Luna & Migueles, 2008;Van Damme, 2013) and less stringent monitoring (Loftus et al, 1989;Parker et al, 2008). These findings indicate that the recognition response associated with misinformation endorsements are not accompanied by significant levels of discrepancy detection or strong context-item match responses.…”

“…The likelihood of stronger context-item matches in true memory responses and the fact that false event memory rates are associated with more liberal responding (Luna & Migueles, 2008;Van Damme, 2013) and less stringent monitoring (Loftus, Donders, Hoffman, & Schooler, 1989;Parker, Garry, Engle, Harper, & Clifasefi, 2008) suggests that misinformation based false memories will be associated with reduced P3b levels relative to true episodic memories. However, it is also possible that some level of discrepancy detection will be present even for endorsed misinformation memories elevating the P3b voltage for these responses.…”

Arousal—But Not Valence—Reduces False Memories at Retrieval

“…Other studies using the same paradigm have found that arousal (i.e., level of activation) induced before encoding, and not valence per se, affects the likelihood to incur memory errors. Nonetheless, evidence is mixed: in some cases, high arousal has been found to elicit higher false recall and false recognition than low arousal [ 6 ], but also the opposite has been found, with high arousal leading to fewer memory errors than low arousal [ 7 ]. Importantly, arousal exerts its influence on memory distortions not only at encoding but also when induced after learning: indeed, it has been found to reduce acceptance of misleading information in a misinformation paradigm [ 8 ].…”

Post-learning arousal enhances veridical memory and reduces false memory in the Deese-Roediger-McDermott paradigm

“…Most studies to date have manipulated emotion or arousal before or during learning (e.g., Buchanan & Lovallo, 2001;Corson & Verrier, 2007;Kensinger & Corkin, 2004;Mather et al, 2006;Storbeck & Clore, 2005;Van Damme, 2013). In such studies, the effects of the manipulation may be on any aspect of the memory process, including attention, encoding, motivation, rehearsal, and consolidation; the effect on a specific aspect of the memory process cannot be isolated.…”

“…Indeed, this is one reason that the animal literature has often employed exogenous administration of arousal-associated hormones (see McGaugh, 2000) and some human studies employ non-emotive, sympathetic arousalinducing manipulations such as muscle tension (Nielson & Jensen, 1994;Nielson, Wulff, & Arentsen, 2014;Nielson et al, 1996) and adrenergic-inhibiting agents such as beta-blockers to study these mechanisms (Cahill, Prins, Weber, & McGaugh, 1994;Nielson & Jensen, 1994). Furthermore, studies that have attempted to distinguish arousal and valence effects in human memory modulation studies using emotive stimuli have implicated arousal as the dimension of primary relevance (Anderson et al, 2006;Corson & Verrier, 2007;McBride & Cappeliez, 2004;Nielson & Powless, 2007;Schwartz, 1975;Van Damme, 2013). Lastly, the arousal stimulus used in the present study has been used in past studies where arousal, as measured by a change in heart rate and/or skin conductance, was specifically responsible for the effects on memory (e.g., Nielson & Meltzer, 2009;Nielson et al, 2005).…”