Mood and False Memories for Events CitaTon: Mirandola C, ±o²alini E (2016) Arousal— But Not Valence—Reduces False Memories at Retrieval. PLoS ONE 11(3): e0148716. doi:10.1371/ journal.pone.0148716 Editor: Alexandra Kavushansky, ±echnion— Israel InsTtute of ±echnology, ISRAEL Received: October 21, 2015 Accepted: January 3, 2016 Published: March 3, 2016 Copyright: © 2016 Mirandola, ±o²alini. ±his is an open access arTcle distributed under the terms of the CreaTve Commons A³ribuTon License , which permits unrestricted use, distribuTon, and reproducTon in any medium, provided the original author and source are credited. Data Availability Statement: Data are freely available through Figshare.com ( h³ps://dx.doi.org/10. 6084/m9.´gshare.2067951.v1 ). Arousal—But Not Valence—Reduces False Memories at Retrieval Abstract Mood a²ects both memory accuracy and memory distorTons. However, some aspects of this relaTon are sTll poorly understood: (1) whether valence and arousal equally a²ect false memory producTon, and (2) whether retrieval-related processes ma³er; the extant literature typically shows that mood inµuences memory performance when it is induced before encoding, leaving unsolved whether mood induced before retrieval also impacts memory. We examined how negaTve, posiTve, and neutral mood induced before retrieval a²ected inferenTal false memories and related subjecTve memory experiences. A recogniTon-memory paradigm for photographs depicTng script-like events was employed. Results showed that individuals in both negaTve and posiTve moods–similar in arousal levels–correctly recognized more target events and endorsed fewer false memories (and these errors were linked to remember responses less frequently), compared to individuals in neutral mood. ±his suggests that arousal (but not valence) predicted memory performance; furthermore, we found that arousal raTngs provided by parTcipants were more adequate predictors of memory performance than their actual belonging to either posiTve, negaTve or neutral mood groups. ±hese ´ndings suggest that arousal has a primary role in a²ecTng memory, and that mood exerts its power on true and false memory even when induced at retrieval. IntroducTon Memory illusions have been in the spotlight of cognitive and neurocognitive scientists for decades. The discovery that memory does not lead to static representations of events is now a milestone concept; memory scientists have begun to investigate in detail the numerous variables that influence the occurrence of distortions, including the emotional state or mood of the rememberer. The present study is focused on the effect of transient mood (negative, positive, and neutral), induced immediately before retrieval, on the creation of memory distortions for everyday scripted events. When individuals are exposed to mood-inducing emotional stimuli, the tendency to incur memory errors — or false memories — may depend on the valence of their affective state. Some studies have shown that negative mood reduces and positive mood enhances memory errors [ 1 – 3 ], and that this is especially true when mood is induced before learning. In these studies, the Deese-Roediger-McDermott paradigm (DRM; [ 4 ]) was employed. The DRM paradigm consists of the presentation of a series of wordlists including semantically related words (e.g., nurse , sick , medicine ). Within each list, words also converge in meaning toward a non- presented lure (e.g., doctor ); the acceptance of this word at recognition represents a false memory. According to the affect-as-information hypothesis, negative mood affects encoding as it favors item-specific processing, which limits the incorporation of non-presented lures; on the other hand, positive mood promotes relational processing, which facilitates the OPENACCESS
