Running Head: Negative Emotion and Memory SpecificityHow Negative Emotion Enhances the Visual Specificity of a MemoryElizabeth A. Kensinger (1,3), Rachel J. Garoff-Eaton (2,3), and Daniel L. Schacter (2,3)(1) Department of Psychology, Boston College(2) Department of Psychology, Harvard University(3) Athinoula A. Martinos Center for Biomedical ImagingAddress correspondence to:Elizabeth A. Kensinger, Ph.D.McGuinn Hall, Room 510140 Commonwealth Ave.Chestnut Hill, MA 02467Phone: 617-552-1350Fax: 617-552-0523Email: [email protected] Emotion and Memory Specificity 2Some studies have suggested that emotion primarily increases memory for “gist,”and does not enhance memory for detail. There are, however, some instances in whichnegative objects (e.g., snake, grenade) are remembered with more visual detail thanneutral objects (e.g., barometer, blender). In the present fMRI study, we examined theencoding processes that lead a person to remember the exact visual details of negativeand neutral objects, and to remember which of two decisions were made about the objects(a size decision or an animacy decision). The enhancement in memory for a negativeitem’s visual details appeared to result from enhanced visual processing: The rightfusiform gyrus, a region known to be critical for processing exemplar-specific details,showed a greater extent and magnitude of activity during the successful encoding ofnegative objects. Activity in the right amygdala also corresponded with memory forvisual detail, although it did not relate to memory for the task performed with the item.These data provide strong evidence that engagement of some amygdalar regions cancorrespond with enhanced memory for certain types of details, but does not ensuresuccessful encoding of all contextual details.Keywords: amygdala, detail, fMRI , fusiform, gist, memory, neuroimagingNegative Emotion and Memory Specificity 3Individuals often claim to remember emotional experiences with rich detail.Laboratory studies repeatedly have demonstrated that people are more likely to reportvivid recollections of emotional events than of nonemotional ones (e.g., Dewhurst &Parry, 2000; Ochsner, 2000; Kensinger & Corkin, 2003). Despite the subjectivevividness associated with emotional experiences, however, numerous studies haverevealed that emotion does not enhance memory for all types of detail. Rather, emotionoften has focal effects on memory, with only some aspects of an experience more likelyto be remembered because of its emotional content (see Adolphs & Buchanan, 2002;Heuer & Reisberg, 2004 for recent reviews). For example, Adolphs and colleagues haveshown that emotion can enhance memory for the “gist” or general theme of presentedinformation, but can reduce memory for details (Adolphs et al., 2001; Denburg et al.,2003). They also have provided evidence that these mnemonic effects are mediated bythe amygdala: individuals with amygdala damage do not show these gist-enhancing anddetail-impairing effects of emotion on memory (Adolphs et al., 2005).Not all effects of emotion on memory, however, are explained by this gist vs.detail framework. Emotion sometimes can enhance the visual specificity with whichinformation is remembered: Individuals are better at distinguishing “same” from“similar” object exemplars when those exemplars are negative than when they are neutral(Kensinger et al., 2006; Kensinger et al., 2007), suggesting that individuals are morelikely to remember exactly what a negative object looked like when presented at study.The primary goal of the present study was to examine the neural processes thatlead to this enhancement in memory for the visual details of negative objects. A growingnumber of neuroimaging studies have attempted to elucidate the neural processes thatNegative Emotion and Memory Specificity 4contribute to memory specificity for neutral items (for reviews, see Schacter, Dobbins, &Schnyer, 2004; Schacter, Gallo, & Kensinger, 2007), but there is a virtual absence ofsuch research concerning emotionally arousing stimuli. Prior research examiningparticipants’ abilities to remember the visual details of neutral items has revealed acritical role of the right fusiform gyrus. Activity in the right fusiform gyrus correspondswith subsequent memory for the exact visual details of a studied object (e.g., being ableto identify not only that an umbrella had been studied, but also to identify whichparticular umbrella had been studied; Garoff et al., 2005), consistent with previousfindings suggesting that the right fusiform gyrus is important for the processing of theexemplar-specific visual details of an object (Koutstaal et al., 2001; Simons et al., 2003;Marsolek, 1999). Based on this prior literature, we hypothesized that activity in the rightfusiform gyrus would correspond with enhanced visual specificity for negative objects aswell as for neutral ones.The critical question was whether amygdala activity at encoding also wouldcorrespond with memory for visual detail. The amygdala often has been shown tomodulate sensory functioning (LeDoux, 1995; Davis & Whalen, 2001; Dolan &Vuilleumier, 2003). Individuals are more likely to detect emotional than neutral objectsin visual arrays (Ohman et al., 2001) or to identify emotional compared to neutral stimulipresented only briefly (Anderson & Phelps, 2001; Phelps, Ling, Carrasco, 2006), andrecent neuroimaging data have suggested that these effects are due to interactionsbetween the amygdala and fusiform gyrus (e.g., Noesselt et al., 2005; Vuilleumier et al.,2004) and between the amygdala and occipital cortex (e.g., Tabert et al., 2001). Whilethese studies have shown a link between amygdala activity and sensory processing, theyNegative Emotion and Memory Specificity 5have not revealed whether connections between the regions influence the visualspecificity of a memory. In fact, there has been significant debate about whetheramygdala engagement would correspond with enhanced or reduced visual specificity:Although it might seem intuitive that such connections would enhance visual specificity,a couple of studies have shown stronger relations between activity in visual processingregions and the successful encoding or retrieval of neutral as compared to emotionalinformation (Dolcos et al., 2004; Sharot et al., 2004). Although these studies have notexamined the visual specificity of a memory, they raise the intriguing possibility that