Neural Processes Supporting Young and Older Adults’Emotional MemoriesElizabeth A. Kensinger1,2and Daniel L. Schacter2,3Abstract& Young and older adults are more likely to remember emo-tional information than neutral information. The present func-tional magnetic resonance imaging study examined the neuralprocesses supporting young (ages 18–35) and older (ages62–79) adults’ successful encoding of positive, negative, andneutral objects (e.g., a sundae, a grenade, a canoe). The re-sults revealed general preservation of the emotional memorynetwork across the age groups. Both groups recruited theamygdala and the orbito-frontal cortex during the successfulencoding of positive and negative information. Both ages alsoshowed valence-specific recruitment: right fusiform activitywas greatest during the successful encoding of negative in-formation, whereas left prefrontal and temporal activity wasgreatest during the successful encoding of positive informa-tion. These valence-specific processes are consistent with be-havioral evidence that negative information is processed withperceptual detail, whereas positive information is processedat a conceptual or schematic level. The only age differencesin emotional memory emerged during the successful encod-ing of positive items: Older adults showed more activity inthe medial prefrontal cortex and along the cingulate gyrusthan young adults. Because these regions often are associatedwith self-referential processing, these results suggest that olderadults’ mnemonic boost for positive information may stemfrom an increased tendency to process this information inrelation to themselves. &INTRODUCTIONIndividuals typically are more likely to remember emo-tional information than they are to remember nonemo-tional information (LaBar & Cabeza, 2006; Reisberg &Heuer, 2004; Buchanan & Adolphs, 2002). Although theneural processes corresponding to this emotional mem-ory enhancement have been thoroughly studied inyoung adults, with activation in the amygdala and theorbito-frontal cortex corresponding with later memoryfor emotional material (LaBar & Cabeza, 2006; Hamann,2001), little is known about how the emotional memorynetwork changes with age. The primary goal of the pres-ent study was to examine the neural processes that cor-respond with young and older adults’ successful encodingof emotional information.On the one hand, there is reason to believe that theemotional memory network may be fairly stable withaging. Some behavioral studies have suggested thatolder adults display emotional memory enhancementscomparable to those of young adults (e.g., Denburg,Buchanan, Tranel, & Adolphs, 2003; Kensinger, Brierley,Medford, Growdon, & Corkin, 2002). Moreover, theregions associated with emotional memory in youngadults (e.g., amygdala and orbito-frontal cortex) tendto be relatively preserved with aging, both structurally(Salat et al., 2004; Salat, Kaye, & Janowsky, 2001; Chow& Cummings, 2000; Tisserand, Visser, van Boxtel, &Jolles, 2000) and functionally (Gutchess, Kensinger, &Schacter, 2007; Williams et al., 2006). On the other hand,however, are studies that have suggested that the va-lence of information (whether positive or negative) mayfundamentally influence the way in which older adultsprocess and remember emotional information. A num-ber of studies have revealed a ‘‘positivity shift’’ withaging; whereas young adults are more likely to remem-ber negative information than positive or neutral infor-mation, older adults may be at least as likely (or evenmore likely) to remember positive information com-pared with negative information (Mather & Carstensen,2005).It has been proposed that this ‘‘positivity shift’’ mayoccur because older adults put more emphasis on emo-tion regulation goals than do young adults, with olderadults having a greater motivation to derive emotionalmeaning from life and to maintain positive affect (e.g.,Mather & Carstensen, 2005). In the service of thesegoals, older adults may focus their attention on thingsthat will elicit pleasant feelings (Mather, 2006; Carstensen,Isaacowitz, & Charles, 1999) and may process positiveinformation in a more self-referential fashion (Gutchesset al., 2007). Although this argument has some empirical1Boston College,2Athinoula A. Martinos Center for BiomedicalImaging,3Harvard UniversityD 2008 Massachusetts Institute of Technology Journal of Cognitive Neuroscience 20:7, pp. 1161–1173support, behavioral data do not always support the con-clusion that older adults’ ‘‘positivity shift’’ results fromchanges in encoding processes. For example, when per-forming a digit parity task in which positive or negativedistractor words were presented in between the twonumbers that were to be evaluated, older adults did notshow a disproportionate interference effect from thepositive words compared to the negative words. Never-theless, the older adults did later remember the positivewords better than the negative words (Thomas & Hasher,2006). These findings suggest that older adults’ positivitybias may not arise from increased attention towardpositive items at the time of encoding. However, whenolder adults’ attentional resources are taxed during en-coding (by use of a divided attention manipulation), theirpositivity effect disappears (Mather & Knight, 2005),suggesting that there may be a link between the way inwhich older adults process positive versus negative atencoding that leads to their mnemonic benefit for posi-tive information.The present functional magnetic resonance imagingstudy used a subsequent-memory paradigm, in whichencoding-related activity to items of different valenceswas analyzed as a function of whether the items werelater remembered or forgotten (for more information onthe use of the subsequent-memory paradigm, see reviewby Paller & Wagner, 2002). Thus, the present study couldexamine whether there are age-related changes in theneural processes supporting the successful encoding ofemotional information, and whether such changes couldexplain older adults’ positivity shift. If age-relatedchanges in encoding processes correspond with thepositivity shift, then it should be possible to demon-strate age-related changes in the neural processes thatpredict subsequent memory for positive information.We therefore compared the neural processes that youngand older adults recruited as they encoded positive,negative, and neutral information. This design allowedus to examine