Neuropsychologia 42 (2004) 791–800 Effects of Alzheimer disease on memory for verbal emotional information Elizabeth A. Kensinger a, , Alberta Anderson b , John H. Growdon b , Suzanne Corkin a a Department of Brain and Cognitive Sciences, Massachusetts Institute of Technology, MIT, Building NE20-392, 77 Massachusetts Ave., Cambridge, MA 02138, USA b Department of Neurology, Massachusetts General Hospital, Boston, MA, USA Received 19 March 2003; accepted 7 November 2003 Abstract In healthy young and older adults, emotional information is often better remembered than neutral information. It is an open question, however, whether emotional memory enhancement is blunted or preserved in Alzheimer disease (AD). Prior studies of emotional memory in AD have included small samples of patients. In addition, studies that failed to find an enhancement effect in AD used stimuli lacking semantic coherence (e.g. lists of unrelated words, some that were emotional and others that were neutral). To circumvent these limitations, the present study examined a large number of AD patients (N = 80) and investigated whether AD patients would show better memory for a verbal description of an emotional event as compared to a neutral one. AD patients were equivalent to young and older control participants in rating the emotional descriptions for valence and arousal. Unlike the control groups, however, memory in AD patients did not benefit from the emotional narratives. We conclude that AD disrupts memory enhancement for at least some types of verbal emotional information. © 2003 Elsevier Ltd. All rights reserved. Keywords: Alzheimer disease (AD); Emotional memory 1. Introduction The panoply of sensory input that we experience in daily life is often infused with emotional relevance. This emo- tional information is usually better remembered than neu- tral information. An extreme example of this enhancement is the formation of a “flashbulb memory,” in which individ- uals retain a particularly vivid memory of an exceptionally emotional event (Brown & Kulik, 1977). This enhancement, however, can also occur in the laboratory, using more con- trolled stimuli and encoding conditions: Emotional memory enhancement has been demonstrated using a range of stim- uli, including words, sentences, pictures, and narrated slide shows (see Hamann, 2001; Buchanan & Adolphs, 2003, for reviews). Numerous cognitive and neural processes likely con- tribute to emotional memory enhancement. At encoding, individuals may elaborate on emotional information more than neutral information. They may be prone to associate emotional items with additional semantic information, or with autobiographical experiences. These encoding strate- gies could lead to a richer representation of emotional Corresponding author. Tel.: +1-617-253-7437; fax: +1-617-253-1504. E-mail address: ekensing@alum.mit.edu (E.A. Kensinger). information as compared to neutral information, and these differences in the richness, or distinctiveness, of the mem- ory could underlie the enhancement effect (see Doerksen & Shimamura, 2001; Ochsner, 2000; Kensinger & Corkin, in press; Kensinger & Corkin, 2003a, for evidence that emotional information is more vividly remembered than neutral information). Individuals also rehearse emotional information more than neutral information; this increased rumination on emotional events could mediate the enhance- ment effect (Christianson & Engelberg, 1999). Emotion may also exert effects at retrieval: emotion may serve as a retrieval cue; for example, a person may initially remember how they felt about an event, and that cue may then allow them to generate additional features about the event. Thus, there may be additional support present for the retrieval of emotional as compared to neutral information. Another critical factor underlying the emotional memory enhancement effect concerns the modulation of consolida- tion processes: emotional information may be more likely to be consolidated than neutral information, thus increasing the likelihood that emotional information will be retained over a delay. This modulation of consolidation may be me- diated by the amygdala (see McGaugh, 2000, for review). Numerous lines of evidence converge on the conclusion that the amygdala is critical for the emotional memory 0028-3932/$ – see front matter © 2003 Elsevier Ltd. All rights reserved. doi:10.1016/j.neuropsychologia.2003.11.011