Emotion-Induced Retrograde Amnesia and Trait Anxiety Andrei C. Miu, Renata M. Heilman, Adrian Opre, Mircea Miclea Program of Cognitive Neuroscience, Department of Psychology, Babes ¸-Bolyai University Emotional arousal can both enhance and impair memory. Considering that both emotional memory and trait anxiety (TA) have been associated with adrenergic activity, the authors investigated whether there is an association between 2 opposite emotional memory biases and the TA. The authors used a procedure recently put forward by B. A. Strange, R. Hurlemann, and R. J. Dolan (2003) to elicit an emotion-induced retrograde amnesia (ERA) coupled to an emotional memory enhancement (EME). The authors contrasted the association between these emotional memory biases and the TA in several conditions involving different levels of encoding and types of recall. The results presented here indicated a significant interaction of the TA with EME and ERA and the dependency of these biases on the consciously controlled use of memory. Keywords: emotional memory, trait anxiety, level of encoding, type of recall Recognition of the multifaceted relation between emotion and memory launched not only a thousand research ships but provided an intellectual basis for the modern cognitive psychology move- ment that integrated emotional “hot” cognitions (e.g., David, Mi- clea, & Opre, 2004; David, Schnur, & Belloiu, 2002). Early influential studies such as those of Henri Pie ´ron and Sigmund Freud argued that emotion predominantly reinforces memory, which led to the fossilization of this view in contemporary psy- chology (for a review, see Borod, 2000). More recent research has shown that this theory is true but limited because there are cases when emotionality is associated not only with memory enhance- ments but also with memory decrements. Human participants usually remember better the emotionally arousing words or pic- tures than the neutral stimuli (Bradley, Greenwald, Petry, & Lang, 1992; LaBar & Phelps, 1998). However, whereas there is a lasting vivid memory for the gist of an emotional event, the memory for the peripheral details of the same event proves decremented (Adolphs, Denburg, & Tranel, 2001). Moreover, human partici- pants are less likely to remember the details of an event if it comes before another emotionally arousing event that creates a state of “mental shock” (Christianson & Loftus, 1987; Christianson, Lof- tus, Hoffman, & Loftus, 1991; Loftus & Burns, 1982). A topic related to the impairing influence of emotional arousal on memory has been independently approached by neuroscientists studying the involvement of the hypothalamo–pituitary system in stress- induced memory impairments (e.g., McGaugh & Roozendaal, 2002; Woodson, Macintosh, Fleshner, & Diamond, 2003). The amnestic potential of emotional arousal has remained nonetheless rather poorly explored in mainstream cognitive psychology until recently (Strange et al., 2003). We briefly describe below how several recent studies reopened this topic. Psychologists have traditionally viewed anxiety as a variant of emotional arousal (see Easterbrook, 1959; Miclea, 2001). Trait anxiety (TA) is a construct based on the assumption that even healthy individuals can differ from one another in their sensitivity to threat (Spielberger, 1983). This individual difference is related to the innate sympathovagal balance, that is, the nonlinear feed- back control exerted by the autonomic nervous system on heart rate and other physiological processes (e.g., LeBlanc, Ducharme, & Thompson, 2004; Maes et al., 2002; Mizuki, Suetsugi, Ushi- jima, & Yamada, 1996; Virtanen et al., 2003; Yu, Dimsdale, & Mills, 1999; for a review, see Kabbaj, 2004). TA has been accord- ingly included in personality inventories (for reviews, see Opre, 2001; Zuckerman, 1991). Participants high in TA display signifi- cantly longer solution times and lower accuracy scores for time- limited abstract reasoning tasks, and they usually obtain lower scores on Raven’s Progressive Matrices (Leon & Revelle, 1985; Mandler & Sarason, 1952; Mayer, 1977; Siegman, 1956). The anxious participants use fewer cues and are less selective in distinguishing task-relevant from task-irrelevant cues, particularly in difficult tasks (Easterbrook, 1959). Another theory has asserted that anxious participants have self- rather than task-centered re- sponses and engage in task interference rather than task comple- tion responding (Mandler & Sarason, 1952). However, neither theory is representative of the performance of persons who differ in their level of TA (Leon & Revelle, 1985). Several cognitive models of emotion have theorized that TA is associated with an attentional and memory bias for threat-related (worry) information because this type of stimuli relates more to the personal preoccupations of individuals high on this trait (Reidy & Richards, 1997a, 1997b). The studies that tested this association on both clinical and nonclinical high-TA samples have obtained equivocal results (Reidy & Richards, 1997a, 1997b; Richards & French, 1991). Some authors have attributed the inconsistency of these results to the type of emotional stimuli used, which were not always relevant or specific to the form of anxiety under investi- gation (Reidy & Richards, 1997b). In a recent study, Reidy (2004) Andrei C. Miu, Renata M. Heilman, Adrian Opre, and Mircea Miclea, Program of Cognitive Neuroscience, Department of Psychology, Babes ¸- Bolyai University, Cluj-Napoca, Romania. This research was supported by Grant E33 from the Romanian Univer- sity Research Council, Bucharest, Romania. We thank Laura Petra and Alina A. Miu for critically reading and significantly improving the style of the article. Correspondence concerning this article should be addressed to Andrei C. Miu, Department of Psychology, Babes ¸-Bolyai University, 37 Republicii Street, Cluj-Napoca CJ 400015, Romania. E-mail: andreimiu@psychology.ro Journal of Experimental Psychology: Copyright 2005 by the American Psychological Association Learning, Memory, and Cognition 2005, Vol. 31, No. 6, 1250 –1257 0278-7393/05/$12.00 DOI: 10.1037/0278-7393.31.6.1250 1250