Narrative and Emotion 1 Running title: NARRATIVE PERSPECTIVES The emotional impact of loss narratives: Event severity and narrative perspectives Tilmann Habermas and Verena Diel Goethe-University Frankfurt a. M., Germany Accepted for publication in Emotion, September 26 2009 © American Psychological Association http://www.apa.org/journals/emo/ This article may not exactly replicate the final version published in the APA journal. It is not the copy of record. Correspondence concerning this article should be addressed to Tilmann Habermas, Goethe- University, Institute for Psychology, Department of Psychoanalysis, Senckenberganlage 15, 60054 Frankfurt a. M., Germany. E-mail: tilmann.habermas@psych.uni-frankfurt.de Acknowledgments. We thank Andrea Silberstein for coding free responses, Claudius Stauffenberg for segmenting propositions, Anna Kenney for editorial assistance, and Peter Vorderer and Volker Hodapp for helpful comments on earlier versions of this paper. Narrative and Emotion 2 Abstract Out of the complex influences of event, narrative and listener characteristics on narrative emotions, this paper focuses on event severity, narrative perspectives, mood, and dispositions for emotion regulation and empathy. Event severity and perspective representation were systematically varied in sad autobiographical narratives to study their influence on quantity and quality of readers’ emotional response. Each of three stories were manipulated to contain elaborated perspectives, only the past protagonists’ perspective (dramatic narration), and very little perspectives at all (impersonal narration). We predicted that event severity influences the quantity of emotional response, while degree of perspective representation influences plausibility and whether emotional responses are sympathetic or interactional, i.e. directed against the narrator. Hypotheses were confirmed except for plausibility, and perspective representation had an effect only on anger against and dislike of the narrator. In a second study, impersonal narration evoked anger at and negative evaluations of the narrator which were related to blaming the narrator for showing too little emotional involvement. The generalizability of findings across emotions and implications for sharing of emotions in everyday and clinical settings are discussed.