Using Affect–Expectations Theory to Explain the Direction of the Impacts of Experiential Emotions on Satisfaction Philippe Aurier University of Montpellier Guergana Guintcheva EDHEC Business School ABSTRACT Traditional research in consumer psychology considers valence (negative vs. positive) to be the primary dimension of an emotion. The direction of the impact an emotion has on consumer satisfaction is considered to be dictated by its valence. Drawing on the theory of affective expectations, in the case of experiential emotions, an alternative conceptualization is proposed in which this direction is seen as being conditional on emotional expectations associated with the consumption experience. When pertaining to consumers’ expectations relative to an anticipated experience, negative emotions can be the output benefit of consumption and contribute positively to satisfaction. Conversely, positive emotions that run against expectations can be a negative output and contribute negatively to satisfaction. On the basis of a qualitative study and a quantitative survey of movie consumption in theaters, it is established that positive (joy) as well as negative (sadness and fear) experiential emotions are positively associated with film expectations, whereas a positive emotion (calmness) runs against these expectations. Contrary to what traditional research predicts, not only joy but also sadness and fear demonstrate positive impacts on satisfaction whereas calmness, a positive emotion, has a negative impact. © 2014 Wiley Periodicals, Inc. Traditional research in consumer behavior considers a symmetrical relationship between stimulus valence (appetitive vs. aversive), emotional valence (positive vs. negative) and their impact on subsequent evalua- tive judgments and behavior (Forgas, 1995; Schwarz & Clore, 1983). A stimulus perceived as appetitive is con- sidered to generate positive emotions that are supposed to have a positive impact on subsequent judgments such as satisfaction, and vice-versa. Most emotion– satisfaction models developed in the literature are built on this principle (Mano & Oliver, 1993; Oliver, 1993; Oliver, Rust, & Varki, 1997; Phillips & Baumgartner, 2002; Rust & Oliver, 2000; Westbrook, 1987; Westbrook & Oliver, 1991). This symmetrical view predicts that a Shakespearean drama could be perceived as an aver- sive stimulus, which generates negative emotions (sad- ness, fear) that decrease satisfaction. Conversely, a bur- lesque comedy (vaudeville) could be perceived as an ap- petitive stimulus that generates positive emotions and satisfaction. However, simple common sense runs con- trary to this view. Most people will consider the dra- matic and sad Shakespearean scenario as an appet- itive stimulus generating both positive and negative emotions such as joy, fear, and sadness that could both contribute to satisfaction. This calls into question the understanding of the mechanisms conditioning the di- rection of the impacts emotions have on satisfaction. The review of the consumer and psychology litera- tures shows that two lines of research have recently challenged the traditional conceptualizations on emo- tions. In the first one, the appraisal framework consid- ers that the meaning of an emotion is not valence alone. Each emotion carries specific systematic motivational and cognitive characteristics that fuel carryover to subsequent judgments and decisions (Roseman, 1991; Roseman & Evdokas, 2004; Smith & Ellsworth, 1985). Consequently, emotions of the same valence may have a different impact on subsequent evaluations (DeSteno, Petty, Wegener, & Rucker, 2000; Lerner & Keltner, 2000; Raghunathan & Pham, 1999). However, a cen- tral proposition of the appraisal framework is that a unique emotion is considered in a given context (Lerner & Keltner, 2000). In the second stream, research on emotion co-occurrence has examined cases where peo- ple experience positive and negative feelings (such as fear and happiness) during the same consumption ex- perience and integrate them into overall evaluative judgments (Andrade & Cohen, 2007; Lau-Gesk, 2005; Psychology and Marketing, Vol. 31(10): 900–913 (October 2014) View this article online at wileyonlinelibrary.com/journal/mar © 2014 Wiley Periodicals, Inc. DOI: 10.1002/mar.20742 900