Abstract: Many consumption decisions are based on hedonic expectations (“Would it be enjoyable?”) and/or hedonic memories of previous consumption episodes (“Was it enjoyable?”). Psychological research suggests, however, that expectations as well as memories may deviate from consumers’ actual experiences. This research focuses on discrepancies in consumers’ consumption-related expectations, memories and actual experiences, and the role of these variables in consumer decision making and implications for consumer learning. In particular, we argue that consumers’ expected and remembered experience converge but may deviate from their actual consumption ex-perience because they draw on different sources of information. Modified from Robinson and Clore’s (2002) accessibility framework, our conceptual framework extends to consumption experience in general and investi-gates the conditions under which consumers are likely to learn or not learn from their actual experience.
One hundred and twenty six undergraduate students participated in this study and were randomly assigned to one of the three experimental conditions. Participants in the predict condition were asked to predict the taste of a well-known (category leader) brand of snack (i.e., ready-to-eat sausage) vs. a less well-known brand (without actually tasting the food) and indicate their preference and choice of the two brands. Participants in the experience condition first tasted the two brands of sausages and then indicated their preference and choice ac-cordingly. Participants in the experience with delay condition were asked to first taste the two brands of sausages and then worked on some unrelated tasks, and finally were asked to recall their taste experience and indicate their preference and choice of the two brands of sausages. Our results show that memories of taste experience parallels with expectations (i.e., well-known brand sausage tastes better than the less known brand), as reported by the participants in the experience with delay and the predict condition. Similarly, participants in the two con-ditions also indicated similar preference pattern (well-known brand is preferred over the less known brand). However, reports of actual taste of the two brands did not differ in the experience condition and participants are as likely to prefer and choose the less well-known brand as the well-known brand. Our findings suggest that memories are likely to confirm expectations and may have little to do with the actual experience. Furthermore, because decisions are often made on the basis of expectation or memory, consumers may rarely learn from their actual experience, unless judgment and decision are made on the spot of consumption and consumers are en-couraged to revise their prior beliefs or expectations.
These findings have important implications for consumer decision making and consumer well being. Be-cause hedonic predictions converge with memories but both have little to do with the actual experience, con-sumers may fail to optimize their actual consumption experience. Moreover, these processes impede learning from experience and may result in the erroneous impression that one’s expectations were right on target.

Key words: consumption experience, expectation, retrospective memory, consumer decision making, con-sumer learning