The use of crying over spilled milk: a note on the rationality and functionality of regret

This article deals with the rationality and functionality of the existence of regret and its in ̄ uence on decision making. First, regret is de® ned as a negative, cognitively based emotion that we experience when realizing or imagining that our present situation would have been better had we acted differently. Next, it is discussed whether this experience can be considered rational and it is argued that rationality only applies to what we do with our regrets, not to the experience itself. Then, research is reviewed showing that both the anticipation of future regret and the experience of retrospective regret in ̄ uence behavior. The in ̄ uence of anticipated regret can be considered rational as long as the decision maker can accurately predict the regret that may result from the decision. The in ̄ uence of experienced regret cannot be considered rational, since decisions should be based on future outcomes, not historical ones. However, in ̄ uence of experienced regret can be called functional since it may result in increased learning from our mistakes. The standard version of the rational choice model has proven its usefulness for understanding and predicting human behavior. But it remains seriously incomplete. Most analysts regard a irrationalo behavior motivated by the emotions as lying beyond the scope of the model. But it is neither necessary nor productive to adopt this view. With careful attention to the things people care about, we can greatly enrich our understanding of why we behave as we do. (Frank, 1988, p. 783) Theories of rational choice view decision making as a cold cognitive process. Decision makers are supposed to rationally calculate for all possible courses of action the utility of each possible outcome, and weigh the utilities with the probability that each outcome will occur. They then choose the course of action that provides them with the highest (expected) utility. Emotions are neglected in these theories. In reality, however, decision outcomes are known to be powerful antecedents of emotional experiences and these emotions may well in ̄ uence the Marcel Zeelenberg, Tilburg University, Department of Marketing, PO Box 90153, 5000 LE Tilburg, The Netherlands; tel: 1 31-13 466 8214; fax: 1 31-13 466 2875; e-mail: [email protected] ISSN 0951-5089 (print) ISSN 1465-394X (online)/99/030325± 15 Ó 1999 Taylor & Francis Ltd 326 MARCEL ZEELENBERG choices we make. This article deals with the rationality and functionality of one speci® c emotion, regret. Regret is the negative, cognitively based emotion that we experience when realizing or imagining that our present situation would have been better had we acted differently. It is the typical emotion we may feel when decisions go awry [1]. Should regret, or more general emotions, be included in our theories of rational choice? The quotation with which this article started suggests we should. The present article elaborates on that recommendation. Several alternative approaches to decision making that take the in ̄ uence of regret into account have been developed. For example, in the 1950s, researchers argued that we sometimes base our decisions on a aminimax regreto principle (e.g. Luce & Raiffa, 1957; Savage, 1951; see also Acker’s 1997 approach of tempered regrets). This principle holds that one computes the maximum of possible regret (de® ned as the difference between the actual outcome of the chosen option and the highest possible outcome of the rejected options) for each option, and then chooses the option where this maximum regret is smallest. The minimax regret rule is useful when there is no knowledge whatsoever about the probabilities of the possible outcomes, because this information is not needed and not taken into account even when it is present. More recently, the economists Bell (1982) and Loomes and Sugden (1982) formulated decision theories that also take the probability of regret into account. The main assumptions of their Regret Theory have been supported in empirical research. These assumptions are that we may experience emotions as a consequence of our decisions. Decision makers experience regret when the outcome of the rejected option would have been better, and rejoicing when the outcome of the rejected option would have been worse (e.g. Mellers et al., 1997; Zeelenberg et al., 1998d). These emotions have an impact on how we evaluate decision outcomes (e.g. Inman et al., 1997). And ® nally, this impact of regret is taken into account before we decide, and thus may play an important role in determining what we choose (a detailed account of this in ̄ uence is provided later in this article). For example, a decision to buy a particular house can result in enormous regret if shortly after the act of buying, house prices drop dramatically. Consequently we may enjoy living in that house less than we would have otherwise. Even in more day to day decision making emotional experiences may play a role. For example, going to the cinemas and seeing an awful movie can really get us down, especially when we later learn that another movie that we considered was extremely good. Finally, even trivial decisions, such as choosing which cash register to line up at in a supermarket, can produce feelings of regret when the line we are in does not proceed as quickly as others do. Anticipations of these emotional reactions may in ̄ uence what we choose, which particular house, movie, or cash register, but also how we choose. That is, negative emotions may prompt us to delay decisions (Beattie et al., 1994) and in ̄ uence the amount and direction of pre-decisional information search (Luce, 1998). Thus, when making decisions we not only predict the utility that will be provided by these options, as assumed in rational choice theory, we also predict the emotions that arise from comparing the result of that option with the results of options forgone. RATIONALITY AND REGRET 327 In this article I review recent developments in the psychology of regret and discuss how regret in ̄ uences our decision making. In particular, I address the issue of whether it is rational to let regret in ̄ uence our choices. First, I consider the experience of regret itself.