Addressing the question of how people create alternatives to reality calls also for the question of why they should do so. Byrne (Reference Byrne2005) provides an excellent overview of the counterfactual thinking literature that is primarily focused on the first question. Admittedly, she goes quite some way into answering the second question, as well. However, in my view, the choice to emphasize mostly the mechanisms by which counterfactuals are created, causes us to miss out on the perhaps most relevant consequence of counterfactual generation. Counterfactuals are often a source of emotion, and it is this emotional reaction that drives many of the implications of counterfactual thinking. For example, counterfactuals serve a clear learning function, as Byrne notes on several occasions. Being aware of an upward counterfactual and the way in which it could have been obtained can clearly help one to learn from mistakes and to do better next time. However, I propose that the awareness of the counterfactual is motivating precisely because we become emotionally aroused by it.

Some of the emotional correlates of counterfactual thinking are discussed in the The Rational Imagination (Byrne Reference Byrne2005). But, the function of these emotional reactions, the role they serve in instigating behavior, have fallen beyond the scope of the book. In my review, I build upon our own work on the causes and consequences of what I would call the prototypical counterfactual emotion, regret (for a review, see Zeelenberg & Pieters Reference Zeelenberg and Pieters2007), and on our recent attempt to formulate a pragmatic approach to the role of emotions in behavior (Zeelenberg & Pieters Reference Zeelenberg, Pieters, de Cremer, Zeelenberg and Murnighan2006). The most important point I want to make is that to fully understand counterfactual generation, one not only needs to know how counterfactuals are constructed, but also how they elicit emotions and subsequently how these counterfactual emotions drive our future choices. The mechanisms reviewed by Byrne, which are captured in a set of seven principles (see p. 200), address counterfactual generation, but remain relatively mute with respect to the behavioral consequences.

We have been studying the relation between counterfactual thinking and the specific emotions regret and disappointment (Zeelenberg et al. Reference Zeelenberg, van Dijk, Van der Pligt, Manstead, Van Empelen and Reinderman1998b). Although both stem from a comparison between “what is” and “what might have been,” regret originates from comparisons between the factual decision outcome and a counterfactual outcome that might have been, had one chosen differently. Disappointment originates from a comparison between the factual decision outcome and a counterfactual outcome that might have been, had another state of the world occurred. Building on this, we realized that one should be able to shape the emotional response to a negative outcome by giving direction to the counterfactuals that are generated. In one study, participants were asked to imagine themselves being in a situation that resulted in a bad outcome that was produced by various elements in the situation, including the protagonist's own choices and also uncontrollable aspects of the situation. Thus, the situation could elicit either regret or disappointment (or a combination of the two). Next, participants were instructed to mentally undo the event either by mutating aspects of their behavior, or by mutating aspects of the situation. Next, they indicated the regret and disappointment they would feel in that situation. Interestingly, participants indicated feeling significantly more regret than disappointment when they mutated their behavior, and the opposite occurred when they mutated situational aspects. The results of this study clearly indicate the malleability of counterfactual generation and the consequences of it. Further studies (reviewed in Zeelenberg & Pieters Reference Zeelenberg and Pieters2007) have shown that the type of counterfactual emotion experienced strongly influences how people deal with negative events.

The non-mechanistic relation between counterfactuals and emotions is also apparent from another series of experiments. Eric van Dijk and I (Van Dijk & Zeelenberg Reference Van Dijk and Zeelenberg2005) found that the presence of counterfactuals (i.e., feedback on the outcomes of unchosen alternatives) does not always produce equally strong emotions. We asked participants to imagine being at a small fair, at which they took part in an instant scratch-card lottery. They could choose one of the only two scratch cards that were left. Participants then learned that they had won a €15 liquor-store token or a €15 book-store token. After this, participants were informed that someone else, who bought the last remaining scratch card, won either a €50 book-store token or a €50 liquor-store token. We found that complex comparisons of outcomes that were in different product categories (i.e., you win a book-store token but miss out on a liquor-store token, or vice versa) resulted in less regret than did relatively easy comparisons. Put differently, reduced comparability of counterfactual and obtained outcome attenuates the emotional response that is evoked by the counterfactual. Interestingly, we also found that for people who have a natural tendency to engage in comparison processes, the emotional response to the counterfactual outcome was also intensified when it was complex (i.e., between categories). Thus, depending on the individual characteristics of the decision maker, the same counterfactual comparison can result in mild or severe emotional reactions.

This work discussed here is of course highly selective, idiosyncratic, and incomplete, and many other relevant articles have been published. What these two articles (Van Dijk & Zeelenberg 2005; Zeelenberg et al. Reference Zeelenberg, van Dijk, Van der Pligt, Manstead, Van Empelen and Reinderman1998b) show, however, is that the specific content of the counterfactual thoughts that people may generate is crucial in determining the emotional response that will follow from these thought processes. Importantly, emotions typically arise when one evaluates an event or outcome as relevant for one's concerns or preferences (Frijda Reference Frijda1986). Emotions do not only have a backward-looking informational function (signaling how we are doing with respect to our goals), but also a future-oriented motivational function (telling us what to do next). Emotions are motivational processes that prioritize certain goals and thereby mobilize and give direction to behavior. Different specific emotions thus motivate and facilitate different behavioral responses to the event that was initially responsible for the emotion. This could explain how and why different counterfactual thoughts, focusing on one's own behavior or on elements in the situation, ultimately, but via the experience of either regret or disappointment, result in completely different behaviors. In this way, the consequences of counterfactual thoughts provide input for understanding their origins. After all, the of emotions serves the purpose of behavioral regulation. Put differently, feeling is for doing.