Ruth Byrne's The Rational Imagination (Reference Byrne2005) is an outstanding work that provides the clearest answers thus far to long-standing questions about counterfactuals such as “What is most mutable?” and “Why is this so?” Thus, in the spirit of broadening the perspective offered by Byrne, we will merely choose to quibble with a single phrase that appears in the book, specifically, “counterfactual thinking requires a comparison of the facts with the imagined alternative” (p. 122). In our view, this assumption, shared by most if not all scholars who study counterfactual thinking and conditional reasoning, is unnecessarily restrictive. Rather, we have suggested (e.g., see Markman & McMullen Reference Markman and McMullen2003; Reference Markman, McMullen, Mandel, Hilton and Catellani2005; McMullen Reference McMullen1997; McMullen & Markman Reference McMullen and Markman2000), and continue to maintain, that individuals do not necessarily engage in counterfactual simulations exclusively to evaluate factual reality. Instead, comparative evaluation may be, and often is, suspended in favor of experiencing the counterfactual simulation as if it were real.
Markman and McMullen (Reference Markman and McMullen2003) proposed a Reflection and Evaluation Model (REM) to account for how counterfactuals can have either contrastive (i.e., displaced away from a counterfactual standard) or assimilative (i.e., displaced toward a counterfactual standard) effects on affect and judgments. The model asserts that two psychologically distinct modes of mental simulation operate during counterfactual thinking: reflection, an experiential, “as if” mode in which counterfactual possibilities are vividly simulated, and evaluation, a comparative mode in which counterfactual standards are used as a reference point against which to evaluate factual reality. Reflection renders standard-consistent cognitions accessible, thereby yielding assimilation, whereas evaluation yields contrast because counterfactuals are used as a standard against which to compare factual reality.
To illustrate, consider how an individual who just missed being killed in a plane crash may experience a great deal of anguish by dwelling on what might have happened (assimilation) (see also Oettingen Reference Oettingen, Gollwitzer and Bargh1996), but at another time feel quite lucky by focusing on the fact that they are, by good fortune, alive today (contrast). Similarly, imagining having won the lottery might elicit positive affect by one's fantasizing about having a great deal of money (assimilation) so long as one suspends their evaluation of the fact that they do not have a great deal of money. On the other hand, evaluation of this fact in light of the counterfactual should lead one's perception of their present state to seem even more impoverished (contrast).
Byrne (Reference Byrne2005) describes an interesting study conducted by Thompson and Byrne (Reference Thompson and Byrne2002) that examined whether there are differences among individuals in their willingness or ability to consider more than one possibility when confronted with a counterfactual conditional. Although the majority of participants kept in mind two possibilities – the presupposed facts and the counterfactual conjecture – about a quarter of the participants kept in mind only a single possibility – the presupposed facts – when they contemplated the conditional. In our view, however, this work neglects to mention a third way that individuals may entertain counterfactual assertions: keeping only the false possibility in mind. Moreover, engaging in such an “as if” type of simulation should have assimilative effects on subsequent responses and judgments. Is there empirical evidence to support such a claim?
An early and clear demonstration was provided by McMullen (Reference McMullen1997), who asked participants to recall a somewhat negative event in their own lives and imagine how things could have turned out better (upward counterfactual) or worse (downward counterfactual) than they actually did. Participants in the reflection condition were then instructed to “vividly imagine what might have happened instead,” whereas those in the evaluation condition were instructed to “vividly imagine the event and what might have happened instead.” Providing clear evidence for affective assimilation, participants in the reflection condition reported positive affect after making upward counterfactuals and negative affect after making downward counterfactuals, whereas this pattern was reversed in the evaluation condition, thereby evidencing affective contrast. Of course, it may be argued that comparison must still be the default effect of a counterfactual (cf. Roese et al. Reference Roese, Sanna, Galinsky, Hassin, Uleman and Bargh2005), and that affective assimilation is only possible under specific conditions. However, McMullen's data are not consistent with that conclusion. When instructed to generate a counterfactual and then vividly imagine that counterfactual, participants exhibited an affective assimilation effect; contrast effects only emerged for those who explicitly evaluated their factual events. This suggests that counterfactuals are not contrasted with their corresponding factual events by default, but rather the explicit evaluation of the factual event must first be made.
How can a counterfactual, which is by definition an alteration of a factual event, not by default act as a contrast to that factual event? Put another way, have we chosen to define counterfactuals too broadly? Are they truly counterfactuals if individuals treat possibilities as fantasies (either positive or negative) rather than as standards of comparison? Consider that McMullen's (Reference McMullen1997) participants were simply instructed to ‘‘think of how something different could have happened rather than what actually happened.’' Only following this counterfactual generation was the reflection/evaluation manipulation initiated. Thus, equivalent simulations were shown to yield both affective assimilation and contrast. In turn, one might argue that an obviously unreal alternative should be subjectively perceived as distinct from a real event, but the work of Johnson and Raye (Reference Johnson and Raye1981) and Anderson (Reference Anderson1984) suggests that the representations of real and imagined events share many characteristics, and that imagined events are sometimes confused with real events. Similarly, Gilbert et al. (Reference Gilbert, Krull and Malone1990) suggested that, in order to be comprehended, a proposition, even an obviously false one, must first be accepted and treated as if it were true. Hence, perhaps a counterfactual, before it can be used as a standard of comparison, must first be comprehended in a manner that treats that counterfactual as if it were true, and only then can it be used as an evaluative standard. In this way, then, an initial assimilation effect would later give way to a contrast effect. In all, we hope that drawing attention to “as if” processing of mental simulations can broaden the theoretical landscape so artfully portrayed in Byrne's The Rational Imagination.
