Tempting Fate

Anticipated Regret and Perceived Probability

We have claimed that people’s tendency to avoid abnormal actions could reflect the regret they anticipate feeling if misfortune results, and that it need not reflect their memory-distorted beliefs about the likelihood of the misfortune’s occurring. Thus, we argued that the greater reluctance to sell a lottery ticket as the drawing approached did not derive from people’s memories of their experiences with selling lottery tickets, but from their subjective utility function: They perceived that the closer the drawing was when they parted with a winning lottery ticket the more intense their regret would be. We confess, however, that it is very difficult to rule out the presence of memory bias even in the lottery example. The difficulty arises with specifying precisely what principle it is that experience teaches. It is true people might not have previous experience with selling lottery tickets, but this does not mean they have not had previous experience with similar situations; that is, situations in which they have experienced a misfortune after engaging in a highly mutable action. The lottery situation might fit the template of a more general situation, and therefore, evoke a more general principle; for example, “Don’t tempt fate.” Perhaps the experience of having unwisely changed a multiple-choice answer at the last minute would be sufficient for people to fear that selling a lottery ticket at the last minute might be similarly ill-fated. It is left to other researchers to pursue more vigorously the possibility that subjective utility functions alone can deter people from pursuing a highly mutable course of action.

Establishing the independent role of anticipatory regret is not easy. The most straightforward means of assessing the independent power of anticipatory regret is to show that people avoid an abnormal action even if they do not believe that the action is any more likely than another to lead to a negative outcome. This standard, however, may prove too strong a test for the anticipatory-regret hypothesis. The problem is this: One of the effects of anticipating that an event will make a person feel bad is that its occurrence seems more likely (cf. Johnson & Tversky, 1983). Subjective expectancies are influenced by subjective utilities, not just by recollected experiences. Thus, as the dread of a negative outcome occurring through a particular action increases so will its subjective probability. There are at least two reasons for this pattern. First, people consider their feelings to be information (Miller & Turnbull, 1990; Schwarz, 1990), and therefore are disposed to overestimate the likelihood of a negative outcome based on their fear of its happening. Second, people estimate the probability of an event at least partly by the vividness with which they can imagine it (Carroll, 1978; Sherman, Cialdini, Schwartzman, & Reynolds, 1985). Thus, anticipating the counterfactual thoughts that a negative event would evoke, because it tends to make the event more vivid, tends to make it more subjectively likely.

In the lottery scenario, Mr. K. receives an offer to buy his ticket a week before the drawing; Mr. T. receives the offer an hour before it. Might the fear these men have of experiencing self-recrimination in the event they sold a winning ticket affect their estimates of the chances that a sold ticket will win? To examine this question, we provided a group of participants with this lottery scenario and asked them the following three questions:

Who would be most optimistic of winning if he resisted selling his ticket?

Who would have the strongest expectation that his ticket would win if he were to sell it?

Who would have the greatest sense that he was tempting fate if he were to sell his ticket?

The results indicated that participants believed there would be no difference in the two men’s hopes of winning were they to refuse the friend’s offer. However, participants predicted that the two men would have different expectancies were they to accept the friend’s offer. Mr. T. (the man who sold his ticket closest to the drawing) was expected to be more optimistic that his (former) ticket would win. This result suggests that the participants’ prediction that Mr. T. would show greater reluctance to sell his ticket was not guided by their belief that his retained ticket is more likely to win. Rather, it seems guided by their belief that his sold ticket would be more likely to win. The decision not to sell, then, is just that – a decision not to part with the ticket, not a decision to keep the ticket. The ticket’s capacity to yield satisfaction is not assumed to be enhanced by one’s retaining it, even though its capacity to yield dissatisfaction is assumed to be enhanced by one’s selling it.

To undertake a risky action one easily imagined otherwise (such as selling a lottery ticket close to the drawing) is to tempt fate. We contend that one tempts fate not only when one opts for one action when there is an alternative action that makes more sense or has a better chance of a positive outcome, but also any time one takes an action that will be easy to undo mentally. To say, “It would be just my luck if I switched from this line to the other, that this line will speed up and the other will slow down” is not to commit oneself to the proposition that not switching means one’s line is likely to speed up. One is just “asking for it” if one takes an action he or she knows, in retrospect, will be easy to imagine otherwise. Similarly, to insist on wearing one’s lucky shoes to the final exam is not necessarily to commit oneself to the proposition that he or she will do better than others who do not own lucky shoes. Wearing the lucky shoes might not even commit the person to the proposition that he or she would do worse if he or she did not wear the lucky shoes. It might commit the person only to the proposition that he or she is just asking for it by not wearing them. If things turn out badly after the person chooses not to wear his or her lucky shoes, he or she will be forced to endure intense self-recriminations. In essence, having lucky shoes means one also has one or more pairs of “unlucky” shoes, the wearing of which will leave him or her fearful of impending disaster and vulnerable to relentless self-recrimination.