Past and future regret and missed opportunities: an experimental approach on separate evaluation and different time frames

“All negativity is caused by an accumulation of psychological time and denial of the present. Unease, anxiety, tension, stress worry – all forms of fear – are caused by too much future and not enough presence. Guilt, regret, resentment, grievances, sadness, bitterness, and all forms of nonforgiveness are caused by too much past, and not enough presence .” (Tolle, 2012, p. 127)

According to Tolle (2012), spiritual teacher and author, the constant need of the mind to remember the past and to think about the future is the main factor that prevents us from experiencing inner peace and personal happiness. It is in fact a very special ability that is still assumed to be a unique ability in which only humans can excel (Cheke & Clayton, 2010; Roberts, 2002). Even though it is responsible for emotional ups and downs, such as joy or regret, and prevents us from experiencing this inner peace, it still has a positive aspect as it drives us to constantly improve and learn from our successes and mistakes (Epstude & Roese, 2008; Roese, 1997). This, in turn, increases the quality of future decisions and enables overall improved living conditions (Zeelenberg, 1999).

The decisions that everyone has to make—such as those involving career, private life, or even leisure time—could involve strong emotional experiences. These includes feelings about the respective decision, such as fear, anxiety about the consequences, and regarding the expectations about feelings that may occur after the outcome, such as relief or sadness. Of all the emotions a person can experience, regret has received the most attention in past research (Connolly & Zeelenberg, 2002). Zeelenberg and Pieters (2007) showed this by collecting academic publications on the topic of regret from 1945 to 2005, and they concluded that from the 1990s, publications on this topic increased dramatically. This clearly shows an increased interest in regret over the last decades.

One reason for the increased interest is that regret has a significant influence on decision-making. It is therefore one of the most frequently experienced emotions in human emotional life and especially occurs when individuals hold themselves personally responsible for the undesirable outcome (Connolly & Zeelenberg, 2002; Zeelenberg, 1999).

This study aims to shed further light on the findings of previous research by Shani, Danziger, and Zeelenberg (2015). As a first step, an overview of important theoretical concepts for the topic is presented. In an experimental study, we seek to extend the findings of Shani et al. (2015) by modifying some of their experiments in order to add new insights and variables to the current literature in this field of research. The overall question is whether missed opportunities in the future elicit more regret than missed opportunities in the past.

When participants only knew about the past discount that they had missed, they felt more regret for purchasing from this importer (M _{past first/2 weeks} = − 0.59, SD = 0.59; M _{past first/6 weeks} = − 0.51, SD = 0.70) than when they heard about a future discount that they will miss (M _{future first/2 weeks} = 0.18, SD = 0.84; M _{future first/6 weeks} = 0.16, SD = 0.88). Evaluating the regret scores for each discount separately, the past miss elicits more regret than the future miss does. Our prediction that future miss would be equally strong in a separate evaluation as it was in a joint evaluation could not be statistically confirmed.

We found a significant difference between the regret scores for the scenarios with one discount and the subsequent scenarios with two discounts within all four-condition groups. Mean regret scores for all conditions can be seen in Fig. 1. Table 2 indicates the test results and p values for the before and after regret scores in each group. It is clear that the mean regret scores are significantly different when introducing the second discount. Therefore, the mean difference is greater when participants first hear about the missed discount in the past and then about the future discount that they will miss (mean diff._{past first/2 weeks} = 0.88; mean diff._{past first/6 weeks} = 0.54). In comparison, the mean differences were lower when participants first knew about the missed discount in the future and then about the missed discount in the past (mean diff._{future first/2 weeks} = − 0.61; mean diff._{future first/6 weeks} = − 0.7). These results show that future miss has a stronger influence on changing people’s decisions.

	Past first/2 weeks	Past first/6 weeks	Future first/2 weeks	Future first/6 weeks
U	− 4.912a	− 3.707a	− 3.642b	− 3.610b
Sig. (two-sided)	.000	.000	.000	.000

^aBased on negative ranks

^bBased on positive ranks

We found no significant effect of missing the future opportunity in 2 weeks’ time or in 6 weeks’ time on regret. Participants who only knew about the future discount in 2 weeks did regret purchasing from this importer slightly more (M = .18) than participants who knew solely about a future discount in 6 weeks (M = .16). However, as indicated in Table 5, this difference was not statistically significant. The prediction that future misses with a longer time distance would cause less regret than misses in a nearer future could not be statistically verified.

We found significant differences between the two responsibility mean scores within the past first and future first conditions with a 6-week time distance. We also found a marginally significant difference within the past first condition with the 2-week time distance. Figure 2 depicts the mean responsibility scores for all four conditions. Table 6 depicts the before and after responsibility mean differences in each group and their significance levels.

	Past first/2 weeks	Past first/6 weeks	Future first/2 weeks	Future first/6 weeks
U	− 1.770a	− 2.469a	− 1.528b	− .903b
Sig. (two-sided)	.077	.014	.127	.014

^aBased on negative ranks

^bBased on positive ranks

The overall picture of the responsibility scores is very similar to the regret scores (Figs. 1 and 2). When participants felt more regret for purchasing from the importer that offered a past discount, this same importer made them feel more responsible for missing the discount. When participants changed their mind about their regret feelings after the new information of a future discount, they were also more inclined to feel more responsible for missing the new discount. These similarities of the regret and responsibility mean score movements apply also to the conditions where the future discount was introduced first.

Between the four conditions, there was predominantly no significant difference in the mean scores. There was only a significant difference in the scenarios with one missed opportunity between the groups Past first/6 weeks and future first/2 weeks (mean diff. = − .425, SE = .161, p = .047, 95% CI = .00, .85). The difference of variances between the four groups for the responsibility scores is therefore assumed to be mostly within the groups. In the matter of responsibility feelings, people seem to have different perceptions or are more indifferent to feeling responsible about missing the discount from one importer or the other.

We found a significant and marginally significant difference between the preference scores for the scenarios with one discount and the subsequent scenarios with two discounts in the past first conditions. Figure 3 depicts the mean preference scores for all four conditions. Table 7 summarizes the test results and p values for the before and after preference scores in each group.

	Past first/2 weeks	Past first/6 weeks	Future first/2 weeks	Future first/6 weeks
U	− 1.915a	− 2.043a	− .243b	− .500a
Sig. (two-sided)	.056	.041	.808	.617

^aBased on positive ranks

^bBased on negative ranks

It can be observed that in the past first conditions, the participants changed their mind after they read the second scenario. First, they preferred to purchase from the importer that did not offer a discount in the past (M _{past first/2 weeks} = .18; M _{past first/6 weeks} = .12) and then they preferred to purchase from the importer that did offer a discount in the past rather than purchasing from the importer that is related to a future miss (M _{past first/2 weeks} = − .20; M _{past first/6 weeks} = − .25). Comparing with the regret scores (Fig. 1), the participants seemed to prefer purchasing from the importer that does not elicit as much regret.

The mean scores in the future first conditions, however, present a different picture. Participants did not change their minds when they additionally read about a past discount that they have missed. In both scenario sequences, they preferred to purchase from the importer that will offer the mug in the future at a lower price even though they felt initially more regret for purchasing from this importer. It is noteworthy that these regret scores were very close to zero (M _{future first/2 weeks} = .18; M _{future first/6 weeks} = .16), which indicates that they were more temped to be indifferent to the importer from which they more regret to purchase. This could be an explanation for the preference choices that do not show any clear decision-making to prevent a future miss or past miss.

In all four groups, the tendency was towards preferring to purchase the mug from the importer B. There were consequently no significant differences in the preference means at the first scenario round, F(3, 212) = .788, p = .502. In the past first conditions, it can be assumed that the participants chose this importer because, at that point, it was the only choice without any discount related opportunities. So they could easily distance themselves from the importer that offered a discount in the past. In the future first conditions, they preferred to purchase from the importer that offered a discount in the future. It seems that there was no strategy to distance themselves from the future miss when this is the only miss of which they are aware.

After the introduction of the second scenario, the preference mean scores differed significantly between the four conditions, F(3, 211) = 5.962, p < .01). A post hoc test shows that there were mostly significant differences between all conditions if the only difference was not the time variable. There was one marginal difference between past first/2 weeks and future first/6 weeks (mean diff. = − .480, SE = .191, p = .064, 95% CI = − .98, .02). It can therefore be assumed that the variance of differences is mainly between the groups. In the past first conditions, people preferred purchasing from the importer connected to a past miss, and they therefore distance themselves further from the future miss. In the future first conditions, the participants did not try to distance themselves from the future miss and preferred purchasing from the importer that offers a discount in the future.

Shani et al. (2015) are presumably the first authors to draw a direct comparison between regret connected to a past miss and regret related to a future miss. They showed that a possible future regret could influence people’s decision-making more than regret caused by a past miss. We tried to shed further light on these findings by testing the strength of their findings in a separate evaluation and different time variables. By presenting to the participant two sequential scenarios with only one discount in the first scenario and two discounts in the second, we were able to collect data for a separate evaluation and a joint evaluation with different orders of the missed opportunities. Thus, we could also evaluate to what extent they changed their minds when introducing a future regret as new information.

In the separate evaluation, we find that people feel more regret and responsibility for a past miss than for a future miss. This remains contrary to the findings of the study to which we are referring. Furthermore, regarding their purchase preferences, participants chose more to purchase from an importer connected to a future miss than from an importer that offered a discount in the past. This also remains in sharp contrast to the findings of the previous study. However, a future miss as second information had more influence on changing their mind about their feelings of regret than the new knowledge about a past miss did. This is informative for the strength of the Future Opportunity Principle and fulfills the prediction that a future miss is more impactful. Regarding the different periods, we found that future misses with a longer time distance did not cause less regret than misses in the nearer future did. One reason could be that the two periods were not large enough to present significant differences in the regret means. Another possible explanation is that the impact of a future miss is so strong that the feelings of regret remain stable regarding the period or are even locked to other external influences. These results have clear practical implications for both consumers and companies (as decision-making actors). Specifically, companies (e.g., retailers) should be very careful when advertising future opportunities for products or services for which justifying a future miss is difficult, as consumers may find it easier to justify missing a future opportunity for offerings that satisfy utilitarian goals than for those that satisfy hedonic goals (Shani et al., 2015).

A limitation of this research resides in the study power. Although our sample size was large enough to test for our predictions, not all of them were totally confirmed. Both parametric and non-parametric statistical analyses were conducted, in line with the previous studies of Shani et al. (2015). Future research could seek to test our predictions with a larger and more diversified sample. Specifically, future research could examine different decision-making individual profiles regarding future and past misses. For example, maximizers are known as decision makers who are as concerned with the potential loss of existing options as with the loss of undiscovered future ones (Moyano-Díaz, Martínez-Molina, & Ponce, 2014; Patalano, Weizenbaum, Lolli, & Anderson, 2015). Future research could also seek to increase the difference between the periods and further test the possibility to influence the feelings of regret with the time variable. This could be important for the corporate sector, should companies need to decide how far in advance they can announce a discount. If customers feel too much regret for missing the future discount, they will have negative associations with the product and enjoy the product less. Also, future research could seek to carry out other statistical analyses, such as moderation and mediation analyses with additional variables, as well as to adopt longitudinal randomized experiments.

The preference choices showed that participants chose to purchase mostly from the importer that will offer a discount in the future. This was also the case when they felt more regret for purchasing from this importer and the alternative was an importer that did not offer any discounts. One reason could be that the participants misunderstood the instruction and thought that they could buy the mug in the future. Shani et al. (2015) mentioned that the participants could have difficulties to imagine losing a future discount opportunity, even to the extent that they are less aware or are completely unaware of the information that they do not have the possibility to purchase the mug for the discount price. In a follow-up study, Shani et al. (2015) mentioned the participants’ persistence to accept that they cannot return the product and get the discount later, even though this was explicitly mentioned in the instructions. A method to circumvent this persistence could be to make the scenario more impossible also for the future miss, for example with a geographic restriction. The mug could be a unique souvenir from a vacation destination and the discount will be offered after the participants are flying back home. This would make it easier for the participant to accept the impossibility to fix this missed opportunity.

Our results show that, when evaluated separately, future misses caused less regret than past misses. Conversely, future misses made participants change their feelings of regret more intensely than past misses did. Moreover, regret levels did not decrease when future misses were deferred. In sum, our findings support the strength of future misses on regret but also show contrasting effects when the options were evaluated separately. Thus, we did not expect a reversal of preferences or feelings of regret in a separate evaluation, yet the results stand in contrast to the results of the joint evaluation. Overall, we seek to shed some light into the comprehension of the psychological mechanisms of past and future regret in decisions that involve different time frames. Which situation elicits more regret still remains a complex issue that needs further examination. This study attempted to move one step closer to the comprehension of these phenomena.