2 Experiments 1A and 1B

2.1 Method

Our data and materials are available online on the Open Science Framework (OSF) here: https://osf.io/vxp56/. These experiments were preregistered and the preregistrations are located here: https://osf.io/urhda/ (Experiment 1A) https://osf.io/8n5aq/ (Experiment 1B).

2.1.1 Participants

Our samples (Experiment 1A, N = 391; Experiment 1B, N = 518) were recruited via Turk Prime (Litman, Robinson & Abberbock, 2016). We excluded further participants (Experiment 1A, n = 212; Experiment 1B, n = 213) who failed at least one of two comprehension questions.Footnote ¹ In all studies, participants were required to be U.S. residents and possess a Mechanical Turk HIT approval rating greater than or equal to 95%. Participants were only able to sign-up for one of the two experiments. All studies in this article were approved by the Office of Research Ethics at the University of Waterloo.

2.1.2 Procedure

Participants read a single first-person vignette and rated their agreement with a proposed course of action; see Figure 1 for the materials. In each vignette, participants were assigned an important goal (developing a cure for a disease in Experiment 1A; building a highway to relieve traffic congestion in Experiment 1B). However, recent information indicated the goal was probably futile (another company had already developed a better drug; a different government agency had already devised a better solution to traffic congestion). Participants decided whether they should continue pursuing the goal even though this would require expending resources.

Figure 1:

Vignettes and action statement for Experiment 1A and 1B. Variations based on sunk cost condition feature black text within square brackets. Variations based on moral condition feature colored text (blue = Immoral, green = Non-Moral) within square brackets.

The vignette varied across participants in a 2 X 2 between-subjects design, manipulating: 1) Whether expending the resources required a moral violation (killing lab monkeys; confiscating and bulldozing citizen’s houses) or did not require this (killing farmed pine trees; bulldozing government-owned land). 2) Whether some costs had already been sunk (many monkeys or pine trees already killed for cure; many houses or government plots already bulldozed) or not.Footnote ²

To respond, participants read a statement asserting they would pursue the goal (e.g., “I would conduct experiments on the 100 monkeys to finish developing my cure.”) and rated their agreement with the action on a 1 (Strongly Agree) to 6 (Strongly Disagree) scale; we reverse coded responses so that higher scores reflected greater agreement. Next, participants responded to two multiple choice comprehension questions and a few demographic questions.

2.2 Results

We examined responses from both experiments using separate cost (sunk, not-sunk) by morality (immoral, non-moral) ANOVAs. Both ANOVAs revealed main effects of both factors. Willingness-to-act was greater when costs were sunk then when they were not (Experiment 1A, F(1, 387)=20.31, p<.001, =.050; Experiment 1B, F(1, 514)=44.59, p<.001, =.080), and for non-moral than immoral actions (Experiment 1A, F(1, 387)=30.02, p<.001, =.050; Experiment 1B, F(1, 514)=14.63, p<.001, =.028). There was also an interaction between these factors in Experiment 1B (highway) (F(1, 514)=5.58, p=.019); as shown in Table 1, the sunk cost effect was larger in the immoral condition than in the non-moral version; this interaction was not found in Experiment 1A (Medical Cure) (F(1, 387)=3.10, p=.079), although the pattern of data was similar (Figure 2).

Table 1:Comparisons of Willingness-to-Act (Experiments 1A and 1B) and Moral Acceptability (Experiment 2) across Sunk Cost conditions (Sunk, Not-Sunk)

Note. Levene’s test was significant (p < .05) for the moral conditions of both experiment 1A and 1B but when tested with a Welch correction applied the results are nearly identical. Student’s t-tests are reported here for ease of reading.

Figure 2:

The distribution of agreement with a moral acceptability statement across the sunk cost conditions of each vignette. The black horizontal bar represents the mean value of each condition. The width of the bars represents the number of participants choosing that value: the wider the portion the greater the number.

In summary, Experiments 1A and 1B demonstrated that sunk costs can impact our willingness to act not only in non-moral contexts, but also contexts where the moral component is clear (e.g., sacrificing animals for scientific testing). Further, we found that sunk costs may have an even larger impact on our decision to act in an immoral context. However, this conclusion is tentative as the financial cost of the resources differed between the immoral and non-moral contexts (see General Discussion). These experiments provide initial evidence that past investments can impact our moral judgment by increasing our willingness to persist in immoral actions.

Next, we asked whether sunk costs could make immoral actions seem less immoral. Our test question in Experiments 1A and 1B did not directly address this because many factors besides morality can affect willingness to proceed with an action. For example, a person might be willing to proceed with an immoral action if they think it is in their self-interest. Hence, to directly focus on morality, we asked about the acceptability of acting.

3 Experiment 2

3.2 Results

We examined responses from both vignettes using separate one-tailed independent-samples t-tests. Both tests revealed sunk cost effects (see Table 1). Moral acceptability was greater when costs were sunk then when there were no sunk costs (see Figure 3). That is, extending the findings from Experiment 1A and 1B, sunk costs impacted participants’ judgments of moral acceptability by increasing the moral acceptability of an immoral action.

Figure 3:

The distribution of willingness to act (WTA) across immoral/non-moral and sunk/not-sunk conditions of Experiments 1A and 1B. The dark bar represents the mean value of each condition. The width of the bars represents the number of participants choosing that value: the wider the portion the greater the number.

4 General Discussion

We found that the sunk cost bias extends to moral judgments. When costs were sunk, participants were more willing to proceed with a futile, immoral action compared to when costs were not sunk. For example, they were more willing to sacrifice monkeys to develop a medical cure when some monkeys had already been sacrificed than when none had been. Moreover, people judged these actions as more acceptable when costs were sunk. Importantly, these effects occurred even though the benefit of the proposed immoral action was eliminated.

Our findings illustrate a novel way that the past can impact moral judgment. Moral research conducted to-date has focused extensively on future consequences (e.g., Baez et al., 2017; Reference Miller and CushmanMiller & Cushman, 2013). Although this makes normative sense as only the future should be relevant to decisions, it is well known that choice is affected by irrelevant factors like past investment (Reference KahnemanKahneman, 2011; Reference Kahneman, Slovic and TverskyKahneman, Slovic & Tversky, 1982; Reference Szaszi, Palinkas, Palfi, Szollosi and AczelSzaszi, Palinkas, Palfi, Szollosi & Aczel, 2018; Reference Tversky and KahnemanTversky & Kahneman, 1974). As such, our findings show that as is true with other (non-moral) judgments, people’s moral judgments are affected by factors that rational agents “should” ignore when making them.

Further, our findings show that a major decision bias (i.e., the sunk cost effect) extends to moral judgment. This finding is broadly consistent with research showing that moral judgments are affected by such biases. This earlier work shows that when making moral judgments, people are sensitive to how options are framed (e.g., Reference Shenhav and GreeneShenhav & Greene, 2010) and prefer acts of omission over commission (e.g., Reference Bostyn and RoetsBostyn & Roets, 2016). For example, people make different moral judgments when the decision is presented in a gain frame than when it is presented in a loss frame, even though these two decisions are logically identical (Reference Kern and ChughKern & Chugh, 2009). Likewise, people judge lying to the police about who is at fault in a car accident (a harmful commission), to be more immoral than not informing the police precisely who is at fault (a harmful omission) (Reference Spranca, Minsk and BaronSpranca, Minsk & Baron, 1991). However, unlike most of these previous demonstrations, our findings directly compare the presence of decision-making biases across moral and non-moral contexts (also see Reference Cushman and YoungCushman & Young, 2011).

In our first experiment, we also found that the sunk cost bias may be stronger in moral decision-making than in other situations. This is surprising. In non-moral cases proceeding with a futile course of action is wasteful. But in our moral version of the scenarios, proceeding is wasteful, harmful to others, and morally wrong. Yet, there was a greater discrepancy between willingness to act in response to sunk costs in the immoral condition. Increasing the reasons to not proceed with the action amplified the sunk cost bias. One potential explanation for this is that people are unwilling to admit their prior investments were in vain (Reference BrocknerBrockner, 1992). People succumb to the sunk cost bias in part because they feel a need to justify their past decisions as correct (Reference KuKu, 2008; also see Staw, 1976). Likewise, moral judgments seem to generate a much greater need to provide reasons to justify past decisions (Reference HaidtHaidt, 2012). Thus, those making decisions in an immoral context might have additional pressures to justify their previous choice that stem from the nature of moral judgment itself.

Another explanation is that the initial investment was of a larger magnitude in the immoral compared to the non-moral condition. In both cases, participants incurred an economic cost, but only in one did participants incur an additional moral cost. People are more likely to succumb to the sunk cost bias when initial investments are large (Reference Arkes and AytonArkes & Ayton, 1999; Reference Arkes and BlumerArkes & Blumer, 1985; Reference Sweis, Abram, Schmidt, Seeland, MacDonald, Thomas and RedishSweis et al., 2018). Perhaps sunk costs exerted a greater effect in the immoral condition because the past investments were greater (i.e., of two kinds: economic and moral, rather than just one: economic). However, as we do not know if the economic resources (e.g., pine trees and lab monkeys) were of comparable value, the discrepancy between moral conditions may entirely stem from the lab monkeys being valued higher and thus larger in investment magnitude. Thus, we are hesitant to draw any strong conclusion from this finding. The difference in sunk cost magnitude could stem from differences in financial costs between the immoral and non-moral contexts.

Our finding that moral violations led to increased willingness to act is reminiscent of the “what the hell” effect, in which people who violate their diet then give up on it and continue to overindulge (Reference Cochran and TesserCochran & Tesser, 1996; Reference Polivy, Herman and DeoPolivy, Herman & Deo, 2010). We see this as similar to persisting in an immoral course of action after costs have been sunk. After engaging in a morally equivocal act, people may feel disinhibited and willing to continue the act even when its immorality becomes clear. Likewise, people may persist in an attempt to maintain the status quo (Reference Kahneman, Knetsch and ThalerKahneman, Knetsch & Thaler, 1991; Reference Samuelson and ZeckhauserSamuelson & Zeckhauser, 1988). These accounts, though, may not explain why sunk costs changed people’s moral perceptions. One possibility is that this resulted from cognitive dissonance between people’s actions and their moral code (Reference AronsonAronson, 1969; Reference FestingerFestinger, 1957; Reference Harmon-Jones and MillsHarmon-Jones & Mills, 1999). For example, sacrificing monkeys to develop a cure may cause dissonance between not wanting to harm but having done so. To resolve this, people might change their moral perceptions, molding their moral code to fit their behavior.

We close by considering a broader implication of this work. The extension of decision biases to moral judgment has been previously construed as supporting domain-general accounts of morality that suggest moral judgment operates similarly to ordinary judgment (Reference GreeneGreene, 2015; Reference Osman and WiegmannOsman & Wiegmann, 2017). This is because if morality is not unique, one could reasonably expect that a factor that affects ordinary judgment would likewise affect moral judgment. Thus, if information irrelevant to the decision at hand (e.g., past investments) influences whether we continue to bulldoze land to build a highway, so too should it influence the same bulldoze decision that requires confiscating the land. This is not conclusive however, and our findings could be interpreted to support domain-specific accounts instead (e.g., Mikhail, 2011). For instance, the sunk cost bias was demonstrably larger in moral judgments. Nevertheless, an interpretation of our results as evidence for a domain-general account of morality must explain how the varying effect of past investment on judgment is a difference in degree but not kind