Motivation and emotion/Book/2019/Sunk cost fallacy motivation

Sunk cost fallacy motivation:
What is the SCF and how does it influence our motivation and decision-making?

Multimedia presentation (3 min)

Overview

Have you ever made a commitment you regret, only to find yourself sticking with it anyway? Why do companies invest in failing projects? Why do gamblers keep gambling? These scenarios are encountered frequently throughout our lives. These occurrences typify the sunk cost fallacy. As you read through, make sure to keep these questions in your mind:

Focus questions What drives people to make further investments following a sunk cost? How can psychological factors help to explain the SCF? What is the evidence to substantiate it? How can it be applied to real scenarios?

What is the sunk cost fallacy?

Figure 1. Some people may endure a university course or career path that they've come to dislike. Consider the reasons why, and how the sunk cost bias can be relevant.

The psychological Sunk Cost Fallacy (SCF) is the process of committing further resources to an endeavour once an initial investment of time, money, or resources has already been contributed (Harkes & Blumer, 1985). Once a cost is incurred it is sunk; it should logically play no role in future outcomes or decisions. The term 'sunk' describes the way that the initial cost is irreversible, and that successive decisions cannot alter it. Individuals are assumed to be rational actors, yet many people will make irrational decisions to justify an investment (Friedman, Pommerenke, Lukose, Milam, & Huberman, 2007). Research on this phenomena dates back to the 1970s (Staw, 1976).

This phenomenon is well-documented, but one area in particular where it is problematic is pathological gambling. This is because many gamblers exhibit 'loss-chasing' in situations where they have lost money and will continue to take risks in order to recuperate their money (Campbell-Meiklejohn, Woolrich, Passingham, & Rogers, 2008). There is an impetus to better understand how sunk costs can manifest, and ways in which their potential for driving irrational behaviour can best be mitigated. The fallacy is also an interdisciplinary issue, having been studied within the fields of economics and sociology.

Examples to consider The sunk cost fallacy can crop up inexplicably in day to day life. Here are two hypothetical examples that many people have experienced. You pay a fee of $300 to attend a short course for six weeks. After the second session you realise that it's tedious, but you continue to attend since you would like to avoid 'wasting' the exorbitant fee. You've been following a popular television series for months, but you notice that recently the overall quality has taken a nosedive. Rather than dropping the show you may continue to watch it as not to waste your previous time investment.

Psychological factors

Optimism bias

Optimism bias is the tendency for individuals to overestimate the likelihood of hypothetical future positive events, and underestimate the potential for negative ones. This bias increases in proportion with uncertainty over future events. Whilst the downsides to this bias seem to outweigh any benefits, it has been argued it evolved as a mechanism to fend off existential fears over health and mortality (Sharot, 2011). At its core, the overestimation of beneficial outcomes may help to explain why individuals needlessly commit further efforts to futile tasks.

Framing effects

Framing effects describe the way in which a choice is framed can impact the decision made by an individual. Typically an outcome is perceived as positive or negative in relation to a neutral reference outcome, where shifts in reference can alter the perceived value difference between available options. For instance, the framing of an investment in terms of gains rather than losses may lead to a greater likelihood of individuals making that investment. Choices tend to fall into two patterns - choosing a non-risk gain over a risky gain with identical expected value, and choosing a risky loss over a risk-less loss (Svenson, Raynard, & Crozier, 1997). A subsequent component of this is the certainty effect, where individuals place more weight on outcomes that are certain relative to ones that are only probable when faced with a choice (Tversky & Kahneman, 1981).

Whilst this is clear factor in the investment of sunk costs, much of the literature has only examined framing in isolated decision-making rather than the compounded sequence of decisions involved in scenarios of escalating commitment.

Escalation of commitment

When individuals face an investment with negative results, the escalation of commitment describes a tendency that people will continue to invest further resources with the assumption that doing so will yield a positive outcome. This highlights the irrationality of tendencies to commit to sunk investments (Staw, 1996). Whilst this only describes one component of the sunk cost phenomena, it was seminal in helping to develop more contemporary frameworks.

Choice-supportive bias

Choice-supportive bias is a tendency for individuals to remember choice-related information to retroactively support personal decisions. Table 1 outlines the main factors involved in this process:

Table 1. Mechanisms Underpinning Choice-Supportive Bias

Factors	Description
Misattribution	Preferential information is falsely attributed to the chosen option. Also known as a source error.
Fact distortion	Characteristics owed to the chosen option are falsely remembered as being more preferential than they are in actuality.
False memories	New attributes are recalled as belonging to the chosen attributes, particularly positive ones.
Selective forgetting	Negative attributes of the selected option are forgotten entirely and positive ones are more distinct in recall.

(Lind, Visentini, Mantyla, & Del Missier, 2017).

Loss aversion

Brafman and Brafman (2008) describe loss aversion as an irrational pattern of behaviour displayed when individuals will make unnecessary investments of time or resources in an effort to simply avoid a perceived poor outcome, even when the mounting costs begin to outweigh the potential losses from the alternative strategy. Arguably, this is similar to the escalation of commitment phenomenon, although here the approach is measured by avoiding potential risks rather than approaching a potential gain. The gain versus loss is an important dichotomy in the sunk cost fallacy.

Cognitive dissonance

A seminal theory in psychology, cognitive dissonance describes the sense of unease that arises when an individual's actions contradict their thoughts and beliefs. Internally, people aim toward a psychological constancy that is threatened when their external actions are contradictory to their internal ideals (Festinger, 1957). People are driven to quickly ease the resultant discomfort. A common method is self-justification, a process where individuals will rationalise their behaviour in order to maintain their self-perception that they are competent and morally adequate (Steele, 1988). There is evidence suggesting that cognitive dissonance has a moderating effect on willingness to continue with a sunk cost investment (Chung & Cheng, 2018).

Combining the factors

Consider the biases that have been discussed thus far, and how they play in to the sunk cost fallacy (see Figure 2).

Figure 2. The psychological process of the sunk cost fallacy.

The initial decision to commit to a sunk cost is informed by the tendency to escalate commitment, and avoid loss. For the former, individuals are attempting to mitigate a poor outcome by making further investment, whereas the latter is avoiding the outcome altogether. The initiation of investment may also stem from cognitive dissonance, where the resulting discomfort from the original outcome may drive recuperation efforts. The initial outcome also acts as a frame of reference to which subsequent decisions are held against; framing effects can explain why efforts are made to make the original decision 'worth it'.

After committing to a sunk cost and making efforts to continue investment, optimism bias may exacerbate the issue as individuals tend to overestimate the success of their continued efforts.

Choice-supportive bias becomes relevant in retrospect - despite the continued efforts arguably being worthless, selective errors in memory can help people to rationalise their otherwise irrational behaviour.

Returning to the examples Short course example: In the first example, the investment is mainly monetary - the $300 fee for the course. Continuing to attend represents the continued investment here, as the $300 is already spent regardless. Cognitive dissonance may start to kick in; dropping out early would be a waste of your money, and you are not a wasteful person. Continuing to attend is the effort made to keep your self-concept as a reasonable person intact. Optimism bias may also lead you to believe that eventually it will be worth it, that the hard work will eventually pay dividends. Framing effects dictate that the actions here are driven by framing the continued attendance as a gain (as in it will pay off) rather than a loss (of time and effort). Television series example: In this scenario the investment is not monetary, the resource being expended is primarily your time. Again, cognitive dissonance is relevant. You consider yourself a person with good taste, but the sharp drop in quality of the TV series contradicts this. You might instead convince yourself that it is still good to restore your internal consistency. Moreover, choice-supportive bias comes in to play. In an effort to rationalise your continued enjoyment of the series, your perceptions may become warped, and you may selectively remember the better moments of the show and disregard the inferior ones. Like the first example, optimism bias may also lead you to believe that eventually the show's conclusion will be good, and enduring it in its current form will be worth it in retrospect.

Evidence

Physiological mechanisms

Figure 3. Location of the ventromedial prefrontal cortex, highlighted in red.

There is also physiological evidence supporting the phenomenon. The ventromedial prefrontal cortex (shown in Figure 3), which is involved in gauging gains and losses in decision-based contexts, undergoes reduced activation in decisions where a prior investment has been made. This implies that it already fulfills its role with the original investment and thus contributes less in subsequent decisions (Haller & Schwabe, 2014). Larger sunk costs induce stronger activity in the lateral frontal and parietal cortices, which are brain areas charged with mediating risk. Smaller incremental cost induces activity in the striatum and medial prefrontal cortices, areas sensitive to reward. This finding favors the certainty effect as an influential factor in driving sunk cost behaviour (Zeng, Zhang, Chen, Yu, & Gong, 2013). Cognitive ability is also negatively correlated with exhibiting sunk cost bias (Haita-Falah, 2017).

Quantitative research

This evidence affirms that the SCF has underlying biological mechanisms, but more quantitative research has demonstrated its effects as well. Sunk effects have been exhibited regarding committed relationships, especially when money and effort has been invested. Furthermore, a sunk investment of time has been demonstrated where individuals will opt to stay in a relationship that has become unhappy after an extended period of time spent together (Rego, Arantes, & Magalhães, 2016). However, this study only confirms the SCF in hypothetical terms wherein individuals were presented with fictitious scenarios. Replicating this in practice would potentially be difficult.

Quantitative research has demonstrated the SCF in practice in other contexts. A survey conducted by Braverman and Blumentahl-Barby (2012) examining cognitive biases in health practitioners found that clinicians tend to actively avoid falling into the fallacy when assessing ineffective treatment for patients. Much as in the previous study this, too was hypothetical. An interesting observation to note from this survey was that while clinicians often actively over-compensated for sunk costs, one in ten still recommended staying on an ineffective course of treatment. These results put forward the possibility that the SCF can be countered with education and training in some areas. It is perhaps that in some cases specialist knowledge can act as a force of reasoning when weighing up sunk costs.

Case studies

Figure 4. The French and British governments continued to fund the Concorde project despite mounting costs.

The Concorde Effect

See Figure 4. During the 1960s, the British and French governments pooled their collective resources with the goal of developing the world's first supersonic passenger aircraft. Despite mounting technical challenges and underwhelming investment from the private sector, the two governments were committed to the project. Such to the point that costs had soared to six times its original budget (Marston, 2000). Eventually it entered commercial service in 1975, serving a niche market of wealthy travelers between London and New York until they were withdrawn in 2003. As a result of its lengthy development a phrase was coined - The Concorde Effect (Conway, 2005). What can this tell us about the sunk cost fallacy in practice?

Gambling addiction

Figure 5. Given the minuscule chance of success, it can be argued that gambling is an irrational behaviour. The SCF has explanatory potential for targeting addiction.

Gambling addiction is a pervasive issue affecting many individuals. The SCF is relevant here as gambling addiction is often fuelled by the tendency to further invest in sunk costs (Fujina et. al, 2018). This is exhibited in the consistent pattern of 'loss chasing' exhibited by gamblers, who will often continue to invest money in an attempt to break even from previously accrued losses (Campbell-Meiklejohn, Woolrich, Passingham, & Rogers, 2008). Gamblers will also be more likely to make risky decisions after experiencing monetary loss (Brevers, He, Xue, & Bechara, 2017). Loss-chasing is such a consistent feature that it is even a diagnostic requirement for gambling addiction in the DSM-5 (American Psychiatric Association, 2013).

The previous cognitive factors can also be applied here. Loss-chasing itself is a form of escalating commitment. Gamblers arguably tend to display optimism bias in their habits - individuals who have experienced mounting losses may optimistically think that this time I'll win, despite inevitably failing to do so given the small chance for success. Given the irrationality of gambling it is of no surprise that it goes hand in hand with the SCF, as in Figure 5.

Despite the evidence that gambling is fueled by the SCF, there is little in the way of literature applying it to the treatment of addiction. Perhaps a future direction for research is to develop interventions that use the SCF as a foundational model for treatment.

Quiz

How much have you been paying attention? Choose the correct answers and click "Submit":

1 Which one is not a factor in choice-supportive bias?

	Selective forgetting
	Misattribution
	Selective remembering
	Fact distortion

2 What did the survey conduced by Braverman and Blumenthal-Barby (2012) demonstrate?

	Doctors can prescribe ineffective medicine
	Clinicians will compensate for the SCF when assessing ineffective treatment
	2 in 10 practitioners still adhere to the SCF
	Clinicians employ biased judgement

3 What brain area is responsible for gauging wins and losses?

	Ventromedial prefrontal cortex
	lateral frontal and parietal cortices
	striatum
	prefrontal cortices

Conclusion

Evidently, the SCF is a complex and multifaceted phenomenon that drives behaviour. To summarise the issue, it is pertinent to return to the focus questions raised at the beginning.

What drives people to make investments following a sunk cost?

A recurring theme of behaviour informed by the SCF is that it is often irrational. After all, the efforts taken to justify a sunk cost cannot effectively alter the initial decision. Individuals are often driven by this need to self-justify their decisions - to ease the cognitive dissonance incurred by poor investments (Harkes & Blumer, 1985). A person committing to a sunk cost does so because admitting to a poor outcome is a threat to their self concept as a competent decision maker (Steele, 1998).

Another consistent theme is the need to recuperate losses, a pattern seen most often among gamblers. If cognitive dissonance invites people to recuperate their self-image, sunk financial investments lead to people wanting to recuperate their money. In both cases the final outcome is manifest as irrational behaviour.

In both cases the main desire is justification; a need to justify investments in time, money, or effort.

How can cognitive biases help to explain the sunk cost fallacy?

The earliest rationale developed for explaining the SCF was the escalation of commitment. Often the initial sunk cost results in a negative outcome, which is what drives the desire to make further investment. The investments are made both to mitigate the initial cost but also to satisfy the tendency to avoid losses. Framing effects tend to dictate the initial course of action, where the framing of choices inform what direction a person may take. Optimism bias is also displayed in this period, where the persistence is unrealistically believed to yield a positive outcome eventually.Finally, all of this is supported retroactively in the individual's mind by choice-supportive bias, where they will continue to maintain their sense of competency by distorting the consequences of their decisions. Much like the persistence justifies the sunk cost in practice, choice-supportive bias justifies it only in perception.

What is the evidence to substantiate it?

Evidence substantiating the SCF has been well documented. This is primarily through physiological study, where examination of brain regions involved in decision making and assessing risk provide a biological corollary for the processes underlying the phenomenon. Whilst no single brain region has dominant explanatory power, areas within the prefrontal cortex are largely responsible. Tangible existence of the SCF has also been demonstrated in other quantitative research, although these examples still only illustrate decision-making processes in concept rather than in practice.

How can it be applied to real scenarios?

Already, the coining of the term 'Concorde Effect' is an example of the ways in which the SCF can be applied to practical scenarios. Individuals encounter it regularly in mundane life, from committing to unhappy relationships or enduring a television series that is quickly deteriorated in quality. The explanatory power of the SCF regarding problematic gambling habits also underpins the potential for it to be used as a framework for future interventions and treatment methods, although at the time of writing this potential has yet to materialise in the form of published research.

In Sum... The sunk cost fallacy can be described as a process of an individual's mind justifying it's own irrational thought processes. Perhaps this chapter can be a useful informational tool so that you might better avoid encountering unfavourable instances of the SCF in the future.

See also

• Choice-supportive bias (Wikipedia)
• Cognitive dissonance (Book chapter, 2011)
• Escalation of commitment (Wikipedia)
• Framing effects (Wikipedia)
• Gambling addiction motivation (Book chapter, 2018)
• Loss aversion (Book chapter, 2018)
• Optimism bias (Wikipedia)
• Self justification (Wikipedia)
• Ventromedial prefrontal cortex (Wikipedia)

References

American Psychiatric Association. (2013). Diagnostic and statistical manual of mental disorders (5th ed.). Arlington, VA: American Psychiatric Publishing.

Arkes, H. R., & Blumer, C. (1985). The psychology of sunk cost. Organizational and Human Decision Processes, 1. 124-140. https://doi.org/10.1016/0749-5978(85)90049-4

Brafman, O., & Brafman R. (2008). Sway: The irrestible pull of irrational behaviour (1 st ed.). New York: Doubleday.

Braverman, J. A., & Blumenthal-Barby, J. S. (2012). Assessment of the sunk-cost effect in clinical decision making. Social Science & Medicine, 75. 186-192. https://doi.org/10.1016/j.socscimed.2012.03.006

Brevers, D., He, Q., Xue, G., & Bechara, A (2017). Neural correlates of the impact of prior outcomes on subsequent monetary decision-making in frequent poker players. Biological Psychology, 124. 30-38. https://doi.org/10.1016/j.biopsycho.2017.01.009

Campbell-Meiklejohn, D., Woolrich, M., Passingham, R., & Rogers, R. (2008). Knowing when to stop: The brain mechanisms of chasing Losses. Biological Psychiatry, 63. 293–300. https://doi.org/10.1016/j.biopsych.2007.05.014

Conway, E. M. (2008). High speed dreams: NASA and the technopolitics of supersonic transportation (1 st ed.). Baltimore: John Hopkins University Press

Festinger, L. (1957). A theory of cognitive dissonance. Palo Alto, CA: Stanford University Press.

Friedman, D., Pommerenke, K., Lukose, R., Milam, G., & Huberman, B. A. (2007). Searching for the sunk cost fallacy. Experimental Economics, 10. 79-104. http://dx.doi.org/10.1007/s10683-006-9134-0

Fujino, J., Kawada, R., Tsurumi, K., Takeuchi, H., Murao, T., Takemura, A., … Takahashi, H. (2018). An fMRI study of decision-making under sunk costs in gambling disorder. European Neuropsychopharmacology, 28. 1371–1381. https://doi.org/10.1016/j.euroneuro.2018.09.006

Haita-Falah, C. (2017). Sunk-cost fallacy and cognitive ability in individual decision-making. Journal of Economic Psychology, 58. 44-59. https://doi.org/10.1016/j.joep.2016.12.001

Haller, A., & Schwabe, L. (2014), Sunk costs in the human brain. NeuroImage, 97. 127-133. https://doi.org/10.1016/j.neuroimage.2014.04.036

Lind, M., Visentini, M., Mantyla, T., & Del Missier, F. (2017). Choice-supportive misremembering: A new taxonomy and review. Frontiers in Psychology, 8. https://doi.org/10.3389/fpsyg.2017.02062

Marston, P. (2000, August 12). Is this the end of the Concorde dream? The Telegraph. Retrieved from https://www.telegraph.co.uk/news/uknews/1352927/Is-this-the-end-of-the-Concorde-dream.html

Olivola, C. Y. (2018). The interpersonal sunk-cost effect. Psychological Science, 29. 1072-1083. https://doi.org/10.1177%2F0956797617752641

Rego, S., Arantes, J., & Magalhaes, P. (2018). Is there a sunk cost effect in committed relationships? Curent Psychology: Research and Reviews, 37. 10-519. https://doi.org/10.1007/s12144-016-9529-9

Sharot, T. (2011). The optimism bias. Current Biology, 21. 941–945. https://doi.org/10.1016/j.cub.2011.10.030

Svenson, O., Crozier, W., & Ranyard, R. (1997). Decision making : cognitive models and explanations. London: Routledge.

Staw, B. M. (1976). Knee-deep in the big muddy: A study of escalating commitment to a chosen course of action. Organizational Behaviour and Human Performance, 16. 27-44. https://doi.org/10.1016/0030-5073(76)90005-2

Staw, B. (1996). The escalation of commitment: An update and appraisal. In Z. Shapira (Ed.), Organizational Decision Making (Cambridge Series on Judgment and Decision Making, pp. 191-215). Cambridge: Cambridge University Press. https://doi.org/10.1017/CBO9780511584169.011

Steele, C. M. (1988). The psychology of self-affirmation: Sustaining the integrity of the self. Advances in Experimental Social Psychology, 21. 261-302. https://doi.org/10.1016/S0065-2601(08)60229-4

Tverskey & Kahneman (1981). The framing of decisions and the psychology of choice. Science, 211. 453-458. https://doi.org/10.1126/science.7455683

Tversky, A., & Kahneman, D. (1986). Rational Choice and the Framing of Decisions. The Journal of Business, 59. 251-278. https://doi.org/10.1007/978-3-642-74919-3_4

Zeng, J., Zhang, Q., Chen, C., Yu, R., & Gong, Q. (2013). An fMRI study on sunk cost effect. Brain Research, 1519. 63-70. https://doi.org/10.1016/j.brainres.2013.05.001

External links

• Post-purchase rationalisation (Medium)
• The sunk cost fallacy (Logically Fallacious)