Hindsight Bias

Skyler Jones & Celeste Wijnbelt

“I knew that they’d win that game,” you say after the final buzzer has sounded. Despite your confidence in your comment, your initial thoughts beforehand were that there was a 50/50 chance that the team would actually win. So why, after the fact, or in hindsight, were you so confident in your statement? Why was the unforeseen suddenly foreseeable? We often quickly believe that the known information we have gathered was obvious. We knew it would happen all along. But did we?

This “knew it all along” phenomenon is more commonly known as hindsight bias, the tendency for people to believe that they could have accurately predicted the outcome of a past event. Studies reveal three primary forms of hindsight bias: memory distortion, inevitability, and foreseeability (Roese, 2012). In memory distortion there is a focus on the falsely gathered understanding of earlier judgement; in other words, the individual “said it would happen.” Inevitability refers to a belief that the past was predetermined; the individual believes “it had to happen.” Finally, in foreseeability the individual attributes the outcome to their inherently subjective belief of personal ability and knowledge; they “knew it would happen.” Combining all three levels, hindsight bias stretches across a range of situations, making it very common.

Hindsight bias occurs across all age groups and is a robust phenomenon that can have significant consequences related to judgement and decision making in the future based on past assumptions. In adults, hindsight bias has been shown to impact decision making in a wide range of areas ranging from medical and legal decisions to minor judgements of personal experiences. In each case, the decision maker, backed by knowledge of a certain outcome, overestimates the likelihood of said outcome.

Early studies conducted by Baruch Fischhoff (1975) suggest that a strong component of hindsight bias is “creeping determinism,” a term used to describe the tendency for individuals to deem observed outcomes as inevitable (Fischhoff, 1975). The reasoning behind this is the propensity for individuals to flatter themselves, so to speak, by claiming that they knew something was to happen all along, though only spoken after the fact. Overconfidence, therefore, is a significant contributor to hindsight bias, as it can often lead individuals to perceive their judgments as more knowledgeable than they actually are.

Components of Hindsight Bias

Theory of Mind

Some scholars believe that hindsight bias is the result of the brain rendering old information inaccessible after updating with newly acquired information (Fischhoff, 1975), while other accounts maintain that hindsight bias occurs due to a biased restructuring of the beginning situation, where the outcome is the cue (Pohl). Another suggestion, however, is that hindsight bias could stem from another phenomenon: theory of mind (Bernstein et al., 2007). Theory of mind (ToM) is the ability of individuals to attribute mental states and conclusions of situations to others and themselves, serving as a cornerstone of interactions between people. Another critical thing to note about ToM is its sparsity among children. Children below the age of 4 often have difficulty understanding that others experience things different from their own reality (Bernstein, 2007), which is oftentimes revealed by ‘false belief tests.’ Bernstein notes that hindsight bias appears to be similar to trends in ToM studies of children; in both, the participant learns that their belief of a situation was incorrect and must concurrently reapproach and re-evaluate their prior naive belief or that of another ‘naive’ individual. Essentially, adults making biased decisions after the fact (hindsight bias) share similarities to children making biased judgements based on ToM trends. The difference, however, lies in the certainty of the decision being made. Where children hold on to their beliefs of a certain outcome through their lack of theory of mind, adults have a limited commitment to their biased outcomes in decisions with hindsight bias. With that in mind, it becomes clear that children might grow out of decision bias associated with theory of mind (as they develop more ToM with age), whereas hindsight bias is a lifelong cognitive fault, perhaps playing a role in other judgments as well. Additionally, it might be more appropriate to view hindsight bias as a continued yet altered version of theory of mind biases in children, a representation of “false hunches’’ or “false judgments” based on tasks, rather than “false beliefs’’ that children retain (Bernstein, 2007). While there is still much to be learned about the interconnectedness between hindsight bias and other judgement and decision-making processes, it is clear that these types of biases in judgement persist throughout whole lives, not just occurring in one stage. However, the similarities between hindsight bias and theory of mind reveal that the sparsity of theory of mind in children could be an origin of the persistent hindsight bias we see in all age groups.

Cognitive Dissonance

Hindsight bias holds several parallels to cognitive dissonance. Cognitive dissonance is the uncomfortable feeling that is thought to occur when an individual does something that is inconsistent with his or her internal values (Debono, 2019). To resolve the dissonance, one will adjust their beliefs to be consistent with their behavior. Consider this example: a woman purchases an expensive designer sweater. A month after the purchase the woman decides that she doesn’t like the sweater all that much anymore, and decides to return it. However, the store claims that the return window has closed, and the woman can only get half-price back for the sweater. As a result, the woman convinces herself that she does like the sweater afterall and decides to keep it. The dissonance in this situation occurs when the woman chose to keep the sweater even though she didn’t particularly like it. To mitigate the dissonance, she convinced herself that she does, in fact, like the sweater.

This series of decisions is reflected in Alafair S. Burke’s paper Improving Prosecutorial Decision Making: Some Lessons of Cognitive Science, which stated “the desire to avoid cognitive dissonance can cause people to adjust their beliefs to maintain existing self-perceptions” (Burke, 2005). Similar to hindsight bias, cognitive dissonance aims to bring one’s beliefs and actions into balance. As humans, we want to believe that our actions reflect our internal values– and when this is not the case, we do everything in our power to realign them (Burke, 2005). While cognitive dissonance and hindsight bias hold many similarities, a few key differences must also be mentioned. Cognitive dissonance occurs concurrently with the decision making process, and pushes one to change their thoughts and feelings about a particular decision as they make it. However, hindsight bias mainly takes place once the decision has already been made. While cognitive dissonance leads to changes in one’s thoughts and decisions in the present moment, hindsight affects how one reflects on the past. In whole, the comparison of these two phenomena have allowed us to gain a better understanding of our decision making processes, and will continue to be useful as we discover the potential problems that can arise under these biases.

The Problem of Hindsight Bias

The consequences of hindsight bias can hinder our ability to effectively gather information and learn from it. Fischhoff proposes that unrecognized hindsight bias can hinder us from effectively judging the past and learning from it (Fischhoff, 1975), which highlights the potential negative impact that hindsight bias can have on our overall judgment and decision making. It is the outcome knowledge that ‘tricks’ individuals into believing they know what the past was about, when in reality, they learned nothing from it. In the words of Roese and Vohs, “hindsight bias can be confused with simple learning from experience.” The so-called ‘decision trap’ of hindsight bias reveals two primary consequences: a myopic attention to a sole cause of a past event and skewed overconfidence in personal judgements (Roese, 2012). The myopia from hindsight bias involves a failure to accurately pinpoint the cause of a problem. This is the result of either a hyperfocus on a false cause or an exaggeration of the impact of the actual cause (that might not be the actual cause again). In real terms, this can lead to “placing more blame on a particular individual than is warranted,” and not recognizing a lapse in personal judgement of a problem. Overconfidence, however, is related more to an exaggeration of trust in a personal ability to analyze a problem and make an educated decision. As a result, individuals may overlook other potential perspectives, which leads to adopting risky positions in future decision making. Overconfidence from hindsight bias creates a reluctance to adopt a different approach or perspective on an issue, since the individuals who ‘knew it all along,’ feel like there is no reason to consider any other perspectives. Consequently, the individual learns less or–in some cases–nothing.

Ways to debias

Hindsight bias often involves people, who already know the true outcome of an event, believing that the event was relatively inevitable. As a result, one of the most common ways to combat hindsight bias is to argue against the reported outcome and attempt to convince oneself that the event could have had a different end result (Gilovich, Griffin & Kahneman, 2002). When a subject argues against the reported outcome of an event, they are concurrently reminding themselves of their original thinking patterns. This process will naturally combat hindsight bias, as it allows one to rediscover their previous argument. However, this technique is not always effective. If one has a hard time coming up with ideas that oppose the reported outcome, the process can be counterproductive. Across his studies, Nobert Schwarz found that when the ease of accessibility to these alternative options decreases– or if people struggle to accumulate ideas– they take this as meaning there are not many examples to be generated (Sanna & Schwarz, 2003) . This process plays off of Tversky and Khaneman’s availability heuristic. If one has a hard time coming up with ideas that oppose the outcome, the process can backfire and they will be more engulfed in their biased thinking patterns than before and believe to an even greater extent that the outcome was inevitable.

Social Applications

As discussed previously, hindsight bias can be apparent in both professional and academic settings. Whether it be a doctor trying to recall why they made a certain diagnosis, or a historian putting themselves in the shoes of a historical figure, it is essential to be able view a decision with foresight and not be swayed with the actual outcome. By analyzing a decision without the additional bias of an end result, people across every field are able to gain a logical and complete understanding of what truly went through the minds of the decision-makers at the time. De-biasing techniques, such as arguing against the reported outcome and making a decision journal, ensure that hindsight bias does not limit one’s focus and accurate interpretations can still be made.

Conclusions

Hindsight bias is a complicated phenomenon that preys on people’s overconfident judgements of a past event. The psychological motivators of this bias have yet to be fully understood, however, some argue that it stems from the brain failing to uphold previous resolutions once updated with new information; others firmly believe that ToM (theory of mind) plays a major role. No matter what might be causing it, it is clear that hindsight bias has multiple dangers, the biggest one being that it will often prohibit decision makers from seeing the possibility of another path. As a result, hindsight bias often prohibits people from learning anything from their previous experiences and forcing them to make faulty judgments in the future. Thankfully, there are ways to prevent hindsight bias and ensure that your feelings about a decision are not based off of the real-life outcome; these include reminding one’s self why they made a particular choice in the first place, which can be done by creating a decision journal. All in all, hindsight bias can affect people of all ages and life stages, and when given the opportunity will thoroughly skew one’s memory. This bias holds great power, but stands no match for someone who is well aware of its effects and possesses the proper tools to fight it. 

References

Ackerman, R., Bernstein, D. M., & Kumar, R. (2020). “Metacognitive hindsight bias.” Memory & Cognition, 48(5), 731. https://doi.org/10.3758/s13421-020-01012-w 

Ash, I. (2017).“Can Exposure to Post-Outcome Information ‘Debias’ the Hindsight Bias?” Proceedings of the Annual Meeting of the Cognitive Science Society, https://escholarship.org/uc/item/3jr8164k

Bernstein, D. M., Atance, C., Meltzoff, A. N., Loftus, G.R., (2007). “Hindsight bias and developing theories of mind.” Child development vol. 78,4 (2007): 1374-94.  https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3649066/ 

Burke, A. S. (2005). Improving Prosecutorial Decision Making: Some Lessons of Cognitive Science. William and Mary Law Review, 47(5), 1587–1634.

DeBono, K. G. (2019). Cognitive dissonance. Salem Press Encyclopedia of Health.

Fischhoff, B. (1975). “Hindsight Is Not Equal to Foresight: The Effect of Outcome Knowledge on Judgment under Uncertainty.” Journal of Experimental Psychology: Human Perception and Performance, vol. 1, no. 3, Aug. 1975, pp. 288–299. https://psycnet-apa-org.stanford.idm.oclc.org/fulltext/1976-00159-001.pdf 

Gilovich, T., Griffin, D., & Kahneman, D. (Eds.). (2002). Heuristics and Biases: The Psychology of Intuitive Judgment. Cambridge: Cambridge University Press. doi:10.1017/CBO9780511808098 

Freeman, M. P. (2010). Hindsight: the promise and peril of looking backward. Oxford University Press.

Sanna, L. J., & Schwarz, N. (2003). Debiasing the hindsight bias: The role of accessibility experiences and (mis)attributions. Journal of Experimental Social Psychology (Print), 39(3), 287–295.

Van Boekel, M., Varma, K., & Varma, S. (2017). A retrieval-based approach to eliminating hindsight bias. Memory, 25(3), 377–390. https://doi.org/10.1080/09658211.2016.1176202