Definition, Examples, & How to Avoid It​

Hindsight bias is our tendency, after an event has occurred, to overestimate the extent to which we could have we could have predicted it (APA, 2023). Put another way, we believe we knew something was going to happen all along, even if we actually didn’t have any idea beforehand. Our present knowledge influences our memory of what we think we knew earlier, and we find it difficult to ignore (Bernstein et al., 2007).​

​You may also see it referred to as “creeping determinism”. This is more of a type of hindsight bias. After we gain knowledge of an outcome, creeping determinism is that feeling that things had to happen this way. The belief that the outcome was inevitable or pre-determined gradually “creeps” into your mind. (Nestler et al., 2008)

On the surface, it can seem like hindsight bias is harmless. So what if we think you predicted the outcome of a baseball game when you actually didn’t? But it can have some more serious consequences, especially for decision-making. Here are some potential problems:

• Overconfidence – Hindsight bias can cause us to overestimate our prediction abilities. When you think you know more than you know, you’re less likely to seek information to inform your decisions. This can lead to risky choices. More on this later.
• Misunderstandings – It can contribute to misconceptions about the causes of past events. If we think we knew something all along, we overlook the complexities and uncertainties in a situation. This can result in oversimplification of cause and effect. We tend to see a single cause for an outcome and neglect other contributing factors and reasonable explanations (Roese & Vohs, 2012). This can lead to unfair blame for poor outcomes and once again, poor decisions.  
• Failure to learn from mistakes  –  When we believe we know why events turned out as they did, we don’t see a need to examine our reasoning process. This impairs our ability to evaluate the past and learn from it. One study found participants that who used hindsight bias were less likely to learn from mistakes and more likely to repeat them in the future (Roese & Vohs, 2012).
• Memory distortion – Studies show that hindsight bias can change the memory you have of your prediction before you learned the outcome. Having inaccurate memory is connected to other biases such as confirmation bias, belief bias, and anchoring. Also, sometimes memory distortion leads us to believe we got information from a more reliable source than we did. For example, you may think you heard something on the news when actually a friend told you (Calvillo, 2012). These effects are never good for effective decision-making.

There are three main causes of hindsight bias. The first is cognitive, which means it involves mental processes. The second is metacognitive, which is about ease of understanding. The third is motivational. We have a motivation to see the world as orderly and predictable and to see ourselves as intelligent and knowledgable (Roese & Vohs, 2012).​

• Memory distortion – This is when we only remember things that support what we know to be true. Or, we even change what we remember so that event outcomes make more sense to us. This is related to the need to see the world as orderly and predictable.
• Knowledge updating – This is our tendency to merge new information into what we already know. For example, once you know the outcome of a football game, you automatically connect that information to things you already knew about player statistics and coaching performance. It reinforces what you already knew, solidifying the belief that the game had to turn out the way it did. But, we are less likely to see any inconsistent information.
• Sensemaking – This is about our desire to find reasons for outcomes. We tend to look for simple reasons to understand why something happened the way it did. We oversimplify cause and effect in our search to make sense of otherwise random or complicated, interconnected causes of outcomes. It’s a kind of “mental shortcut” to make it easier to process information.

The term metacognition refers to thinking about thoughts themselves. Hindsight bias makes it feel easy to come up with points to support an outcome. This ease of reasoning makes it seem like events were easily foreseeable or certain to happen. Here’s an example. In a study, one group of participants was asked to come up with two reasons for an outcome, and another group was asked to come up with ten reasons. Since it’s much easier to come up with two reasons, this group showed much higher hindsight bias. They felt more certain they would have been able to predict it (Roese & Vohs, 2012).

Two motivations contribute to hindsight bias— predictability and self-esteem. 
​

• Need for predictability – Because of our need to believe the world is safe and predictable, we tend to underestimate the role of chance and randomness in events. It can feel threatening to know that sometimes things happen for no reason, just a result of random chance. Order helps us feel in control, and more secure. 
• Ego/Self-esteem – Simply put, telling yourself that you could have predicted what others didn’t, or even creating a false memory that you did predict an outcome, makes us feel confident and self-assured. (Something we all want more of.) There is also social pressure to appear intelligent and competent. 

Avoiding hindsight bias is hard. One study found that even when study participants were told to ignore an outcome, they were unable to (Henriksen & Kaplan, 2003). But some things can help reduce its effects.

• Consider the opposite – By considering how things could have turned out differently, we become less attached to what did happen and more open to possibilities (Roese & Vohs, 2012).
• Examine the data – When you have more information, you will be able to get a fuller view of the factors that contributed to a decision or outcome. 
• Seek other perspectives – Getting perspectives that are different from your own can help you make more balanced and informed decisions.
• Record your decision process in a journal –  Recording your thoughts and predictions before you know the outcome is a good way to fact-check yourself after you know what happened. This way you can’t unconsciously “change” your memory.
• Focus on the decision-making process rather than the outcome – This helps to identify areas of improvement. Did you gather information and consider alternatives? Why did the decision make sense at the time?

• Success – When we’re successful, we tend to attribute that success to our actions or talents, and think “I knew I would succeed”. But actually, chance or being given advantages could have played a big role.
• Elections – Our memory of election predictions is often affected by hindsight bias. In one study, participants were asked to predict the percentage of votes candidates would receive, and then recall their prediction after the election was over. 88% of the participants remembered their predictions as closer to the outcome than they actually were (Chen et al., 2019). 
• Relationships – When a couple splits up, we may find ourselves thinking “I knew it wouldn’t last.” You may even be able to give the warning signs from the beginning of the relationship that you swear you saw. But before the break-up, you didn’t voice any concerns.
• Sporting events – After attending a game or match, you insist that you knew who was going to win all along, even if the game was close or unpredictable.
• Stock market – After a stock price rise or fall, people often claim they knew it was going to happen, even though they can’t present any concrete information that led to their prediction.

Thousands of studies have been conducted on hindsight bias since the phenomenon was first reported in 1975 by psychologist Baruch Fischoff. Hindsight bias experiments typically fall into one of two types. The first one involves comparing predictions between two groups, one that knows the outcome and one that doesn’t. The other involves asking individuals to predict an outcome, and then recall their prediction after the outcome is known (Roese & Vohs, 2012). Here’s an example of each:

• Rate the outcome – In this experiment, participants were asked to read accounts of historical events or clinical cases. Then they were given several outcomes and asked to rate how likely each outcome was. However, one group was told what the outcome was before they rated, while the other group was not. Participants in the group that knew the outcome consistently rated that outcome as more likely, but the other group didn’t (Fischhoff, 1975). This shows how our estimate of likelihood is swayed by hindsight. ​
• Recall your estimate – A well-known experiment asked college students to rate the chances that the U.S. Senate would confirm Clarence Thomas to the supreme court. Before the confirmation hearing, 58% of the students said he would be confirmed. But after the confirmation, the same students were polled again, and this time 78% said that they had predicted his confirmation (Dietrich & Olson, 1993). This shows how confirmation bias can influence your memory.

Hindsight bias makes us believe we have better intuition than we do. When we believe that we always knew something would turn out the way it did, even when we didn’t, it gives the false impression that we are really good at intuitively reading a situation without examining the data. If your intuition is that good, why bother gathering information and thinking things through?

As described earlier, overconfidence is a common result of hindsight bias. We all know of someone who engages in risky behavior due to overconfidence. We may even see it in ourselves. When it comes to hindsight bias, overconfidence causes us to overestimate our ability to read situations and we don’t give enough weight to chance. This also leads to overlooking other perspectives and even ignoring information that contradicts us (Roese & Vohs, 2012)

​For example, let’s say an investor makes a successful trade that makes them a lot of money. They may attribute this success to their skill and ability to predict the market, and completely ignore the role of chance. This belief creates an increased likelihood of taking higher risks in the future which could end up in huge losses.

Hindsight bias is sometimes referred to as a “decision trap”. A decision trap is when we become stuck in a pattern of decision-making that is ineffective or even harmful. It usually involves some kind of mental shortcut that makes decisions easier when things are uncertain.

​Hindsight bias can be a trap because it gives us a false perception about how good we are at making decisions. This is obviously related to the overconfidence that tends to go along with hindsight bias. When you’re confident you could have predicted an event, or if an outcome seems obvious, you don’t look for new ways to analyze the situation. We oversimplify its complexity and overestimate the quality of our decisions.

Hindsight bias in healthcare was first noted in the 1970s when researchers observed doctors believing they knew all along how patient cases were going to turn out. This view can be especially concerning in a healthcare setting. When doctors are overconfident in their diagnosis and treatment, it can lead to incorrect treatment decisions as well as a failure to learn from mistakes.

​One study found that doctors who were given information about a patient’s diagnosis were more likely to believe they would have made the correct diagnosis, even if they hadn’t considered that diagnosis initially. This overconfidence can lead to missed or delayed diagnoses, which can have serious consequences for the patient (Motavalli & Nestel, 2016).

If you’d like to keep learning more, here are a few books that you might be interested in.

• Anderman, J. (2009). 9/11 on the ‘net: Risk and Uncertainty in Hindsight. Intersect: The Stanford Journal of Science, Technology, and Society, 2(1), 1-20.
• APA. (2023). Apa Dictionary of Psychology. American Psychological Association. Retrieved May 4, 2023, from https://dictionary.apa.org/hindsight-bias 
• Bernstein, D. M., Atance, C., Meltzoff, A. N., & Loftus, G. R. (2007). Hindsight Bias and Developing Theories of Mind. Child development, 78(4), 1374. https://doi.org/10.1111/j.1467-8624.2007.01071.x
• Calvillo, D. P. (2012). Working memory and the memory distortion component of hindsight bias. Memory, 20(8), 891-898.
• Chen, Y. H., Cheng, H. P., Lu, Y. W., Lee, P. H., Northoff, G., & Yen, N. S. (2019). Can knowledge of election results change recall of our predictions? Neural correlates of political hindsight bias. Plos one, 14(10), e0220690.
• Dietrich, D., & Olson, M. (1993). A demonstration of hindsight bias using the Thomas confirmation vote. Psychological reports, 72(2), 377-378.
• Fischhoff, B. (1975). Hindsight-foresight: The effect of outcome knowledge on judgment under uncertainty. PsycEXTRA Dataset. https://doi.org/10.1037/e459202004-001
• Henriksen, K., & Kaplan, H. (2003). Hindsight bias, outcome knowledge and adaptive learning. BMJ Quality & Safety, 12(suppl 2), ii46-ii50.
• Motavalli, A., & Nestel, D. (2016). Complexity in simulation-based education: exploring the role of hindsight bias. Advances in Simulation, 1(1), 1-7.
• Nestler, S., Blank, H., & von Collani, G. (2008). Hindsight bias doesn’t always come easy: causal models, cognitive effort, and creeping determinism. Journal of Experimental Psychology: Learning, Memory, and Cognition, 34(5), 1043.
• Roese, N. J., & Vohs, K. D. (2012). Hindsight bias. Perspectives on Psychological Science, 7(5), 411–426. https://doi.org/10.1177/1745691612454303