We’ve all been its victims. We ‘knew it all along’; we look back on events and can clearly see that something was obviously going to happen. Or at least, you ‘knew it all along’ now that it’s happened. This is the weird world of ‘hindsight bias’: when we convince ourselves that an event—maybe a football score or an election result—was more predictable than it really was after you know the outcome.
On the surface it may seem harmless enough, though annoying if you’re on the end of someone else’s ‘I could’ve told you that’, but it can be a real problem. It can stop us learning from our mistakes and make us overconfident. And can we do anything about it?
Psychologists studying hindsight bias have come to the conclusion that there are three different levels that combine to give the overall effect; each of these levels stacks up: ‘Forseeability’ on top, then ‘inevitability’ and finally ‘memory distortion’, each influencing the one below to give the overall bias.
Imagine a football game. Before the game, you may estimate a 60% chance your team would win. They did.
Later, if asked what you would have said, you remember thinking much closer to 80%. The difference between these two is the hindsight bias. You trick yourself to be more confident of your earlier, misremembered, estimate: “I said it would happen.” This misremembering of earlier judgments or predictions is called ‘memory distortion’.
The second level ‘inevitability’ is when we look back on events and it’s easy to identify the cause and predict the outcome. For example, taking a taxi in a new country. The taxi driver takes a long route to get somewhere, costing more. You subsequently find out and make a series of links in a chain of events; you’re a foreigner in a foreign land, you don’t know what the quickest route is, how much it should cost and taxi drivers want to make as much money as possible. You feel ripped off and that “under these circumstances, it was inevitable.”
The final level is ‘forseeability’. This is subtly different to inevitability, and is focused on your own knowledge, beliefs and ability. After hearing an event has happened you believe that you, personally, could have foreseen it: “I could have told you that.”
The main effects of hindsight bias are divided into ‘myopia’ and ‘over confidence’. Myopia is an error in identifying the cause of something, by focusing in the wrong area, or exaggerating the impact of the correct cause. Overconfidence is the result of overlooking other explanations, or causes, and making future bad decisions, having not learned from experience.
It can have serious consequences for decision making, which is why it has been an area of intense research in many areas including politics, business, law and medicine. For example it could lead to a doctor relying on their first instinct—because they were right last time—rather than investigating other possible causes, leading to incorrect, or late, diagnosis for whatever is ailing you.
Alternatively, you might employ someone based on a good CV and reference, but they turn out to be a bit useless, and after a year you have to ‘let them go’. Senior management, or HR, may look into this—using the currently available information—and see ‘warning signs’ (made more clear by hindsight bias) that you should have noticed before employing them. This could result in you getting a bad review at your next appraisal, on the basis of one hire, and someone else’s hindsight bias.
Expertise appears to shield some people from hindsight bias, at least in fields where there is frequent performance feedback. The evidence for this is not that great, but it seems that the bias is reduced, though not eliminated when people have greater expertise.
Another, more general, way to counteract hindsight bias is by considering other possible causes or outcomes. Doing this goes someway to forcing people to reconsider what is known and what was known. This reduces peoples tendency to ignore information that doesn’t fit with the story they initially pieced together around some event.
By being aware of hindsight bias and considering the alternative outcomes and causes, we may be able to counteract it to some extent. Though not completely eliminate it. But you could probably have told me that.
Reference: Roese, N.J. & Vohs, K.D., 2012. Hindsight Bias. Perspectives on Psychological Science, 7(5), pp.411–426.