One of the roles in my career as a data analyst slash statistician slash decision scientist has been to help people make better decisions. As a result, over the past five years I’ve dabbled a bit in neuroscience and psychology (and by “dabbled” I don’t mean performed brain surgery on the weekends , but rather studied and read about the subjects) and have discovered the fascinating science of decision making. Specifically, I’ve encountered and learned about Cognitive Biases and Decision Making Heuristics. For a quick introduction, there is a good book that was released recently called You Are Not So Smart that offers a good introduction but for the best discussion, pick up a copy of Thinking, Fast and Slow by Daniel Kahneman the man who helped, um, write the book on the subject, and by that I mean he won a Nobel Prize for his work with Amos Tversky on the study of decision making.
There are upwards of 50 different cognitive biases that we all fall prey to – in politics, household finance, work, you name it – but “The Big Five” as I like to call them are especially common and damaging to our ability to make sound and rational decisions. Specifically, there is:
· The Confirmation Bias – where we actively seek out information that supports our opinions and ignore and discount information that contradicts it.
· The Availability Heuristic – where we tend to weigh evidence and discount statistical evidence based upon its recency or how emotionally impactful an event was.
· The Hindsight Bias – also known as the 20/20 or the Monday Morning Quarterback effect where we assume that given evidence we have now that we would have acted differently in the past, even if we didn’t have that present day evidence at the time.
· The “Post hoc ergo propter hoc” fallacy – better stated as the Correlation does not imply causation fallacy says that just because two events occur doesn’t mean that one necessarily caused the other.
· The Clustering Illusion – also known as the Texas Sharpshooter Fallacy is the mistaken conclusion drawn from patterns in random data. In short, we mistake coincidence for patterns and ignore all of the evidence. In order to keep my “favorite” list, I’ll also argue that variants of the Gambler’s Fallacy (I’m due) and The Silent Evidence Bias (focusing on the signal and ignoring the noise or analyzing winners and ignoring data from losers) could be rolled up into this group.
For obvious reasons surrounding recent events, I want to talk a bit more about The Hindsight Bias. I’ll hopefully discuss other biases when I get the chance to select a recent event from current events to better illustrate, and I’m sure I’ll have a nice big sample space to choose from given that it is Primary season for selecting our next President.
The Hindsight Bias basically occurs when you have information to make a decision today, and assume (mistakenly I obviously need to add) that you would have made the same decision in the past, even though you didn’t have the information at the time. I can’t believe I bought stock in Pets.com, it was obviously a dog of a stock. But at the time, all the analysts pegged it as a can-miss stock. You can’t predict when a CEO will commit fraud and tank a stock so you can’t really blame yourself or feel stupid if you owned stock in Enron. Obviously the Quarterback should have run for that critical first down instead of tried to thrown it – the cornerbacks had the play covered – even though at the time the QB was 9 for 9 on passing attempts on the current drive and apparently tearing up the opponent’s defense. Yes, OBVIOUSLY – if you had known then what you know now – you would have made a different decision, but guess what – you DIDN’T know when what you know now, so how would your decision have been any different?
The obvious impetus for writing this essay on the Heuristic Bias has been the recent scandal at Penn State. The example is the argument that Joe Paterno could have alerted the authorities by calling the PA ChildLine service an anonymous tip line that is used for reporting incidents of child abuse. It’s a great service, that allows people who are afraid of reprisal or are worried that an incident may not be properly reported, and while it seems obvious that someone (in this case, Paterno) should have called that number, there are two big problems with this assumption. The first assumption, of course, is that Paterno was aware of the service. Sure, we all know about it NOW because we googled it, it was passed along via social media networks and everybody is now aware of it. But how many of us were aware of the service beforehand? Again, most of us know about this service now AFTER THE FACT, and we only discovered because it has been widely distributed via the internet. I wonder if Joe Paterno, an 80-year old man who by his own admission doesn’t use e-mail, doesn’t use the Internet and would likely not have a clue what Facebook is, would be aware of this service. Oh, sure, he know NOW after everybody told him – but chances are that if the majority of people don’t know about this resource, then someone isn’t doing a good enough job making everyone aware of it. Now there is a second assumption here which is even more important, and which clearly illustrates the Hindsight Bias. And I’ll just answer the question about why Paterno didn’t call the number with a question of my own – In what year did the state of Pennsylvania start up this service and put the phone number into operation? Because while the number exists in the year 2011, if it was only set up recently, if it only went into operation sometime after the initial incident was reported, then Paterno would most likely have reached a Penn State Sub Shop if he dialed the phone number that wasn’t in existence. I’m not trying to diminish the seriousness of the incident that occurred, but it simply isn’t logical to assume that because someone so easily googled and found the phone number of a service that existed in 2011, that a man who was not internet savvy could have simply done the exact same thing back in 2002.
So what causes the Hindsight Bias? I’m tempted to argue that it could be due to a lack of empathy. Not a lack of sympathy or the ability to feel emotions towards someone but rather the inability of others to “put themselves in someone else’s shoes.” You might hear some people claim that race relations have gotten so much better so why do African Americans still complain about racism? Well, it’s very difficult for someone to understand racism and it subtle effects until you’ve actually been the victim of it yourself. I think this inability to consider the backgrounds and experiences of other people is a big problem in this country, and it often leads to the Fundamental Attribution Error where people tend to blame other people’s problems on a character flaw, while they blame their own problems on a situation that was out of their control. If I mess up, it was bad luck, but if you messed up, you’re incompetent. While it’s convenient to blame the Hindsight Bias on our little bubbles, I think there is a much better explanation and it has to do with how we process and store information. We are very linear thinkers, and once we’ve learned a piece of information, it tends to stick with us. There is a quote (which I’m going to attribute to Oliver Wendell Holmes but don’t quote me on it) that says “The mind, once expanded to the dimensions of larger ideas, never returns to its original size.” It’s usually used to refer to how we grow when we learn new things, but unfortunately we tend to FORGET how small we used to be before we figured things out. Can you remember what it was like before you knew how to read? Can you remember learning how to drive? Can you remember what life was like before kids? Can you imagine yourself as the person that you were before you were the person you are now? Oh, sure we all like to say how young and foolish and dumb we used to be. And like I said earlier “If I knew then what I knew now” is a common phrase, but you DIDN’T know then what you know now, so you WOULD have done exactly the same thing! Don’t try to pretend that your flux capacitor is in working order and you’re going to go back in time and tell yourself the secret to life, because if time travel DID exist, someone would already have gone back and screwed things up for you even more.
It’s often helpful to have a journal, not only because it’s cathartic to write things down in the present, but also because it is illuminating to remember what you were thinking in the past. Often you will find that your present memories do not agree with your past musings. What the heck are you supposed to do to resolve THAT kind of cognitive dissonance? Believe it or not, most people will, when confronted with their past memories in written form, be MORE LIKELY to trust their current memories than their past written record. You are biased in your memories by your current experiences and you view and reconstruct your past to fit your current understandings. I don’t mean figuratively, I mean LITERALLY – you actually reconstruct the neurons in your brain and create NEW memories of your past to replace old memories of your past. To use an old VHS metaphor (if you can recall a day before DVDs) you literally go back and “tape over” one episode of your favorite show with another episode. It’s like Lucas re-edited Star Wars to make Han shoot first, Spielberg digitally replaced guns with phones in E.T. and you went back and altered your memories to make you look like less of a doofus than you were. Unfortunately, having become aware of the Hindsight Bias through my studies, although I’m still susceptible to it, I’m less likely to fall victim, so my memories of being a young doofus are still intact.
Even if you are AWARE of the Hindsight Bias, even if you have a photographic memory and even if you’re highly empathic and can read minds to the point that you can share a person’s memories, you’re STILL likely to make these mistakes. Call it wishful thinking (he should have thrown the slider or swung at that pitch) or selective memory (I’ve always been this cool) it is a problem that will cloud your thinking, and it is a problem that will cause lots of conflict in your interpersonal working. One reason why if you’re coming up for a performance review with your boss why you should have been documenting everything you did over the past year. What was said a year ago has likely morphed into something completely different and the two of you are likely going to be on two different pages. Like the old game of “telephone” the more times the message gets past down the lane, the more it gets distorted until the original message looks nothing like the first. And when confronted with the original, you’ll say “that’s NOT what the message said.”
Just remember…the person that you were 20 years ago no longer exists - so be careful what you say about them because they’re not able to speak in their defense. Have a little respect for the dead – after all, in a way it was your funeral.
No comments:
Post a Comment