Confirmation bias is when you search for confirmation of something you believe. Finding it then bolsters your belief. In The Black Swan, author Nassim Nicholas Taleb recounts a psychological experiment in which subjects were given the number seqence 2, 4, 6, and asked to guess the rule generating them by producing other three-number sequences that followed the same rule. The experimenter would answer “yes” or “no” in response to each sequence and from that the subjects would formulate their rule.
In this case the rule was “numbers in ascending order,” a simple rule which few of the subjects discovered. To do so, they would have had to offer a number series in descending order (to which the experimenter would have said “no”). Being focused on trying to confirm whatever rule they had come up with, the subjects never thought to try to disprove it and thus never asked the right questions…
This practice of seeking evidence that disproves one’s theory is called skeptical empiricism, and is one Taleb advocates as a means of increasing one’s objectivity in perceiving reality. However, it is so much against our nature that it requires a fair degree of concentrative energy. Our habit, our nature is to go for confirmation rather than falsification. Given man’s fallen state I can readily attribute this to the pride of the flesh, delighting in the cleverness of its own ideas and not at all pleased at the idea of being wrong
“The way of a fool is right in his own eyes…” Pro 12:15
“He who corrects a scoffer gets insult for himself.” Pro 9:7
“Do you see a man wise in his own eyes? There is more hope for a fool than for him.” Pro 26:12
Taleb says you can find confirmation for just about anything you want to believe. Confirmation in circumstances, confirmation in “references,” confirmation in events. Confirmation from other people.
The day before I read this section of the book, I was talking with someone who was advocating a health product whose method of operation and results I found difficult to believe. When I expressed my skepticism the person offered several incidents of the personal testimony sort as “proof” the product was legitimate and worked as advertised. As soon as I read about confirmation bias, I realized I’d just seen it in action.
In Frank Peretti’s semi-autobiographical novel, The Visitation, there is an incident where the protagonist was certain that God wanted him to go to the Billy Graham headquarters in Minnesota and offer his services. He was totally unqualified and really had no “services” to offer (except perhaps janitorial), nevertheless he was convinced it was God’s will and direction that he do this. He found confirmation on the side of a boxcar in a train he happened to pass as he started out on his bus trip to Minnesota. There on the side of the car was a number that just happened to be the same as the street number of the Billy Graham offices. Proof, he exulted, that he was indeed following God’s lead.
Unfortunately when he arrived at the headquarters he was turned away without appeal… Which left him confused. Had God not been leading him after all? I’d say no. It was merely confirmation bias at work.
Complicating this tendency to want to confirm one’s theory or belief rather than to disprove it is the tendency to focus on the incidents that do confirm, while blotting out those that do not. Taleb calls this the silent evidence. You hear of the 10 people who were cured of cancer using this innovative technique, not the 1100 who died using it. You hear of the 100 writers who succeeded using such and so marketing technique, not the thousands who did not.
Sometimes, scientists just throw out the experiments that don’t confirm their theories while trying to force the ones with promise to do so… The recent CRU emails give some examples of this, and I distinctly recall an article I read a few years ago by Richard Lewontin, maybe, about exactly this. We are aghast at the practice, yet if we’re honest I think most of us will find we do the same thing, if on a lesser scale perhaps.
I’ll use an example that I’m familiar with. Let’s say I fear that deep down I believe that I’m not really a very good writer (my theory). I can get twenty very positive comments on my writing, from people I know and respect and yet, it’s the one negative comment, often from a total stranger, that I recall most vividly. Why? Because it’s corroborating my “I can’t write” theory. That’s also why the negative comments are the ones that tend to surface when I’m struggling to write the next book, corroborating my resurrected fear that I really can’t write after all. “See? Not only am I having trouble with the work in progress but some reviewer on Amazon confirmed that I really am just an imposter.”
Thankfully God’s growing me out of this ridiculous scenario, and this whole idea of confirmation bias is a very helpful concept in doing so. It also answers questions I’ve had about doctrinal or faith-based differences between believers. But more on that tomorrow.