Monday, February 21, 2011

Logical Fallacies 18: Confirmation Bias

Video here:
http://www.youtube.com/watch?v=Ddk9MoFAah8&feature=watch_response

Personally, I agree completely with the assessment of Carl Sagan that this one is best summed up in the words of Francis Bacon.
            “The general root of superstition is that men observe when things hit, and not when they miss, and commit to memory the one and pass over the other.”
            In other words, when one has a set of preconceptions, one is predisposed to pay more attention and have more recall of times when events occur in line with them and to pay less attention to and have less recall of events that do not.  A white supremacist is more inclined to notice acts of violence committed by members of other ethnic groups than those committed by Caucasians.  A chauvinist is more inclined to catch anecdotal accounts of women doing and saying unintelligent things than of men doing so.  This fallacy is called “Confirmation Bias.”
            Consider a fellow standing at a roulette wheel.  If he tries to bet that a 47 will be rolled at some point, what are the odds that the casino will allow him place the wager?  Not high.
            My mother really likes piano.  For several years, she tried to get me to like it as well.  Instinctively and habitually compelled as I was to abide her wishes, when she wished me to continue applying myself to piano, I abided, despite the total absence of any personal interest whatsoever.
            A few months ago, I tried to discuss this with her.  The discussion reached a dead end when she explained to me that she had been certain that, if I just gave it a fair try, I would love it.
            So in other words, she was on the lookout for evidence that she wouldn’t have to push anymore, but not on the lookout for evidence that it might be time to stop pushing anyway.  She was watching for reason to believe that I was finally beginning to love piano, but not watching for reason to believe that I probably never would.  She was on the lookout for evidence that she was correct; not evidence that she was mistaken.  Therefore, any such evidence was overlooked.  Apparently spending years doing something doesn’t necessarily count as giving it a fair try.  And what was the evidence she was looking for that I had given it a fair try?  What manifestation would have tipped her off here?  Well of course, there was only one such manifestation: me suddenly loving piano.  She assumed that, if I gave it a fair try, I would love it, and her evidence that I had yet to give it a fair try was the fact that I didn’t love it yet.  So in other words she was going to keep pushing it and pushing it until I loved it.
            My grandmother used to write to me once in a while.  One sentence found in her letters over and over again was, “I suspect that tough times are coming.”  Of course, if a month went by and tough times didn’t come, she would simply say that they weren’t here yet, but they would be one day, and if a month went by and tough times did come, her attitude would be, “I knew it.”
            Of course tough times are coming.  Tough times are always coming.  So are easy times.  That’s the way tough times and easy times are.  They come.  They go.  They come again.  They go again.  Such is life.
            This innately human tendency is why science cannot rely on anecdotal evidence to establish correlations.  To accomplish this end, it must insist on empirical evidence.  That is, it must demand consideration of both the times when two variables change in unison, and the times when they don’t.

No comments:

Post a Comment