Sandy beaches often have a line of debris left by the last high tide. These lines look different on different beaches, reflections of the local environment. They might contain rocks, shells, seaweed, jellyfish, flotsam or garbage, egg cases from skate or conch, and so on.
When I was about 11, I spent a week at a beach on which amber occasionally washed up. After a little training, I got pretty good at seeing the amber. On a different beach, the prize was fossilized shark’s teeth, and again I got good at spotting them amid the pebbles.
Given a little training and motivation, the mind pulls out interesting things from the background chaos. What is the wheat and what is the chaff changes based on your needs.
Suppose you’re an emergency room nurse and comment on what a crazy night it’s been and a coworker says, “That’s always the way it is with a full moon.” Now that your mind has been primed, you may notice this coincidence often. But seeing this as more than just a coincidence without good evidence is confirmation bias. Confirmation bias becomes a problem when you sift through the evidence that you come across and select only those bits that confirm what you already believe. You don’t seek information but confirmation.
The hypothesis “God answers prayers” can also be supported by confirmation bias—those prayers that more or less come true within some broad time range are counted as successes, and those that don’t are either ignored or repositioned with, “Sometimes, God says no.” Psychics and horoscope watchers will similarly list successful predictions and ignore or forget the failures.
I listened to the weekly Reasons to Believe podcast from Creationist Hugh Ross for a while. It was little more than a selection of the few bits of evidence from the thousands of scientific articles that week that could be interpreted to support his old-earth Creationist views. Seeing this for what it is—an answer to the question, “What in this week’s news would support my Creationist preconceptions?”—would be fine. It’s when we imagine that this is objective science that we delude ourselves.
So that we evolution-accepting atheists don’t get too smug, Sam Harris proposed the Fireplace Delusion, a chance to have our own preconceptions challenged. It’s a good exercise by which to see your mind being offended and the defenses it puts up to maintain its initial position.
The mind is built to favor evidence that confirms an existing opinion over disconfirming evidence, and to combat this bias, science tries to disconfirm theories rather than confirm them. You can’t prove a scientific theory right, but you can prove it wrong. This reversal—testing our opinions with disconfirming challenges rather than selecting confirming evidence—is a good example to follow.
We can prime our mind, like we’re looking for shark’s teeth on the beach, to pull in only what we want to see, but we delude ourselves when we do so.
Photo credit: Wikimedia
Related posts:

• See all the definitions in the Cross Examined Glossary.

Related links:

• “Confirmation Bias,” You Are Not So Smart blog, 6/23/10.
• “Confirmation Bias,” The Skeptic’s Dictionary.
• “The Fireplace Delusion,” Sam Harris blog, 2/2/12.
• Glenn Morton, “Morton’s Demon,” Talk Origins, February 2002.