Note: This post is an initial exploration, to be expanded as the way forward reveals itself.

—

Confirmation bias describes our tendency to notice and give greater credence to evidence that fits with our existing beliefs. According to psychologist Gary Klein, confirmation bias leads us to:

  1. Search only for evidence that confirms our beliefs

  2. Prefer evidence that supports our beliefs

  3. Best remember information in keeping with our beliefs

  4. Interpret evidence in a way that supports our beliefs

  5. Rely on favored beliefs to misunderstand what is happening in a situation

  6. Ignore opportunities to test our beliefs

  7. Explain away data that don’t fit with our beliefs

Now compare the above varieties of confirmation bias with the following tendencies to:

  1. Search for evidence that confirms what we believe may be the case.

  2. Prefer evidence that supports what we believe may be the case.

  3. Best remember information in keeping with what we believe may the case.

  4. Interpret evidence in a way that supports what be believe may be the case.

  5. Rely on what we believe may be the case to understand what is happening in a situation.

  6. Put off opportunities to test what we believe may be the case until we have conducted a preliminary search for evidence that supports our belief.

  7. Find reasons our tentative belief may still be right even though some data don’t support it

Confirmation bias sounds worse in the first list than in the second, because it’s serving the cause of “our beliefs”, a phrase that coveys conviction that something is indeed the case. In contrast, the confirmation tendencies in the second list merely serve tentative beliefs that something “may be the case” - more hunches than firm belief. And if we want to explore whether a tentative belief is correct, the best approach is to follow the hunch and see what happens. This is how scientific discoveries and great inventions often happen - by persevering through the doubt, at least for awhile.

—

12/27/20: In this post, I’m using ‘hunch’ as shorthand for any type of tentative belief, such as conjecture, speculation, hypothesis, or Big Idea. A hunch is like an unfamiliar seedling in one’s garden. To find out if it’s a weed or flower, you have to water the seedling until its true nature is revealed. Likewise, hunches need to be nurtured for a while to see what they will lead to. If we try to disconfirm our hunches too early, we may miss out on learning something new. So if I have a hunch that Siegfried likes me, I’ll try to confirm that by asking him out. Or if I’m a scientist, I won’t give up on my testing my Big Idea if the initial research findings are negative.

So, how does one tell the difference between good and bad confirmation bias? There’s no set answer, because context matters. As a general rule, we shouldn’t dismiss or ignore evidence that disconfirms our beliefs. We should be alert to the possibility that we are wrong. But there are times when it’s more productive and worthwhile to focus more on confirming than disconfirming our beliefs, to act as if they were true or at least to seek evidence that supports them. The trick is to know when to stop giving our cherished beliefs the benefit of the doubt and when to let them go.

References:

The Curious Case of Confirmation Bias:The concept of confirmation bias has passed its sell-by date. Gary Klein, Psychology Today.  May 05, 2019

Why we interpret information favoring our existing beliefs. The Confirmation Bias, explained. The Decision Lab