Politics and Cognitive Science (1)

I liked this post on politics, confirmation bias, and the blogosphere by Hal at Hellblazer. As I guess I've said before, I think that studying the results of cognitive science can help us in our conduct of our inquiries, especially our political inquiries.

I want to nit-pick a bit with Hal's post, though I don't think that this nit-picking undermines the substance of his point. Hal seems to claim that confirmation bias is our tendency to look for and/or prefer evidence that supports our preferred or antecedently-held beliefs. If I'm not mistaken, Hal is actually conflating two different phenomena. Confirmation bias per se is, as I understand it, our tendency to conclude that whatever hypothesis is under consideration has been confirmed. Confirmation bias affects us even when we don't give a tinker's damn about the hypothesis and have no antecedent opinion about whether it is true or false. For example (I'm doing this from memory, so I hope I don't botch the details here), experiments have been done in which people were asked to evaluate the hypothesis practicing the day before a tennis match increases the likelihood that you will win the match, and they were given a set of ambiguous data to use in their evaluation. The subjects tended to conclude that the hypothesis was true. However, another set of subjects was given the same ambiguous data and asked whether it confirmed the hypothesis practicing the day before a tennis match increases the likelihood that you will lose the match. But again, the subjects concluded that the hypothesis was true! This is not an isolated experimental result; that we suffer from confirmation bias is well established.

We also have a tendency to favor hypotheses that we already believe to be true, and to favor hypotheses that we want to be true...but these are different cognitive shortcomings. Remember: we've got a million of 'em...

Confirmation bias often works together with these other cognitive biases, of course...and this (undoubtedly) often happens when we are thinking about politics and other emotionally-charged issues. And when you are fighting multiple cognitive shortcomings at the same time...well, the deck is really stacked against you.

Anyway, as I said, I don't think that these technical points detract significantly from the value of Hal's post.