From my Mother Jones piece:
If you wanted to show how and why fact is ditched in favor of motivated reasoning, you could find no better test case than climate change. After all, it's an issue where you have highly technical information on one hand and very strong beliefs on the other. And sure enough, one key predictor of whether you accept the science of global warming is whether you're a Republican or a Democrat. The two groups have been growing more divided in their views about the topic, even as the science becomes more unequivocal. So perhaps it should come as no surprise that more education doesn't budge Republican views. On the contrary: In a 2008 Pew survey, for instance, only 19 percent of college-educated Republicans agreed that the planet is warming due to human actions, versus 31 percent of non-college educated Republicans. In other words, a higher education correlated with an increased likelihood of denying the science on the issue. Meanwhile, among Democrats and independents, more education correlated with greater acceptance of the science. Other studies have shown a similar effect: Republicans who think they understand the global warming issue best are least concerned about it; and among Republicans and those with higher levels of distrust of science in general, learning more about the issue doesn't increase one's concern about it. What's going on here? Well, according to Charles Taber and Milton Lodge of Stony Brook, one insidious aspect of motivated reasoning is that political sophisticates are prone to be more biased than those who know less about the issues. "People who have a dislike of some policy—for example, abortion—if they're unsophisticated they can just reject it out of hand," says Lodge. "But if they're sophisticated, they can go one step further and start coming up with counterarguments." These individuals are just as emotionally driven and biased as the rest of us, but they're able to generate more and better reasons to explain why they're right—and so their minds become harder to change. That may be why the selectively quoted emails of Climategate were so quickly and easily seized upon by partisans as evidence of scandal. Cherry-picking is precisely the sort of behavior you would expect motivated reasoners to engage in to bolster their views—and whatever you may think about Climategate, the emails were a rich trove of new information upon which to impose one's ideology.
This is probably the central reason why I find motivated reasoning theory so powerful (and disturbing). Intelligence doesn't protect us from it; intelligence may enable it! When I started out writing about the politics of science, more than half a decade ago, I assumed the people who denied the facts were ignorant, or even manipulative and devious. Each time I met them in person, though, or spoke with them on the phone, I was forced to reevaluate--they had very sophisticated arguments, which they believed very earnestly. They just happened to be wrong. Motivated reasoning explains this phenomenon nicely--and it helps us understand our ideological opponents without resorting to pointless demonization.