I don't know about you, but I'm pretty sure Jonah Lehrer knows a lot more than me about psychological research on cognitive bias.

He's a pretty knowledgeable dude, but he has major incentives to overstate research - the incentives which literally just days ago got him a job at The New Yorker, which is a pretty big plum for any journalist. When you read the original papers (I jailbroke the OP paper in question), you find he overstates things.

For example, this paper: the biases were specifically chosen to be resistant to intelligence, and the intelligent performed as well as the stupid - they were just wrong in predicting they'd perform better than the stupid (as they do in every other area of life). This is interesting if you're trying to formulate an integrated theory of IQ and the two processes, as Stanovich is doing (Stanovich's 2010 book is very good reading), but does it bear the spin Lehrer is putting on it? No.