I've spent the morning looking around the Web to bring you today's news snippets, but then I came back to ScienceBlogs and realized that the best posts on cognitive science are being made right here.
Jonah Lehrer has an excellent analysis of Malcolm Gladwell's Blink. Having just finished the book myself, I tend to agree with most of what Jonah says. "Instead of delving into the scientific details," Jonah says, "Gladwell always goes with the telling anecdote."
I might add to that: I've caught Gladwell in at least one muff of the details. When discussing Nalini Ambady's work with thin-slicing and teacher evaluations (the idea that watching short video clips of teachers is as accurate as end-of-semester evaluations), he claims that there is "no difference" between 5- and 10- second video clips. That's not actually true: the 5-second clips were less reliably correlated with teacher evaluations, and in fact even the 2-second clips gave better results than the 5-second clips.
I'm not sure I agree with Lehrer's conclusion, that Gladwell's biggest problem is discussing the mind without talking about the brain (Why not talk just about the mind? Does every book about psychology research have to include brain scans?), but otherwise I agree with his analysis.
Chris at Mixing Memory has two posts about Gentner and Nunez's work on time-space metaphors. In the first post, Chris makes an effective case that Gentner's results are little more than examples of priming. In the second, he suggests an experiment to resolve the question. I'm not sure I'm convinced by Chris's line of reasoning -- after all, if the words in the "priming" condition are different from those used in the "test" condition, and the results are still signicant, then what's being primed? Is it not the metaphor itself? However, more data is always welcome, and I'd love to see an experiment like the one Chris proposes.
Finally, Shelley at Retrospectacle has a great post about brain-computer interfaces. This is cool stuff. Even if it is awkward and impractical in many applications, it'd be neat if some of these ideas got off the ground.
- Log in to post comments