Economics professors at Cornell and Indiana U. have found a possible correlation between watching TV before the age of three and autism. The evidence looks even more circumstantial than the study linking early TV viewing to ADHD, but still interesting: really what they’ve found is a correlation between diagnosis of autism and the number of rainy days in a particular county for a given period, which is known to correlate with hours kids spend watching TV. I wonder if they also looked at birth month and whether that has an effect — if it did that might imply a critical period of only a few months. (Thanks to Andrea for the link.)
Update 3:30pm: Here’s the actual study. Plus, Steven Levitt offers some skepticism at the Freakonomics blog. (Thanks to Judith for the links.)