This may get disturbing...

[kengr]

Harvard Implicit Association Test

I tried the demo about fat/thin. I'll probably try some of the others.

Comments

[mazarinade.livejournal.com]

Have a care with that one. It seems to be designed for a push-poll bias based on the normal distribution of dominant handedness. And the push-polling on the black/white weapon/harmless object one was somewhat suspicious.

[siegeengine.livejournal.com]

I think it's kinda cool, but also I agree with the above that some of what they're trying to do is a little suspect. I may be reading into it though. It is fun, but nerve wracking. ;)