Functional imaging of the human brain: A window into the organization of the human mind - Part 2

Nancy Kanwisher, MIT

[Garbaz]

Regarding the "Why of the why" at the end:
Isn't this at least in part simply due the inherent differences between the two tasks of object and face recognition? In the former case, you have a highly variable space, requiring much more generality, but less precision, whereas the latter has very little variance, faces are overall pretty similar, but requires very specific attention to nuances between stimuli.
What would be interesting to test, is whether humans would develop a similar specialized region, if tasked with a different task of similar specificity, like differentiating leaves of a specific type of tree, from an early age/over a long period of time. Same goes for CNNs of course, but I'd expect the results there not to differ, given there is nothing inherent about faces for them.

[rickharold]

I love this stuff. So awesome. Thx for sharing