well one can often encounter arguments of how this or that function is 'hardwired' by evolution, which makes very little sense in light of a: evolution's slowness, and b: early brain damage to those regions not always resulting in loss of that function or any strong disadvantage. (perhaps the region where the task usually ends up implemented gets slightly tuned for the task, but that's it)

The relevance is that the very implementation of e.g. visual memory may differ between individuals, to the point of structuring the data in radically different ways. It seems that as brain develops, there's a great deal of hill-climbing of some kind performed by the brain to implement each particular function, and different hills end up climbed even for pretty ordinary functions.