The human eye only really focuses on an area about the size of one word, but moves quickly (saccades) to focus on whatever part you want to see at that moment. The rest of your vision (peripheral vision) has limited functionality to quickly guide a saccade towards any part of it, to detect changes (raisin an IRQ) and an extremely low resolution of general vision (enough to make our . You can't even read one word of text while looking at the one next to it, and if you think you can, it's because you already know what it says. Part of this effect seems to be a lower physical resolution and part of it is because your visual cortex spends its neurons interpreting the center more precisely rather than interpreting more area more loosely.

I don't think that's entirely accurate, because this can also apply to perceiving entirely new UIs you've never seen before. Familiarity helps, but I don't think it's entirely that.