The visual system is capable of creating very detailed visual images. These images are often (always?) more detailed than it is possible for the visual system to create in real time. Much of the detail is supplied from complex processing algorithms applied to data stored in long term memory. A good example of such images is the detailed image of regular brick walls. I can stand in the midst of 3 brick buildings and see what appear to be very precise images of the brick walls. The images seem to respond in real time to changes in my point of view due to moving my head and/or walking/running. The visual images are much more precise than what is possible via real time processing of the data from my retina I happen to be near sighted (ordinary myopia). When not wearing contacts or other corrective lenses, I see fuzzy images of brick walls. Why does my processing of the data from the retina not create better images? I think it could do so. Might there be some evolutionary advantage gained by generating images which approximate the fuzzy reality implied by the fuzzy data provided from my retina?
The precision we've obsessed to achieve is higher than our evolutionary state. It's a failing rather than an advantage.
Your mind may fill in details of an image, but it's not going to invent an image out of whole cloth. As a myope, the image you see without correction is so bad that there's not enough for your mind to build on.
damn, thats a good question. Same with the blind spot, bespectacled people dont see a fuzzy extrapolated fill-in blob, but a clear one