I fully agree with your observations about good or bad photography being dictated in part by its easy visual readability. But the article presented here seems

I fully agree with your observations about good or bad photography being dictated in part by its easy visual readability. But the article presented here seems unscientific and misleading to me. They say "the brain avoids looking at" and then the "evidence" that they produce is an AI prediction generated by a closed-source (meaning unverifiable) software. It would be correct if they say "3M's prediction ignores glass skyscrapers", but that might just as well be a bug in their algorithm. As a counter-example, the Keppler skyscrapers in harborfront Singapore are mostly made from glass to fight off the heat, yet they are pleasing to look at and have very interesting reflection patterns. And they are certainly remembered by locals as a great navigation landmark.