Sometimes it seems very obvious that a given paper is by specific authors (either because of style, or because of how familiar it is with their previous work) but I've had many experiences where I later learned that my supposition was completely wrong. Similarly, when you encounter a paper that doesn't have any obvious cues (which is the overwhelming majority of them) then it's pretty much impossible to tell whether it's an author you admire or someone you've never heard of -- and this is a good thing.
Some conferences don't use blind submissions, and yes: I have felt an awful lot of influence there. "Surely [famous Turing-award winning authors] don't need me to double-check their proof."
This is a good initiative, but a catch is that if he's the only person deliberately spelling some words British and others American, the spelling choice becomes a unique identifier.
Though, it could work as long as no one within the group knows who is using the varied spelling.
I'm not sure how to fix this problem though. In our phage newsletter we try to avoid using names and universities to focus on the paper/topic/finding itself, but I keep finding myself looking at author names and affiliations before diving into any paper.
I know it's "wrong" and I recognize myself doing it, but I still do it all the time.
People should copy the writing style of famous auctor's then to expose the system and keep people honest because then they know there are copycats.