I wonder how you measure the quality difference for higher rates? 384K, 512K and beyond? I hear from audiophiles that there is a very distinct difference, but there is absolutely no basis for it in science.
Oversampling in studio recording is mostly about eliminating aliasing in software that's producing distortion, and it's only relevant in that context: I don't think it's nearly so relevant on, say, an EQ.