You've definitely got a point. But that ~20kHz figure to my knowledge was not just determined by "can the ear hear this" but by "can the person hear a differenc

You've definitely got a point. But that ~20kHz figure to my knowledge was not just determined by "can the ear hear this" but by "can the person hear a difference". But yeah, theoretically, it's possible that all that science was only conducted with, say, headphones. Like with all things science, everything is only falsifiable, never 100% provable. But yeah, I think we're indeed on the same page, the claims that audiophiles make are often falsified already. Or make so little sense that it's akin to "eating an apple in the morning shifts Helium's absorption spectrum by a few nm". You can attempt (and succeed) to falsify that, but it's such a waste of time and energy...