I'm not confused. I'm simply stating in a rather polite way that some people tend to imagine they're hearing differences that don't exist.
When double-blind tested according to scientific standards, most of these bogus claims tend to break down.
But more importantly, if some Joe comes along and tells me he can hear the difference between ADPCM and PCM because "PCM has more bits" or some idiotic lay argument like that... He may indeed believe he's hearing a difference, but he'd be imagining it. He may have fooled himself into thinking this when the truth is that he just doesn't understand how the exact same analog signal can be reconstructed from fewer data.
The problem is, I hear all sorts of arguments predicated on the misconception that fewer data always means lower fidelity... without any comprehension from the other guy as to how fewer data can reconstruct a fundamentally identical signal.
If a person claims they can hear the differences between two identical analog signals reconstructed from two different types of digital compression, then they are either lying, imagining, or have a problem with their hearing... because it is the reconstruction of the analog fundamental which is of utmost importance. Understanding that, as I've stated repeatedly, requires an understanding of what actually is required in a digital encoding system to fully reconstruct an analog signal in a way that is perceptually transparent to the listener.