Not expecting reality but just certain aspects of it. For me it mostly comes down to smooth extended frequency response on and off axis with minimal background noise and the ability to get fairly loud without excessive distortion, and an appropriate amount of room reverberation. Like others have said it can actually be better than reality a lot of the time. With ambient sounds that might plausibly be happening in or around my house my system can truly deceive me at times.
How much “suspension of disbelief do you need?”
We (or most of us) believe that it’s very difficult if not impossible to hear an exact representation of the the sound of a live performance on a recording.
The question is how much do you have to delude yourself into thinking it’s the real thing your listening to, to satisfy yourself.
To some it has to to be as close as possible. But others can make allowances for defects in the sound in order to enjoy the presentation.
‘How much do you need?