We generally do not track technical differences like sampling depth or even file type (lossless vs. lossy). As a rule of thumb, you can think of the case for different recordings like this: For every time the artist sat down with their guitar and played a song, there should be one (and only one - generally) recording.
An additional concern is that it’s really hard to be sure that your high-resolution music is actually high-resolution, i.e. if it wasn’t just upsampled from a CD-quality recording (for the case of lossless music stores, it’s not that rare to get FLAC downloads where it’s immediately obvious from a spectrogram that they are just converted 128kbps MP3s).