In Chopin and other Romantic composers the pieces cry out for interpretative details that are only hinted by the score - rubato, subtle dynamics, inner voices, pedalling etc. Listening to other pianists unlocks these hints for the student interpreter who can then learn how to do it on the piano. In the process, they then develop their own taste and views of what works for them in the music.
...listening to other pianists in lieu of learning a particular piece can also infect you with their individual style of playing, which is a point I made in a previous post: as a result you're more likely to copy at least part of these pianists' style of playing, which means it will not be original at best, and you will copy their bad habits at worst. Either way, you will most likely end up with an unconvincing overall interpretation of the piece, and the piece will take longer to learn well.
In contrast, I find that it is much more beneficial and effective if you learn the piece yourself first, going 100% with what the music says, working at it with your teacher. This is, after all, how the master pianists would've learned the piece to begin with. Chopin wrote plenty of details in his music (where to change tempos, dynamics, articulation, pedalling...). Once you do that, you can start putting in more details that are not in the score (such as rubato, tone colour changes, phrasing...), at which stage you can start listening to what other pianists do for more creative inputs (it is interesting to note that, oftentimes, one finds in the score the explanation as to why certain pianists play the piece a certain way).
Allow me to share with you my personal experience many years ago with learning Chopin's Ballade 1. I had heard a recording by Youri Egorov a lot before I learned the piece myself. When I finally got around to learning it, I actually did some of the details he did (especially as regards rubato, tempo, and phrasing) without realising it. In effect, it took me 1 whole year working with my teacher before I was able to come up with something convincing.
When I looked back, I realised that essentially I had to remove every details I've added to my playing from the Egorov CD and go solely by the music, before I started adding details myself - with my teacher's guidance of course. I realised that, if I hadn't heard the recording, I'd probably be able to come up with a more convincing performance in a shorter period of time.
So, sure, recordings can be wonderful tools for learning a particular piece, if the user/learner can distinguish between what constitutes the actual style of music and the performer's interpretation. I find that usually with Chopin's music, it's largely determined by how well you know the score of the piece since he wrote so much detail in it.
This is not to say that one should not listen to recordings at all of course! It's just that the purpose of hearing other people's performances should be to enlighten, to entertain, to enrich, and not so much to simply learn about how to play a particular piece they happen the be playing.
However, the same experiment could be probably be done with Bach and it might turn out all right - because Bach does not depend on performer interpretations to the extent that Chopin does.
Uh, no - it's the other way around. Think about it: how much musical indications did Chopin write in his score? Much more than Bach, that's for sure. No tempo indications. No articulation indicated. Next to no dynamic markings. What about ornaments - where do you put them, and how do you play them? And rubatos exist in Bach's time too and is also frequently used in performances today, yet Bach never wrote "rubato" anywhere. And what about pedalling? Do we use it or not? If so, where? The list goes on. And on.