I've heard IM and GM commentators on chess tournaments, from the Opera Euro Rapid to PogChamps 3, suggest that retreating moves, such as backwards knight and bishop moves, are unintuitive. I've felt this way while doing certain puzzles as it takes me a long time to find the retreating move after a successful first few moves. But I'm curious as to why this is the case. Is it human psychology, or does it have something to do with fundamental principles and piece development? Unfortunately, I don't have a specific board state or game scenario to this display, but any answers would be appreciated.
I think it's human psychology; we associate forward moves with attack and backward moves with defense. After all, chess originated as an abstract model for a real war.
That said, retreating moves don't need to be unintuitive, as long as we're on the defense. I don't have a specific situation in mind either, but retreating a queen or rook to help defending your king from a mating attack feels familiar.
I've felt this way while doing certain puzzles as it takes me a long time to find the retreating move
That's a good observation: in puzzles, you are almost always on the offense. It's always 'White/Black to move and win material', 'White/Black gives mate in x moves'. In rare cases, it's 'White/Black to move and draw' but even then it's often finding an offensive combination to equalize or force a draw by repetition or stalemate.