This part of music theory is very flawed IMO and maybe someone can change my mind but there is all this talk about resolving dissonances when studying voice leading and most textbooks seem to be saying that certain tones need to resolve in certain ways and while that makes sense to me what doesn't make sense to me is that those tones need to resolve in the same voice. For example, if the 7th of the dominant 7th chord should resolve to the scale degree 3 in the next chord then why should it do so or need to do so in the same voice?
It can be quite jarring to the ear when a dissonance in one voice resolves in a different one. As an easy way to demonstrate this, play an ascending C major scale on some instrument, but when you arrive at B, end the scale in some octave different from the expected one. So, for example, on a piano: C4-D4-E4-F4-G4-A4-B4-C2.
But keep in mind that the rules of theory you're studying were very much bound up in the aesthetics of the times when they were developed. The modern ear, by virtue of composers who chose to bend and break the rules, can tolerate dissonances that in earlier times would have seemed bizarre. The same goes for unexpected or "incorrect" voice-leading.
These rules are still taught, because they give useful guidelines for beginning to understand how Western music works.
...what doesn't make sense to me is that those tones need to resolve in the same voice
The rationale is about voices. It will help to think about the origin of contrapuntal music: The Church and vocal music. An important part of voice leading - I suppose it might be better called part writing - is to create singable lines. That means singable in the literal, not theoretical sense.
If you switch over to a harmonic, homophonic style of music - which I suspect you are in regard to this question - then the integrity of the "vocal" lines for resolution is less important. Although a lot of non-vocal music will still show basic voice leading principles, because the classical style was very circumspect and conservative for a long time.
One place I remember reading about the resolution of tones being "split" between "voices" is in Ralph Kirkpatrick's book about Domenico Scarlatti. But, keep in mind Scarlatti's style was very eclectic. He did a lot of wild stuff compared to his contemporaries, someone like Handel.
To the extent that "splitting" resolution between "voices" works, I think you would technically attribute it to the octave equivalence of pitch classes. In other words, if all the octaves of a pitch class have a kind of equivalence, then it doesn't matter so much which octave a tone resolves in.
Here's my uneducated amateur take on the matter. The ear's tracking of voices corresponds to tracking moving objects based on the sound they make. Objects such as a deer or a lion. The timbre or spectral shape and other characteristics of a sound play a role in the tracking, not only the main pitch. If your ear is tracking a deer, and you expect the deer to go somewhere, but all of a sudden there's a lion in the place where you expected the deer to go... maybe that feels like a natural ending? But from an object-tracking perspective, you'll probably assume that there was some sort of a discontinuity on the deer's path. And metamorphosis from deer to lion might not be the most natural explanation.
If you place your resolution in a different voice, most likely this will be heard as a discontinuity.