Now I think I understand why your switch swap test failed.
The power window switch has 4 sets of contacts. In the center ("rest") position, contact set 1 is closed and connects one side of the motor to one side of the battery (let's call it "battery minus"). Contact set 3 is also closed and connects the other side of the motor to battery minus as well. Contact sets 2 and 4 are open (effectively disconnected). So, both wires to the motor are connected to the same point (battery minus); therefore the motor does not operate.
When you push the UP end of the switch, contact set 3 and 4 are not affected. Contact set 1 opens and contact set 2 closes, connecting that end of the motor to "battery plus". There is now a complete circuit, starting at battery plus, through closed contact set 2, through the motor, through closed contact set 3, and finally to battery minus. The motor runs in one direction.
Similarly, when you push the DOWN end of the switch, contact set 3 opens and contact set 4 closes. There is now a complete circuit, starting at battery plus, through closed contact set 4, through the motor (in the opposite direction relative to before), through closed contact set 1, and finally to battery minus. The motor spins in the opposite direction.
To substitute a "known good" switch, you need 2 switches. For example, if contact set 3 is bad (or the wire to it is broken), and you substitute or parallel a switch for contact set 2, nothing will happen when uyou operate the test switch, so you would assume the factory switch must not be the problem. But this might be a wrong conclusion.
I hope I didn't make this more complicated than necessary. I wish I had a way to make a drawing in the message box.
If you had a broken wire or socket, both windows should have a similar problem. Just one bad window suggests a bad contact or runner (wire) wiithin the switch.