Spatial Sound is “expanding” the sound channels, injecting left sound into right and vice-versa. It gives the impression of a more enveloping sound when using 2 speakers.
When using headphones you’d need HRTF instead, and this is notably what Valve Audio lib is doing and why HLA sounds so realistic for example. In effect you can locate sound source origin as if they were coming from the real world.
However I don’t believe FS2020 developers are using any specific VR sound processing which works only in VR but not in 2D. Instead, they’re using the Wise audio lib and/or using the Wise effects or the Win10 effects. And with FS2020 Spatial Sound ON, it is clear to me my right ear for example is earing inversed phase waveform derived from the left sound channel, and vice-versa. It also amplifies sounds so that you can hear for example the engine louder or quieter depending on the orientation of your head, whereas you’re supposed to hear the same level regardless of the head orientation (just the direction changes).
Now in VR, the directionality is supposed to be handled by the software and the Wise sound lib is supporting this naturally: tell it location of source and ear, then add some occluders and reverb specs, and it is supposed to do the math. The question: is FS2020 using 2 ear positions or 1 only? is it using HRTF or not? (search online what it is if you’re wondering).
All in all, I find it sounds more natural regarding sound source orientation and perceived level balance between the 2 ears regardless of direction (including looking up/down) with FS2020 Spatial Sound OFF.
Can you lease detail which Win Settings are you referring to?