I haven't kept up with the technical developments, but when I was experimenting with an Oculus dev unit a few years ago, I remember a good deal of work going in

I haven't kept up with the technical developments, but when I was experimenting with an Oculus dev unit a few years ago, I remember a good deal of work going into creating the frameworks and tools to generate good, positional audio as an important part of creating immersion. Makes sense that this would be both useful and easier to do well since there actually is positional data for where sounds ought to be coming from. Ideally you'd have the audio equivalent of the tools used to simulate generation, occlusion, reflection, and refraction of light on the graphical side. But adding it to stereo tracks via post-processing is a lot harder to do well. It reminds me of "simulated surround" effects on my old home theater boxes but with more panning around instead of just trying to route certain frequencies to the proper speakers.