Well, IIRC PD advertised for and hired new audio engineers a while back during the final stages of GT6 updates. A new synthesis engine was developed for engine (and exhaust) sounds, the flavour of which (type of synthesis) was likely "physical modelling" IMHO.
I seem to recall that the last few cars added in the final GT6 updates showed signs that the (possibly-unrefined) synthesis engine had been implemented at that point. GTS obviously uses it throughout the car range now.
The synthesis approach, especially if it's a physical-modelling one, offers many advantages over sample-based ones:
The disk-space footprint would be minuscule compared to that required for multiple audio samples for each car.
Individual components of the overall sound can be tweaked independently-and-quickly with no need to resample anything.
Creating sounds for new engines (cars) should theoretically be super-quick as the specs (cylinder numbers, capacity, exhaust length, turbo present / not-present etc.) could be entered into the synthesis engine, placing the sound in the ballpark. The refining process for each engine's particular sound would IMHO require the most time, but that could all be done in software whist comparing with real-world recordings.
So, to answer the OP's question, I do think we're hearing the benefits of the synthesis engine predominantly; the new audio engineers are likely icing on the cake. That said, they might've had a hand in developing the algorithms, in which case they deserve credit too, but I suspect that their primary contributions would be in the traditional AE realm - choosing and tweaking ambient effects (the sense of space in tunnels or lack thereof out in the open), balancing multiple sound sources (crowd noises and the various vehicle sounds etc.) and so on.
This is all pretty much speculation on my part as an AE, so take it with a grain of salt if you like.