Which is why the research into this topic focuses more on headbands, which mostly protect the forehead, than the entire head. See http://www.copenhagenize.com/2009/10/australian-helmet-scien... for a summary of Australian research and a sketch of the head gear. They quote:
> But more impressive were the estimates of introducing protective headwear for car occupants. The authors of the report estimated that the annual reduction in harm would be in the order of $380 million. The benefit of padding the head is that the head is protected from strikes with unpadded automotive components, exterior objects and in vehicles that predate any eventual introduction of padded interiors."
The first thing anyone says in a small ish accident (especially as a cyclist/motorcyclist) is SORRY MATE DIDN'T SEE YOU THERE
So imagine that blocking a good section of your forward vision, and a helmet blocking anything to the sides... motorcycles have very small blind spots as the rider can swivel anywhere with no cage blocking them. Helmets are a minor contributor. But cars don't have that luxury.
Sadly never made it to production.
Race tracks don't have most of those hazards, and control most of the pedestrian hazards, with the limited pedestrians on the hot side of the pit wall assuming much of the responsibility for their own safety.
This would be offset by a reduction in head injuries (with a possible increase in neck and spinal injuries). I think I'd bet a full study would show an overall decrease in safety.