To understand how a helmet works we first have to look at the source of the injury, in this case we consider blunt impact (note that this can differ from ballistic impact and blast events, which we’ll cover in future posts). Blunt impact can result from falls, vehicle accidents, or being hit by a blunt object, but in all these cases the main consideration is acceleration of the head.
A space shuttle hits 3g (3x normal acceleration due to gravity) during launch, and fighter jet pilots can approach 10g with use of a G-suit, going beyond this can result in loss of consciousness. So it may come as a surprise that current military blunt impact standards allow up to 150g of head acceleration, and this is actually low compared to sports helmets like bike and ski helmets that allow up to 250 or 300g. The reason for this is time, while pilots can endure sustained single digit g’s, the typical head impact is over in much less than a tenth of a second.
The correlation between head acceleration and risk of injury dates back to the 1960s and the Wayne State Tolerance Curve, the development of which involved dropping cadavers down an empty elevator shaft. This correlation allowed helmets to be developed that would keep your head below critical levels (e.g. 300g for just a few milliseconds). Over time these helmets have proven effective at preventing the type of life threatening injuries they’re designed for.
In more recent times mild TBI (mTBI) or concussion has begun to take more focus, with studies showing that concussions can occur at lower G levels. However, there isn’t such a clear cut threshold for mTBI as there is for catastrophic head injury (to give an idea of this, concussions were associated with impacts ranging from 16.5g to 177.9g in one study of football and hockey players). Part of the difficulty is that rotational forces, impact directionality, and person to person variation start becoming much more important. There is also a wide range in the severity of mTBI, and it’s speculated that even sub-concussive impacts, those which don’t result in any diagnosable injury, can have long term effects. This all makes it difficult to truly optimize a helmet for mTBI prevention, but in general it seems that lower head accelerations will probably lead to better outcomes.
 Duhaime, et al. “Spectrum of acute clinical characteristics of diagnosed concussions in college athletes wearing instrumented helmets." J Neurosurg. / October 2, 2012