Lets look at this from the point of view of the cartridge:
The stylus tip is vibrating, the energy of this exication is supposed ot be damped linearly with respect to velocity and frequency by the damping structure such that there is no overshoot or frequency non-linearity. this is pretty much an impossibility, but good cartridges get damm close.
The energy not damped is then transmitted through the cartridge body through the coupling between the body and headshell and towards the next coupling , i.e the bearings.
Each of these couplings is a mechanical impedance mismatch ie they couple better at some frequencies than others. This means tha at the frequencies that they don't couple well, they reflect energy back to the cartridge.
The neoprene foam creates a high impedence (lossy) coupling between the cartridge and headshell. It absorbs the energy being transmitted from the cartridge. The issue is that it also deflects quite easily which can interfere with the cartridges ability to accurately track transient information.
If the foam deflects instead of the damper, the coils dont move accurately and the rebound of the foam against the vinyl cause a movement of the coils that are not part of the recorded information.
The audio effects of this device are that in arms that do not have sufficient damping of the arm tube, you will hear better intertransient silences and definition but in all cases you will hear a dilution of transients.
It is my opinion that you have as low an impedance in the coupling between cartridge and arm such as to draw any extraneous energy away from the generator. Then the judicious use of armtube thickness, internal damping and bearing design will result in no energy being reflected back into the cartridge generator assembly.
The folks at SME are not dummies.
If you have an excitable cartridge, use the foam, it is certainly cheaper than buying a new cartridge.
HTH