Rubber vs. Foam Surrounds?

The focus will be midrange performance over all else and I was looking at the old Audax PR17 drivers as the closest equivalent to what I'm aiming for, even though the end goal is rather different. These drivers have an almost flat surround that is made from foam and a very minimal xmax of +/- 0.5mm.

So perhaps I should re-frame my question and ask the gurus here, why does the Audax PR17 use a foam surround rather than rubber?
I recently swapped out twin 4″ SB12NRX25-8 with rubber surrounds for twin 4″ SB12NRXF25-8 with foam surrounds in a M-T-M configuration. The purpose was to compare the performance as mid-range drivers in the band 150Hz-2.5kHz. SB Acoustics claim "Foam surround for maximum transparency and midrange resolution"
https://sbacoustics.com/wp-content/uploads/2021/03/4in-SB12NRXF25-8.pdf
From the rubber surround version data sheet "High damping Butyl rubber surround for non-resonant sound character"
https://sbacoustics.com/wp-content/uploads/2021/03/4in-SB12NRX25-8.pdf
The foam surround appear much stiffer than the rubber surround which is very flexible. This seems to be confirmed by the foam Fs = 63Hz vs the rubber surround Fs = 57Hz. The two driver appear identical otherwise. Surprising to me, the foam does sound better than the rubber to my ears in this purely mid-range application. More natural and less "shouty" at all volume levels.
I don't buy the claim that foam was used for cost saving reasons.
For woofers installed in boxes, the behaviour of the surround to deal with the large pressure variations and excursions associated with low frequencies would be important. How is any surround deformation (billowing) with the pressure variations affecting compliance - is it modulating fs, for example?

For midranges, how well is the surround material terminating the cone to damp resonances within the cone?

For midwoofers, is it possible to satisfy both the above satisfactorily - or is it a compromise too far?
Like many others I have had foam surrounds disintegrate in the past (KEF B139) and have avoided them till recently. My inclination is to use rubber surround drivers if large excursion is likely but my recent experience make me willing to use foam mid-range specific drivers for sound quality reasons.
Foam is lighter and more flexible.
Not always the case as my comparison above described.
Rubber gives better damping to cone resonances, more robust and usually lasts longer.
I agree with robustness and longevity. Rubber is elastic thus stores and returns energy.
 
Also keep in mind that some modern foam surrounds are actually made from foamed rubber, not the traditional polyurethane foam, which had a reputation for deteriorating quickly under more unforgiving environmental conditions. Foamed rubber can also be "dialed in" by density to exhibit certain desired dampening characteristics. This can be used to control cone resonances, breakup and reflected components back into the cone edge (midrange bump), plus the material is lighter in mass than many thinner solid rubber surrounds. Scanspeak has some 7" drivers with foamed rubber surrounds.

I prefer midrange drivers with foam surrounds, mainly becuase they're typically less dampened in the upper midrange and usually tend to sound more open. This makes cone design much more critical, since the surround won't dampen much of the less desired resonance and breakup modes higher up.
 
The frequency response plots for the rubber and foam surrounds of the SB Acoustics 4" drivers appear identical below 200Hz but the foam surround response appears more even in the frequency range 200Hz-4kHz and has better dispersion.
Interestingly only the 4" Norex cone drivers are offered with foam surrounds. All of the larger cone drivers of the extensive catalogue have rubber surrounds.
The foam surround does appear to be a darker hue than the typical polyurethane foam surrounds of yore but I can't determine if they are foamed rubber. I do agree they sound more open, especially on vocal recordings.
 
Comparing the specs of the foam surround version against the rubber surround version, the biggest disparity is the compliance Cms = 1.18 mm/N for the foam vs Cms = 1.49 mm/N for the rubber. Assuming the spiders are the same, the rubber surround produces the higher compliance. This supports the conclusion the the foam surround optimises mid-range performance for this design. Other manufacturers may do things differently.