The fact that selection pressure for mental ability is everywhere present is an excellent point; thanks.

Even though intelligence helps everywhere,* both the benefit and cost from increased intelligence can vary. For example, brains consume quite a bit of calories--and turn them into heat. Everyone is going to have to pay the caloric cost of powering the brain, but the cooling cost of keeping the brain at a healthy temperature is going to vary with climate. Foresight is going to be more useful the more variable local food availability is.

* Well, actually, this should be poked at. The relationship between intelligence and reproductive success could easily be nonlinear, even among early hunter-gatherers and farmers. It's not genetically favored to be smart enough to outwit one's genes! (The effects of widespread female education and careers are too recent to be relevant for this conversation.)

I always maintain "If there is a quantitative difference, I sure as hell hope we never find it."

? We can already measure intelligence, and have good estimates of heritability from cross-generational intelligence testing. We've found the quantitative difference. All that's left to find out is how it works under the hood, which is knowledge we could use to re-engineer things to make them better. Why stop at discovering that piece?