There was research in bees some years ago that also tested this hypothesis and found evidence for a genetic link between robustness and intelligence:

http://www.scientificamerican.com/article/old-and-wise/

Also see this from last year:

http://www.lse.ac.uk/newsAndMedia/news/archives/2015/07/Link...

"They found that, within twin pairs, the brighter twin tends to live longer than the less bright twin and this was much more pronounced in fraternal (non identical) twins than in identical twins."

The intriguing question here is the identity of the dominant mechanism behind the web of associations in human demographic data that link education, intelligence, social status, wealth, and life expectancy. The safe assumption has long been that better lifestyle choices and better use of (and access to) medical services were the driving force here, with smarter people doing better in both of those. Exercise and calorie intake have large effects in comparison to other factors that can be pinned down. It is generally thought that genetics plays little role in mortality until very late life; if it turns out an intelligence-robustness link is meaningful in comparison to exercise and calorie intake, that'll be a turnaround for the consensus. Though extracting compelling evidence from the statistics will be a real feat of arms.

> It is generally thought that genetics plays little role in mortality until very late life

I don't understand this at all. Risky behavior is largely driven by being male, a genetic effect. Within males, it is still largely driven by genetics. Terminal diseases that strike young people are also often genetic problems. How could it be generally thought that genetics plays little role in early-life mortality? What are all the early deaths that are viewed as not implicating genetics?

So, I found US death data by age and sex for 2007 ( http://www.cdc.gov/nchs/data/dvs/mortfinal2007_worktable310.... )

Here are male deaths, expressed as a percentage of female deaths, for every year of age from 0 to 30:

     0  127%
     1  119%
     2  130%
     3  131%
     4  142%
     5  138%
     6  123%
     7  129%
     8  122%
     9  125%
    10  126%
    11  152%
    12  143%
    13  145%
    14  178%
    15  193%
    16  199%
    17  248%
    18  278%
    19  313%
    20  317%
    21  317%
    22  342%
    23  310%
    24  314%
    25  291%
    26  261%
    27  269%
    28  250%
    29  231%
    30  237%

Odds are good (I haven't checked) that the male population in this age range is larger than the female population. But it's going to be larger by less than 10%, which is completely neglible in the face of the factor-of-2-or-3 differences in deaths.

This isn't "little role in early-life mortality", it's a gargantuan genetic effect you'd have to devote quite a lot of effort to ignoring. (And if you meant something different by "early life", male deaths stay in the range of 150-200% of female deaths right up until the early 60s. Female deaths finally exceed male deaths once people hit 80 years old.)

> Female deaths finally exceed male deaths once people hit 80 years old

Absolute numbers-wise, simply because you are running out of males.

This is of course one of those things that is interesting, but of no relevance to the future of your life and life span. If all goes well in the research space, the pace at which rejuvenation therapies arrive over the next 20 years will be the dominant driver of your mortality rate for the rest of your life - even just one narrow treatment after the SENS model capable of repairing one form of cellular damage that contributes to aging should have considerably more impact than exercise or calorie restriction.

> which rejuvenation therapies arrive over the next 20 years will be the dominant driver of your mortality rate for the rest of your life

I think you are being overly optimistic. Within 20 years I don't think there will be much change in that field, because you'll need clinical trials, proof of safety and so on, and if it's not ready right now in terms of what needs to happen conceptually and at the lab scale level, we already know it won't be on the market in 10-15 years from now.

If it was up to the US, maybe things would be that slow, but I think you are not properly appreciating how fast CRISPR is advancing. The CEO of BioViva has already injected herself with edits relating to telomeres, and it's about ready for embryo editing: the main barrier was off-target mutations (this was the problem with the Chinese embryo editing last year), and that has effectively been solved with some tweaks to the enzyme to yield ~0% off-target mutations. As stem cell treatments show, the bioconservatives in the USA can try to slow things down, but the USA is not the only place in the world with biomedical expertise these days.