Life expectancy for human has significantly increased during the last century or so. We all know that there are many reasons that are not linked with "evolution", but I am wondering if such change in life expectancy has been observed "naturally" (not counting animals in zoos or domestic ones). Does evolution favor long lasting individuals?
As far evolution is concerned, there is no benefit whatsoever in having long lived individuals. Evolution only "cares" about individuals while they are capable of reproducing. Survival past the maximum age of reproduction is irrelevant from the evolutionary point of view since it does not make you any better at reproducing, and therefore your genes will not be selected for.
This is highlighted by the various species that die to give birth to or feed their young. One of my favorite examples is the giant pacific octopus (Enteroctopus dofleini) whose females lay their eggs and then die, leaving their body to feed the emerging young.
Now, a mutation that makes an individual both longer lived and capable of reproducing at a greater age could well be selected for. It just always needs to be combined with the ability to reproduce. By itself, dying older will confer no selective advantage unless it is combined with producing more offspring.
The life expectancy of an animal (or anything) is just one aspect of how 'fit' they are to survive. having them live too long can be a detriment to the species.
There is no simple 'better/worse' relationship of fitness to life expectancy. In simple population models predators which reproduce to much or live too long will tend to eat all the prey and starve to death for instance.
The most famous example of life extension molecular biology is the discovery of a 'Methuselah worm' by Cynthia Kenyon about 10 years ago. With a single mutation these C. elegans worms can live about 50% longer, and seem to be pretty energetic while doing it.
It would be interesting to put some of these Methuselah worms into a wild population and see what happens to their life span as they interbreed. C elegans is a highly programmed organism - each cell has a predetermined plan for development and lifespan, which made this worm relatively easy to find.
For human beings (the best studied animal on Earth) there is no simple genetic relationship to longevity for most of us, but there appear to be some simple combinations that are responsible for the oldest living humans:
a genetic model comprising 150 SNPs in order to compute the predisposition of an individual toward EL. Their model successfully predicted exceptional longevity in a different sample of centenarians (individuals that live to age 100) with 77 percent accuracy. This demonstrates that EL is strongly associated with complex combinations of genetic variants.
So for humans this experiment has sort of been done. Longevity phenotypes are possible but don't overall convey a reproductive advantage and so while they hang out in the gene pool, they aren't enough of an advantage to reproduction to be spread about widely...