LONDON, England / NATURE / News / August 24, 2010
Despite its anecdotal support and intuitive appeal, the grandmother hypothesis lacked much quantitative proof showing that it was possible for longevity to....
'Grandmother hypothesis' takes a hit
Pinning longevity to benefits women bestow on their grandchildren may not be plausible
says a report by Ewen Callaway published in NATURE.
The evolutionary biologist William Hamilton initially proposed the idea in a 1966 paper2 that built on the theoretical work of George Williams and Peter Medawar.
Kristen Hawkes, an anthropologist at the University of Utah in Salt Lake City, and her colleagues.
These researchers found that among Tanzanian hunter-gatherers, the Hadza, mothers faced a trade-off between foraging for food for themselves and any weaned offspring, and caring for new infants. But if grandmothers helped with foraging, they were rewarded with healthier, heavier grandchildren who weaned at a younger age. Over evolutionary time, this fitness boost could have selected for women who survived long past menopause, an anomaly among humans' evolutionary kin.
"Chimps almost never live into their forties in the wild, but most humans, if they're lucky enough to make it to adulthood, live beyond the childbearing years," says Hawkes.
Further support for the grandmother hypothesis came from studies of other subsistence cultures, as well as from historical records, although not all studies back up the hypothesis.
Friederike Kachel, an evolutionary biologist at the Max Planck Institute for Evolutionary Anthropology in Leipzig, Germany.
Doing the maths
Despite its anecdotal support and intuitive appeal, the grandmother hypothesis lacked much quantitative proof showing that it was possible for longevity to evolve from grandmothering, says Kachel, whose team ran a mathematical simulation to test the theory's plausibility.
Their agent-based model, which simulates the actions and interactions of individuals, begins with a population of 1,000 people whose lifespans and reproductive windows are an inherited trait that mutates over time. Half are female. Initially, the females die when they stop reproducing, at 50. In some simulations, when females are not nursing their own children, they provide fitness benefits to their grandchildren by reducing the age at which they're weaned.
After about 500 generations, the model demonstrated that the assistance of a grandmother during infancy shortened the interval between the times their daughters give birth, and led to shorter reproductive windows. However, compared with simulations in which grandmothers did not help out, the benefits never result in a change in longevity.
In hindsight, Kachel says, the result isn't as surprising at it might seem. Natural selection is strongest early in life, and its influence on a trait wanes as an organism ages. Therefore, the benefits of grandmothering would have to be enormous to extend human lifespan.
Hawkes is taking the new paper in her stride. She thinks the model includes all the right life history and evolutionary factors, and can't wait to sink her teeth into the software code. But she has doubts about some of the assumptions Kachel's team made about the rates of mortality and fertility. "I bet a lot that if we modify some of those assumptions, we'll get a different result."
Frank Marlowe, an anthropologist at Florida State University in Tallahassee, thinks that the grandmother hypothesis is still the best in town, even if it isn't perfect. However, any successful theory of longevity must also account for males, he says, "even if males are only dragged along with females who were the target of selection".
Kachel is agnostic on the matter of what other forces might have lengthened humans' lives, but our longevity could be a cultural artefact due to the disappearance of traditional causes of mortality, such as predation.
"Grandmothers are important in many societies and they do have an important social role, but that doesn't prove they are the cause of longevity," she says.
© 2010 Nature Publishing Group
Grandmothering and Natural Selection