Published on:

The heritability of having many friends

My latest Mind and Matter column in the Wall
Street Journal:


Not only is the capacity for forming large social networks in
monkeys partly genetic, but some of the genes that affect this
ability may now be known. So suggests a new study of an isolated
population of free-living macaques on an island off Puerto

Lauren Brent of Duke University and her colleagues collected information on the social networks
of about 100 monkeys, measuring how much time they spent grooming
each other and how close together they sat. The researchers
recorded not only how many “friends” each had but how many “friends
of friends.” The latter measure, in both social animals and human
beings, has proved valuable in determining rank and well-being.

Because the family tree of this isolated population of monkeys
is well known, Dr. Brent and her colleagues could tease apart the
influences of nature and nurture. They found that, correcting for
things such as sex and age, a monkey’s “tendency to interact
frequently in an affiliative manner with partners who themselves
interact affiliatively at high rates with others” could be at least
partly predicted from its genetic parentage.

Moreover, based on which version the monkeys had of the genes
SLC6A4 and TPH2, the researchers were able to predict which
individuals spent a lot of time grooming partners who themselves
spend a lot of time grooming their own partners. The statistical
significance of this finding remains tentative, but it looks as if
having the rare version of both genes made monkeys less social,
while either mutation alone had no effect.

The mutations making monkeys less social have been tied to
anxiety and a tendency to avoid risks. This may explain why they
persist. Although well-connected monkeys generally have more
offspring, anxious monkeys may be more vigilant to threats.

It’s a measure of the way the intellectual world has evolved
that 50 years ago such a study would have shocked the scientific
world and brought down a storm of protest.

When the great California Institute of Technology geneticist
Seymour Benzer set out in the mid-1960s to find mutations in
fruit flies that affected behavior, rather than mere anatomy, he
was ridiculed for challenging the consensus that all behavior must
be learned.

Benzer told the geneticist Max Delbrück about the plan to find
behavioral mutants; Delbrück said it was impossible. To which
Benzer replied: “But, Max, we found the gene, we’ve already done
it!” (Benzer’s mother was more succinct: “From this, you can make a
living?”) He was soon able to identify mutations related to
hyperexcitability, learning, homosexuality and unusual circadian
rhythms, like his own: Benzer was almost wholly nocturnal.

Since then, thanks to studies of human twins and a rash of
genetic investigations in animals, it has become routinely accepted
that most things, including personality, sexual orientation and
intelligence, are to some degree affected by genes. The University
of Virginia’s Eric Turkheimer has declared what he calls the “first
law of behavior genetics”: that all human behavioral traits are

In 2009 a study of human twins by James Fowler of the
University of California at San Diego and colleagues found evidence
of genetic heritability of social networks similar to that now
reported in monkeys. Frankly, there’s nothing surprising here. We
all know people who are intrinsically less sociable than others,
because of their personalities not their circumstances.

As always in such nature-nurture debates, the red herring to
avoid is “determinism.” A discovery that genes affect behavior is
no more or less deterministic than a discovery that family or
education does so. Whether you are antisocial because your mother
was unemotional-a fashionable theory in the 1960s-or because of a
mutation tells you nothing about whether your condition can be
remedied by some intervention.

By Matt Ridley | Tagged:  rational-optimist  wall-street-journal