SCIENCE sciencemag.org 16 JULY 2021 • VOL 373 ISSUE 6552 275
GRAPHIC: N. CARY/
carry over into subsequent generations, this
transgenerational legacy in social interac-
tions between similar genotypes (albeit
different individuals) from year to year has
implications for understanding processes
that coevolve with social behavior. Social
inheritance of association networks causes
a strong covariance between the many con-
sequences of social structure and inherited
genetic variation over generations. The
consequences of such social inheritance for
social transmission and the evo-
lution of social interactions are
A prominent example of so-
cial transmission relates to the
spread of pathogens and other
microbiota ( 1 , 6 ). When social
structure persists across gen-
erations through inheritance,
and susceptibility or immunity
is also inherited, it creates a
link between these and could
either increase vulnerability to
disease (through disease persist-
ing in susceptible groups across
generations) or decrease risk
(through immunity or reduced
susceptibility within clusters).
For example, imagine a patho-
gen whose transmission prob-
ability is determined by specific
gene-for-gene matching with
hosts; under this model of so-
cial inheritance, parents and
offspring are much more likely
to be exposed to the same spe-
cific lineages of microbes and
vice versa, and coevolutionary
interactions between host and
pathogen genomes may be ac-
celerated. Helpful bacteria are
also spread socially for many species, par-
ticularly skin and gut microbiota ( 7 ), for
which transmission occurs both vertically
(from parent to offspring) and horizontally
(between peers). As such, inheritance of so-
cial interactions across generations, in which
individuals inherit social associations from a
parent, should contribute to stabilizing the
composition of these communities through
promoting persistence of particular socially
spread microbiota and reducing mixing
In almost all social systems, diverse types
of information are also socially transmit-
ted between individuals. Information can
spread across social networks as individu-
als gain new information, adopt the behav-
ior, and then transmit this to others ( 6 , 8 ).
Unlike infectious diseases, however, these
“behavioral contagions” frequently depend
on much more than exposure to contagious
individuals. Individual decisions surround-
ing whether to adopt a socially informed
behavior may, for example, depend on
whether they conform with the majority of
the population, or which of their specific as-
sociates are performing the behavior, even in
relatively simple animal systems ( 9 ). These
“social learning strategies” ( 10 ) can cause
behaviors to spread more efficiently on heav-
ily clustered social networks rather than
diffusely connected networks that often fa-
cilitate disease transmission ( 8 ). Therefore,
just as the inheritance of social ties that
Ilany et al. demonstrate forces social net-
works into clustered structures that persist
across generations, this may also promote
the spread and establishment of socially in-
formed behaviors within groups compared
with systems without social inheritance.
Furthermore, social inheritance and the
resulting cross-generational clustered net-
works of this kind may support the develop-
ment of behavioral traditions or animal “cul-
tures,” which require the initial social spread
of behaviors and the maintenance of these
particular behaviors over time as individuals
learn these from their group members ( 11 )
both between and within generations.
There is a rich theoretical literature that
stresses the relevance of social network
structure for the evolution of cooperation
( 12 ). The type of social inheritance demon-
strated by Ilany et al. implies that interac-
tions between specific genotypes persist
across generations and hence greatly in-
crease the persistence of such interactions
as well as the chance for reciprocation of
cooperative actions. The transgenerational
carryover of social structure also implies
that competition will be occurring between
the same interacting genotypes more than by
chance, and that this presents the opportu-
nity for reduced competition through prior
familiarity—sometimes called the “dear en-
emy effect”—to occur across generations.
Although the study of Ilany et
al., and prior modeling ( 2 ), has
focused on social inheritance
through kin-structuring, an-
other open question is whether
the same process could operate
in non–kin-structured systems,
too. For example, any animal
(or plant) population for which
offspring show limited dispersal
might result in an increased ten-
dency for offspring to interact so-
cially with the same individuals
whom their parents interacted
with, or even with the offspring
of those individuals whom their
parents interacted with. As such,
the existence of social inheri-
tance provides a general poten-
tial for the social choices of the
parents to directly influence the
social setting of their offspring.
Future work should seek to ex-
amine how widely specific social
relationships are inherited in
range of population structures
and what implications this has
for the rate of evolution of the
many processes that depend on
social network structure. j
REFERENCES AND NOTES
- J. Krause, R. James, D. W. Franks, D. P. Croft, Animal
Social Networks (Oxford Univ. Press, 2015).
- A. Ilany, E. Akçay, Nat. Commun. b, 12084 (2016).
- A. Ilany, K. E. Holekamp, E. Akçay, Science 373 , 348
- J. A. Firth, B. C. Sheldon, L. J. N. Brent, P. Roy. Soc. B Biol.
Sci. 284 , 20171939 (2017).
- E. W. Wice, J. B. Saltz, Nat. Commun. 12 , 3357 (2021).
- A. Kucharski, Rules of Contagion: Why Things Spread—
And Why They Stop (Profile Books, 2020).
- A. Sarkar et al., Nat. Ecol. Evol. 4 , 1020 (2020).
- D. Centola, How Behaviour Spreads: The Science of
Complex Contagions (Princeton Univ. Press, 2018).
- J. A. Firth, Trends Ecol. Evol. 35 , 100 (2020).
- W. Hoppitt, K. N. Laland, Social Learning An Introduction
to Mechanisms, Methods, and Models. (Princeton Univ.
- L. M. Aplin et al., Nature 518 , 538 (2015).
- S. Gokcekus, E. F. Cole, B. C. Sheldon, J. A. Firth, Biol. Rev.
Camb. Philos. Soc. (2021). 10.111/brv.12757
This work was supported by the Biotechnology and Biological
Sciences Research Council (BB/S009752/1) and Natural
Environment Research Council (NE/S010335/1).
t 1 t 2 t 3
t 1 t 2 t 3
A network view of social inheritance
How social inheritance occurs
A focal mother ( ) holds social bonds ( ) of different strengths with her
associates (Stage t1). She produces an offspring ( ) who is socially bonded to
her ( ) and inherits the mother’s strongest social bonds ( ) (Stage t2).
The focal offspring maintains these strong bonds even after it loses its maternal
bond and gains new bonds (Stage t3).
How social inheritance influences social processes
An informed individual ( ) passes a new behavior ( ) to the focal mother ( )
and another ( ) (Stage t1). Newly informed individuals pass on this behavior
( ) (Stage t2). The focal offspring is exposed to the same information as its
mother and also adopts the behavior owing to social inheritance (Stage t3).
0716Perspectives.indd 275 7/9/21 5:20 PM