Social
competence:
an
evolutionary
approach
Barbara
Taborsky
1
and
Rui
F.
Oliveira
2,3
1
Behavioural
Ecology,
Institute
of
Ecology
and
Evolution,
University
of
Bern,
Wohlenstrasse
50A,
CH-3032
Hinterkappelen,
Switzerland
2
Instituto
Superior
de
Psicologia
Aplicada,
Unidade
de
Investigac¸a
˜
o
em
Eco-Etologia,
Rua
Jardim
do
Tabaco
34,
1149-041
Lisboa,
Portugal
3
Champalimaud
Neuroscience
Programme,
Instituto
Gulbenkian
de
Cieˆ
ncia,
Rua
da
Quinta
Grande
6,
2780-156
Oeiras,
Portugal
‘Social
competence’
refers
to
the
ability
of
an
individual
to
optimise
its
social
behaviour
depending
on
available
social
information.
Although
such
ability
will
enhance
social
interactions
and
thus
raise
Darwinian
fitness,
its
evolutionary
and
ecological
significance
has
been
largely
ignored.
Social
competence
is
based
on
behavioural
flexibility.
We
propose
that
the
study
of
social
compe-
tence
requires
an
integrative
approach
that
aims
to
understand
how
the
brain
translates
social
information
into
flexible
behavioural
responses,
how
flexibility
might
be
constrained
by
the
developmental
history
of
an
indi-
vidual
or
by
trade-offs
with
other
(ecological)
compe-
tences,
and
how
social
plasticity
feeds
back
on
fitness.
Finally
we
propose
a
hypothesis
of
how
social
compe-
tence
can
become
a
driver
of
social
evolution.
Behavioural
flexibility
as
phenotypic
plasticity
Adaptation
to
the
environment
is
a
universal
characteris-
tic
of
living
systems.
According
to
classic
evolutionary
theory,
adaptation
by
natural
selection
relies
on
heritable
phenotypic
variation
produced
by
genetic
variation.
How-
ever,
when
the
rate
of
genetic
evolutionary
change
is
out-
paced
by
changes
in
the
environment
the
need
for
adaptive
change
without
genetic
mutation
emerges
[1].
In
this
scenario,
the
evolution
of
phenotypic
plasticity
is
favoured,
that
is,
a
certain
genotype
produces
different
phenotypes
depending
on
environmental
conditions
[2].
Among
ani-
mals,
behavioural
traits
exhibit
a
greater
plasticity
than
morphological
and
physiological
traits
and
plastic
changes
are
reversible
within
an
individual’s
lifetime
(‘behavioural
flexibility’).
This
makes
behavioural
flexibility
a
powerful,
immediate
mechanism
allowing
organisms
to
adapt
to
changing
environmental
conditions,
which
may
or
may
not
be
followed
by
other
flexible
adjustments
of
physiology
or
morphology.
Many
of
these
responses
are
simple
reflexes
and
fixed
action
patterns
elicited
by
a
stimulus
in
the
environment,
when
it
deterministically
predicts
an
appro-
priate
response.
However,
when
environmental
complexity
and
variability
increase,
the
capacity
to
adaptively
modify
behaviour
as
a
function
of
experience
and
context
is
need-
ed.
Although
some
degree
of
context-dependent
beha-
vioural
flexibility
may
be
achieved
with
genetically
determined
rules,
behavioural
flexibility
will
often
depend
on
cognitive
abilities
(understood
as
the
acquisition,
reten-
tion,
and
use
of
information;
[3])
that
allow
individuals
to
adapt
behavioural
output
to
specific
situations
in
a
com-
plex
and
variable
world
(e.g.,
see
[4]).
Interestingly,
the
evolutionary
study
of
behavioural
flexibility
has
rarely
been
framed
within
the
scope
of
phenotypic
plasticity,
but
rather
in
terms
of
cognitive
evolution
and
ecology
[3,5].
This
is
most
probably
due
to
the
fact
that
in
contrast
to
morphological
and
life
history
traits
(which
have
been
the
main
focus
of
phenotypic
plasticity
studies,
and
whose
plasticity
results
from
pro-
cesses
during
development
and
is
usually
non-reversible)
behavioural
flexibility
involves
rapid
changes,
is
labile,
and
is
present
during
the
whole
life
of
the
animal
[6,7].
Opinion
Glossary
Behavioural
reaction
norm
(BRN):
the
set
of
behavioural
phenotypes
that
a
single
individual
produces
in
a
given
set
of
environments
[8].
This
is
in
contrast
to
‘reaction
norms’
describing
typically
irreversible,
developmental
plasticity
of
a
single
genotype.
Thus
BRNs
describe
fast
responses
(within
a
life
time)
by
an
individual
to
variation
in
the
environment.
Developmental
plasticity:
variation
in
the
traits
of
individuals
that
results
from
processes
during
development
as
a
consequence
of
environmental
variation
and
that
is
typically
irreversible
[6].
Epigenetic
modifications:
changes
in
gene
function
that
do
not
involve
a
change
in
the
coding-sequence.
Examples
of
such
changes
are
DNA
methylation
and
histone
acetylation,
both
of
which
modify
genome-regional
gene
expression
without
changing
the
sequence
of
the
affected
genes.
Immediate
early
genes
(IEGs):
genes
that
show
rapid
and
transient
expression
in
response
to
a
wide
variety
of
extra-cellular
stimuli
and
in
the
absence
of
de
novo
protein
synthesis.
Their
products
act
as
transcription
factors
on
late
response
genes
or
as
effectors
that
change
the
functioning
of
the
cell,
and
thus
they
represent
the
first
genomic
response
to
stimuli
[60].
Neural
plasticity:
structural
and
functional
changes
of
the
nervous
system
as
a
result
of
input
from
the
environment.
Two
major
mechanisms
of
neural
plasticity
operate
at
different
time
scales:
structural
rewiring
of
neural
circuits
is
slow
and
long-lasting
and
induces
dramatic
behavioural
changes,
whereas
biochemical
modulation
of
existing
neural
networks
is
postulated
to
mediate
fast
and
transient
changes
between
motivational
states
that
promote
gradual
changes
in
behavioural
expression.
Neurogenomic
state:
transcriptome
profile
of
the
brain
areas
of
interest
for
a
given
behavioural
state
(e.g.,
expression
of
a
given
social
phenotype).
Phenotypic
(behavioural)
flexibility:
changes
of
the
(behavioural)
phenotype
that
can
be
reversed
within
an
individual’s
life
time
[6].
Phenotypic
(behavioural)
plasticity:
an
umbrella
term
subsuming
different
classes
of
plasticity,
including
developmental
plasticity,
phenotypic
flexibility,
and
life-cycle
staging
[6].
Social
competence:
the
ability
of
an
animal
to
optimise
the
expression
of
its
social
behaviour
as
a
function
of
the
available
social
information.
Social
context:
any
social
stimulus
that
can
vary
across
a
gradient
(e.g.,
group
composition,
offspring
begging,
or
competitor
density)
[8].
Social
information:
any
information
that
is
generated
by
the
behaviour
of
another
organism.
Corresponding
author:
Taborsky,
B.
(barbara.taborsky@iee.unbe.ch).
0169-5347/$
–
see
front
matter
ß
2012
Elsevier
Ltd.
All
rights
reserved.
http://dx.doi.org/10.1016/j.tree.2012.09.003
Trends
in
Ecology
and
Evolution,
December
2012,
Vol.
27,
No.
12
679
However,
a
phenotypic
plasticity
approach
to
behavioural
flexibility
would
provide
a
new
and
powerful
framework
to
understand
the
adaptive
nature
and
evolution
of
animal
behaviour;
namely
by
introducing
the
concept
of
‘beha-
vioural
reaction
norms’
(BRN;
see
Glossary)
as
a
tool
to
visualise
and
quantitatively
analyse
individual
plasticity,
and
to
unravel
individual
by
environment
(I
E)
interac-
tions
underlying
it
[8].
For
example,
recently
this
approach
facilitated
the
establishment
of
a
single
framework
to
integrate
the
study
of
animal
personality
and
behavioural
plasticity,
two
phenomena
usually
studied
separately,
thereby
enhancing
the
understanding
of
their
adaptive
significance
[8].
This
approach
will
also
help
to
unravel
the
proximate
mechanisms
of
behavioural
flexibility,
by
extending
the
framework
used
to
study
the
causes
of
phenotypic
plasticity
to
behaviour
(e.g.,
[2,9]).
Adaptive
behavioural
flexibility
in
the
social
domain
The
social
domain
is
arguably
the
most
complex
and
fluctuating
component
of
an
animal’s
environment
as
it
involves
interaction
with
other
behavioural
agents
with
inherently
associated
higher
levels
of
unpredictability.
An
animal
interacting
with
its
non-social
abiotic
or
biotic
environment
will
often
modify
this
environment,
thereby
creating
ecological
feedback
on
the
individual
itself
forcing
it
to
flexibly
adjust
its
behaviour
(e.g.,
a
foraging
individual
changes
the
local
resource
density
enforcing
an
adjustment
of
subsequent
foraging
decisions).
Such
ecological
feedback
is
ubiquitous
and
can
even
span
generations
(e.g.,
[10]),
but
nevertheless
the
degrees
of
freedom
of
this
feedback
will
be
finite.
In
contrast,
feedback
in
the
social
domain
can
have
nearly
infinite
degrees
of
freedom,
as
feedback
will
depend
not
only
on
social
context
but
also
on
the
number,
identity,
and
the
external
(e.g.,
rank
and
size)
and
internal
states
of
the
agents
that
social
behaviour
is
directed
towards.
Con-
sequently,
dealing
with
social
complexity
requires
the
evolution
of
cognitive
mechanisms
that
allow
the
individ-
ual
to
assess
the
internal
(‘emotional’)
state
of
other
organ-
isms
and
the
social
context,
and
to
integrate
and
process
these
stimuli
not
just
as
a
result
of
direct
effects
of
percep-
tual
information,
but
rather
as
a
function
of
what
that
perceptual
information
means
to
the
individual
at
that
moment
in
time
[11].
Therefore,
social
decision-making
depends
on
some
kind
of
social
experiential
knowledge
that
allows
organisms
to
evaluate
stimuli
and
to
determine
the
appropriate
behaviour.
Thus,
more
than
in
any
other
behavioural
domain
it
is
expected
that
social
behaviour
should
exhibit
high
levels
of
plasticity.
The
ability
of
individuals
to
regulate
the
expression
of
their
social
beh aviour
in
order
to
optimise
their
social
relationships
is
referred
to
as
‘social
competence’.
It
should
be
stressed
here
that
flexibility
in
the
expression
of
social
beh aviour
is
necessary
but
not
sufficient
for
social
competence,
since
the
latter
implies
not
only
variation
in
the
response
to
the
same
social
stimuli
depending
on
additional
social
information,
but
also
the
demonstration
that
this
variation
is
adaptive
(i.e.,
optimises
the
response).
Social
competence
allows
organisms
to
express
appropriate
responses
to
demands
and
to
generate
and
capitalise
on
opportunities
in
the
social
environment
[12]
thereby
ultimately
enhancing
their
fitness.
For
example,
social
competence
will
enable
individuals
to
avoid
engag-
ing
in
costly
social
interactions
or
being
ejected
from
their
social
groups.
Social
competence
has
been
shown
to
influence
the
performance
of
different
social
behaviours
across
different
social
contexts
(Box
1).
There
are
two
possible
scenarios
that
can
explain
this
joint
effect.
Either
social
competence
is
composed
of
a
collection
of
social
skills,
each
of
which
evolves
rather
independently,
or,
alternatively,
social
com-
petence
may
be
an
ability
underlying
all
social
sub-
domains
(e.g.,
interactions
with
competitors,
mating,
and
social
foraging)
resulting
in
positive
within-individual
cor-
relations
of
performance
across
different
social
contexts
(Figure
1).
Although
some
evidence
supports
the
former
hypothesis
(e.g.,
independence
of
social
skills
that
contrib-
ute
to
mating
success
in
cowbirds,
such
as
song
quality,
courtship
persistence,
and
competitive
ability
[13])
most
of
the
available
evidence
supports
the
second
alternative,
namely
the
fact
that
the
social
environment
experienced
during
ontogeny
can
affect
the
appropriateness
of
suites
of
social
behaviours
belonging
to
various
different
social
contexts
[14–17].
Animal
social
competence
To
date
the
study
of
social
competence
has
been
mainly
a
domain
of
the
social
sciences,
with
a
strong
focus
on
causal
relationships
between
social
factors
and
the
development
and
expression
of
social
competence
in
humans
[12,18,19].
However,
indications
of
social
competence
are
also
well
known
from
non-human
animals.
Interacting
animals
respond
to
the
presence
of
bystanders
(‘audience
effect’)
by
changing
their
signalling
behaviour
according
to
the
type
of
audience
and
social
context
(e.g.,
[20,21]).
By
contrast,
bystanders
extract
information
from
observed
interactions
(‘eavesdropping’)
that
they
use
in
subsequent
interactions
Evolving neurogenomic machinery
underlying social plascity
Social competence
Selecon
Territory
defence
Mang Social
foraging
Fitness
TRENDS in Ecology & Evolution
Figure
1.
Social
competence
is
hypothesised
to
be
an
ability
underlying
all
social
behaviour
that
is
contingent
on
social
information
and
thus
to
affect
behaviour
in
many
or
all
social
contexts.
This
should
give
rise
to
positive
within-individual
correlations
of
social
performance
across
different
social
contexts;
social
performance
is
under
positive
selection
and
affects
fitness.
Differently
coloured
rectangles
represent
different
individuals;
the
width
of
these
rectangles
represents
an
individual’s
social
performance
in
the
different
social
contexts.
Opinion
Trends
in
Ecology
and
Evolution
December
2012,
Vol.
27,
No.
12
680
Box
1.
How
to
measure
social
competence
If
we
want
to
study
individual
variation
in
social
competence,
we
need
an
approach
that
captures
the
universal
nature
of
this
trait.
We
propose
to
decide
upon
a
representative,
ecologically
relevant
set
of
social
situations,
of
which
we
understand
the
adequacy
of
the
possible
involved
behaviours.
We
should
expose
individuals
to
these
situations
in
standardised
trials
testing
them:
(i)
in
different
social
contexts
(e.g.,
dominance
relationships,
mating,
and
brood
care)
to
test
for
the
presence
of
a
general
social
ability;
(ii)
in
different
social
roles
(e.g.,
dominant
and
subordinate)
within
the
same
context
to
test
if
this
ability
is
based
on
plasticity;
(iii)
in
an
unknown
social
situation
(e.g.,
an
ecologically
relevant
situation
that
had
not
yet
been
encountered
during
ontogeny)
to
check
for
the
ability
to
generalise
across
social
situations;
(iv)
at
different
life
stages
to
test
if
individual
differences
in
social
competence
persist
over
time.
It
is
important
to
have
a
priori
predictions
about
which
social
behaviours
will
be
optimal
in
a
given
test.
Classifying
those
behaviours
as
socially
competent
that
are
expressed
by
individuals
with
the
highest
Darwinian
fitness
may
be
misleading,
because
a
high
fitness
may
result
from
competences
outside
the
social
domain
(e.g.,
solitary
foraging
ability).
We
illustrate
this
research
agenda
by
an
example
from
the
cooperatively
breeding
cichlid
Neolamprologus
pulcher
(Figure
I).
These
fish
had
been
reared
either
with
or
without
the
presence
of
older
conspecifics
to
investigate
whether
experiencing
a
more
complex
social
environment
results
in
better
social
competence
[16,17].
Three
challenge
tests
(T1–T3)
were
conducted
to
assess
social
competence
along
the
four
above
mentioned
axes
(Table
I).
In
T1
juveniles
were
assigned
either
the
ownership
of
a
critical
resource
(a
shelter)
or
of
an
intruder
(asymmetric
competition).
In
nature,
subordinate
group
members
defend
private
shelters
within
the
breeders’
territory
against
other
group
members.
Shelter
access
can
decide
over
survival
as
predation
pressure
is
intense.
The
intruder
role
reflects
the
situation
of
a
juvenile
subordinate
in
search
of
an
own
private
shelter
after
it
is
no
longer
allowed
to
access
the
natal
breeding
cavity.
T2
simulated
a
symmetric
contest
over
a
shelter,
in
which
both
opponents
had
been
assigned
the
role
of
the
shelter
owner
prior
to
the
test.
This
mimics
a
situation
in
which
a
shelter
has
to
be
defended
against
an
at
least
equally
motivated
opponent
(symmetric
competition).
In
T3
larger,
adult
subordinates
were
forced
to
achieve
acceptance
at
the
territory
of
an
unfamiliar
breeder
pair,
a
situation
fish
would
encounter
after
dispersal
from
the
natal
territory
[74].
In
this
situation
subordinates
should
strive
for
acceptance
as
brood
care
helpers.
(a) (b)
0
0.4
0.8
1.2
1.6
*
*
T1: Asymmetric compeon
*
*
*
T2: Symmetric compeon
T3: Integraon in social group
Shelter owner
Intruder
Winner
Loser
Subordinate
20
20
25
10
10
15
15
0
0
2
4
6
8
5
+F -F
+F -F
+F -F
Threat display
Submission
Open aggression
Submission
Submission at breeders' shelter
+F -F
+F -F
0
5
TRENDS in Ecology & Evolution
Figure
I.
(a)
Neolamprologus
pulcher
that
had
been
reared
together
with
older
conspecifics
showed
more
of
the
expected
appropriate
behaviours
in
an
asymmetric
competition
test
(blue),
a
symmetric
competition
test
(red),
and
a
previously
unknown
social
challenge,
namely
being
exposed
to
an
unfamiliar
breeder
pair
(green).
(b)
Neolamprologus
pulcher
lives
and
breeds
in
social
groups
of
3–38
individuals
that
are
structured
by
a
size-dependent
dominance
hierarchy.
Related
and
unrelated
smaller
individuals
help
the
dominant
breeder
pair
rearing
their
offspring.
Table
I.
Three
behavioural
tests
to
assess
the
differences
in
social
competence
between
N.
pulcher
reared
with
or
without
older
conspecifics
(parents
and/or
helpers).
In
all
tests,
fish
reared
with
older
conspecifics
showed
significantly
more
of
the
behaviours
predicted
to
be
appropriate
in
a
given
situation
(Figure
I),
whereas
no
significant
differences
were
found
in
behaviours
predicted
to
be
inappropriate
[16,17]
Challenge
test
(i)
Social
context
(ii)
Different
roles
within
a
context
(iii)
Is
the
situation
known?
(iv)
Life
stage
Predicted
appropriate
behaviour
Predicted
inappropriate
behaviour
T1
Competition
Owner
Yes
Early
juvenile
Threat
display
Open
aggression,
submission
Intruder
Submission
Escape
T2
Competition
Winner
Yes
Late
juvenile
Open
aggression
Threat
display,
submission
Loser
Submission
Escape
T3
Integration
by
subordinate
in
a
new
group
-
No
Adult
Submission
near
prospective
breeding
cavity
Aggression,
escape
Opinion
Trends
in
Ecology
and
Evolution
December
2012,
Vol.
27,
No.
12
681
with
the
observed
individuals
[22,23].
Familiarity
with
an
opponent
is
also
known
to
reduce
(‘dear
enemy
effect’
[24])
or
increase
(‘nasty
neighbour
effect’
[25])
fighting
beh av-
iour
depending
on
the
relative
threat
imposed
to
territory
holders
by
neighbours
versus
strangers.
Previous
social
experience
can
affect
subsequent
behaviour
as
in
the
case
of
‘winner–loser
effects’
demonstrated
across
different
animal
taxa
[26,27],
where
previous
winners
are
more
likely
to
win
successive
contests,
and
losers
will
more
likely
lose
again
even
against
different
opponents.
A
simi-
lar
effect
is
known
from
a
cooperative
context
where
individuals
give
more
help
to
social
partners
if
they
had
received
help
by
another
conspecific
before
(‘generalised
reciprocity’
[28]).
Experim ents
have
shown
that
Norway
rats
optimise
their
social
beh aviour
as
a
function
of
social
information
obtained
in
previous
fights,
as
acting
contin-
gently
on
this
information
increased
an
actor’s
fitness.
Previous
winners
won
successive
fights
after
a
shorter
time
despite
a
reduced
amount
of
aggressive
behaviour,
and
losers
faced
a
decreased
risk
of
injury
despite
a
reduced
amount
of
submissive
behaviour
shown
[29].
Most
of
the
abovementioned
effects
of
social
information
on
behaviour
have
been
observed
to
be
widespread
in
natural
populations
[24,25,27,30]
and
their
existence
was
confirmed
by
targeted
field
experiments
[21,24–27].
Thus
we
propose
that
social
competence
allows
animals
to
efficiently
navigate
the
complexities
of
their
social
envi-
ronment
in
order
to
survive,
reproduce,
and
raise
their
offspring,
and
therefore
it
should
be
seen
as
a
key
deter-
minant
of
the
Darwinian
fitness
of
individuals.
Social
competence
versus
general
cognition
Since
social
competence
can
give
rise
to
consistent
expres-
sion
of
appropriate
flexible
beh avioural
responses
across
different
social
contexts,
it
appears
similar
to
the
concept
of
general
intelligence.
In
humans,
performance
on
di-
verse
cognitive
tasks
shows
robust
positive
correlations,
so
that
individuals
scoring
high
on
one
cognitive
ability
are
also
lik ely
to
score
well
in
others.
This
positive
manifold
of
correlations
has
been
interpreted
as
being
caused
by
a
latent
single
factor
named
general
intelligence,
or
simply
‘g’
[31].
In
comparative
cognition
it
has
been
debated
whether
cognition
is
organised
into
separate
modules
or
whether
there
is
a
general
problem-solving
ability
that
pertains
all
behavioural
domains
and
ecological
demands.
Although
comparative
analyses,
mainly
focused
on
pri-
mates
and
birds,
suggest
the
lack
of
cognitive
modules
and
support
a
general
intelligence
scenario
[32–35],
experi-
mental
approaches
in
single
species
produce
conflicting
results.
Different
developmental
studies
show
that
early
social
experience
can
promote
social
competence
and
so-
cial
learning,
but
leaves
the
performance
in
non-social
cognitive
tasks
unaffected.
For
example,
after
maternal
deprivation,
laboratory
rats
performed
more
poorly
than
normally-reared
rats
in
three
social
learning
tasks,
where-
as
in
two
spatial
learning
tasks
the
treatment
groups
performed
equally
well
[36].
Similarly,
communally-
reared
mice
had
a
better
social
competence
during
their
adult
life
than
single-mother
reared
mice,
but
they
did
not
differ
in
spatial
learning
and
memory
capacity
tests
[37,38].
Zebrafish
reared
among
more
diverse
social
partners
(mixed
strains
versus
single
strain)
were
more
aggressive
later
in
life,
but
non-social
behaviours
such
as
general
activity,
predator
evasion
behaviour
and
stress
recovery
were
unaffected
[39].
Finally,
in
humans
intelli-
gence
quotients
(IQ)
and
social
intelligence
tests
for
assessment
of
the
theory
of
mind
were
not
correlated
[40].
Part
of
these
contradictory
results
may
be
explained
by
the
interpretation
of
comparative
studies
that
have
identified
a
species-level
composite
general
intelligence
factor.
As
for
human
intelligence,
it
is
important
to
distin-
guish
between
‘g’
as
a
general
first-order
factor
or
dominant
eigenvalue
in
a
factor
analysis,
from
‘g’
as
a
psychological
construct
that
reduces
cognitive
abilities
to
a
single
quanti-
tative
cognitive
or
biological
process
[41].
Alternative
models
have
been
proposed
that
illustrate
how
a
psycho-
metric
‘g’
can
be
explained
by
dynamic
interactions
between
cognitive
processes
during
development,
rather
than
by
a
single
underlying
process
[41].
Together,
these
results
suggest
that
it
is
unlikely
that
social
competence
is
a
sub-
domain
of
general
intelligence,
but
further
studies
are
needed
to
clarify
this
issue.
An
evolutionary
framework
for
the
study
of
social
competence
If
we
understand
social
competence
as
a
general
ability
affecting
individual
performance
in
a
social
environment,
it
should
have
the
key
properties
of
an
ecological
perfor-
mance
trait
[42],
namely
(i)
the
existence
of
inter-individ-
ual
variation
in
performance
ranging
from
low
to
high,
and
(ii)
this
variation
should
be
positively
correlated
with
fitness
[43].
Furthermore,
for
it
to
be
an
evolvable
trait,
potential
evolutionary
costs,
benefits,
and
trade-offs
of
social
competence
should
also
be
identified.
(i)
Variation
in
performance
Persistent
individual
variation
in
social
performance
can
be
due
to
genetic
disposition
or
it
can
be
induced
during
ontogeny.
Genetic
disposition
has
so
far
been
addressed
only
in
humans:
twin
studies
have
detected
significant
heritability
of
social
competence
as
assessed
by
questionnaires
([44,45]
and
references
therein).
A
much
larger
body
of
literature
has
shown
that
variation
in
social
competence
can
be
environ-
mentally
induced
and
originates
from
developmental
plasticity
(Table
1).
However,
only
a
few
of
these
studies
tested
performance
across
different
contexts.
For
example,
communally-reared
mice
exposed
to
more
social
contacts
showed
more
appropriate
social
behaviours
in
the
contexts
of
dominance
interactions
and
brood
care,
compared
to
single-mother
reared
mice
[15].
Also,
cichlid
fish
(Neolamprologus
pulcher)
reared
by
alloparents
behaved
more
appropriately
in
a
competitive
context
or
when
striving
for
the
acceptance
as
helper
by
a
breeder
pair,
than
individuals
reared
among
siblings
only
(Box
1,
[16,17]).
(ii)
Correlation
with
fitness
Several
studies,
which
had
induced
differences
in
social
performance
by
manipulations
of
social
cues
during
development,
found
associated
differences
in
fitness
correlates
(Table
1).
For
example,
in
some
rodents,
being
exposed
to
a
more
complex
social
Opinion
Trends
in
Ecology
and
Evolution
December
2012,
Vol.
27,
No.
12
682
Table 1. Summary of studies in which (i) the social experience during ontogeny was manipulated, (ii) challenge tests were conducted after the manipulations to test for induced
differences in social performance, and (iii) effects of differential social behaviour on fitness correlates were reported
Species Rearing environment Life stage of
experience
Social context of
challenge test
Phenotypic effect
a
Effects on fitness
a
Refs
Neolamprologus pulcher
Parents and siblings vs
siblings-only
Dependent young Competition More appropriate use of aggression
and submission
Shorter fights [16]
Competition More appropriate use of aggression
and submission
Losers less often evicted [17]
Integration in
social group
More submissive behaviour at
breeding cavities
Higher tolerance by breeders [17]
Zebra finch, Taeniopygia
guttata
With vs without males
present in breeding colony
Dependent young Mate choice, mating Prefer opposite sex over same sex Higher mating success [79]
Cowbirds, Molothrus ater
Dynamic vs static flock Adult males Mate choice, mating Larger and more variable ‘singing network’ Higher mating success [13]
Lab mouse, CD1(ICR)
Communal vs single-mother
rearing
Dependent young Dominance
interactions
Higher propensity to interact socially Shorter time to find hierarchy
position
[14]
Lab mouse, CD1(ICR)
Synchronous vs asynchronous
birth-spacing in communal nest
Dependent young Dominance
interactions
Higher aggression/lower affiliative behaviour Shorter time to find hierarchy
position
[80]
Lab mouse, CD1(ICR)
Mixed sex vs female only litter Dependent young Maternal care Shorter pup retrieval latency,
higher aggression towards intruder males
Better maternal care [81]
Lab mouse, Balb/c
Communal vs single-mother
rearing
Dependent young Maternal care Shorter pup retrieval latency, more care,
better nest quality
Better maternal care, F2 had
larger litters
[15]
Dominance
interactions
Lower aggression towards intruder males
Lab rat, Porton
Mixed sex vs female only litter Dependent young Maternal care More elaborate nests and more building
behaviour
Fewer and smaller litters produced [82]
Lab rat, Sprague–Dawley
With vs without mother present Dependent young 3 social learning tests Enhanced social learning ability [36]
Short separation
from pups
Shorter latency to show maternal care Better maternal care
Oldfield mouse,
Peromyscus polionotus
With vs without
younger-sibling pups
Subadult daughters Mating, reproduction Better nest quality Higher reproductive success [46]
Rhesus macaque,
Macaca mulatta
With vs without mother present Dependent young Staging intra-group
conflict
Better in acquisition of dominance rank Higher rank achieved [83]
a
Directions of the reported results refer to the respective treatment group that was exposed to the more complex social condition, and that is the first condition mentioned in the column ‘Rearing environment’.
Opinion
Trends
in
Ecology
and
Evolution
December
2012,
Vol.
27,
No.
12
683
