The repurposed social brain Carolyn Parkinson and Thalia Wheatley * Department ofPsychological andBrainSciences,DartmouthCollege, 6207MooreHall,Hanover,NH03755, USA Human social intel lig encedepends on adiversearray of perceptual , cognitive, andmoti vati onal capaciti es. Some of these capacities dependonneural syst ems that ma y have evolve d thr ou ghmodificationof ancestral systems wit hnon-soci al or more limit ed social func- tions(evolutionary repurposing). Socialintelli gence , in turn, enablesnew for ms of repurposing wit hin the life- timeof an indi vi dual (cultural andinstrumental repur- pos ing), whichentailinnovatingov er and exploiting pre-existingcircuitry tomeet problems ourbrainsdid not evolve tosol ve. Consider ing the se repurposi ng processes can provide insi ght into the comput ations tha t br ai nregions contri but etosoc ial inf ormati on processing, generate test able predicti ons that usef ully constrai n soci al neur osci ence theory, andreveal biol og- ical ly imposed constrai nts on cultural inventions and our ability torespondbeneficiall y tocontemporary challenges. Neural repurposing acrossandwithinlifetimes Evo lut ion use swhatever is in the room.Working asa tinkerer, it bor rowsand modi fiespre- exist ing structures [1] fornewuses.Forinstance, it is thought that feat hers original ly evol ved forthermoregulation, were co-opted forflight, and underwent subs equent adaptati ons to inc rease their uti lit y forflying[2]. Most complexbiologi- cal structures are thought toemergethr ough simila r cascadesofre-purposingand adapt ati on[2,3] .Through descentwit h modi ficati on(seeGlossary)[4], functional organization emergesas ‘apatchworkofmakeshifts piecedtogether... fromwhat was avai labl e whenoppor- tunityknocked’ [5].Neurosci ence has increasi ngly em- br aced the conce pt ofevo lut ionary reusein theori es ofbrainfunction[3,6–10] .Comparativelyle ss we ll devel- oped is anunderstandingofho w hu m an beha vi orand brainorgani za ti on are shaped by anal ogous pr ocesses thatoperatewi thin the li feti me of an indi vidua l, through whic h pre-exis tingneural architectureisredeployedto processcultural inventions [11] and toflexiblymanipu- lateotherminds.Wesuggesthere that thesephenomena can alsobe understoodasth e rep ur pos ing of limited ingredients toservenovel ends, and discu ss howthis perspective can help inform our understanding of human sociality. Neuralrepurposing acrosslifetimes EvolutionaryrepurposingHumansocialityissomewhatofamystery.Itisremark- ablecomparedwithmanyotherspeciesthathaveonlyaggressiveorreproductiveencounterswithnon-kin.From childhoodweprefercollaborationtoactingalone,with isolationcausingnegativehealthoutcomes[12,13]. Human socialnetworksarealsovastandunusuallyenduring[14], builtoftenmerelyoncommoninterestswithotherwise would-bestrangers.Whatunderliesthehumanpropensityto coo per ateandformlastingnon-reproductivetieswith non-kin? RepurposingrewardOnepossiblemechanismunderlyinghumansocialconnect- ednessisevolutionaryrepurposinginvolvingoxytocin pathways(i.e.,oxytocin,vasopressin,andtheirreceptors [15]).Thefunctionsofvasopressinincludepromotingpair- bonding,territoriality,andselectiveaggressiontoward Opinion Glossary Cultural inve ntio n: away of th i nk i ng or beh avi n g t hat is s har e d by gr oup members and transmitted across genera tions(e. g., lit era cy,arithmetic, rel igi on, tec hnolo gic al ski lls , music , art, soc ial str uct ur es) thr ough commu- nication or soci al lear ning. Cultural repurposing: theprocess by whi ch cul tur al inven tio nsareacquiredby co- opt ing pre -exist ing set s of brain circui ts (ne uronal ni che s) [11].Th e term ‘cul tural repu rposing’ is us ed here to refer to what Dehaene and Cohen have termed ‘neu rona l recy clin g’ [11] so as to more cle arl y dis tin guish thi s proces s from othe r kind s of repurpo sing (i.e ., evol utio nary and inst rume ntal repurpos - ing ) dis cussed in this articl e. Descent wit h modific ati on: Darwin’s sugg es tion that ne w morphologi cal st ru ct ures tend to be ge nera ted by borrowin g and modi fying pr e- existi ng stru cture s (i.e., thro ugh evolutio nary repu rpos ing/ tink erin g) [1,4]. Evolutionary repurposing : the pr oces s through which a st ructure wi th a pre- exi sti ng fun cti on is co- opt ed for a noveluse over the cou rse of evolu tio n, wit h or withou t cha nge s to its ori gin al for m. Instrumental repurposing : the creative and intention al evoca tio n of evolved resp onses in novel cont exts . Neur onal niche : a set of neur al ci rcui ts wi th suff icient ly related antecedent functi on al it y and ad equate pl as ti ci ty to supp or t the ac quisit ion of a gi ven cultural invention [11]. Oxytocin: a m am ma li an n eu r op ep ti de involv ed in modula ti ng s oc i al an d repr oduc ti ve beha vi or. Acro ss mammalian spec ies, ox yt ocin is involved in promotin g mate rnal beha vior (e.g ., lact atio n, moth er-i nfan t bond ing) ; in some mammal ia n spec ies, oxyt ocin supp or ts the format ion and maintenance of additi onal kin ds of selective soc ial bonds (e. g.,pair-b onds [15,16] ). Psyc holog ical dista nce: th e ext ent to wh ich an ob jec t isremo ved fr om t he curr ent, firs t-hand expe rience of the perc eive r. Obje cts can be psyc holo gica lly dis t ant in ter msofmu lt iple di m en s ions (e.g ., spac e, ti me , soci al ti es , hyp oth eti cal ity ). Psy cho log ica l dis tan ce and pro ximity promote att entio n to abst ract deco ntex tual ized ment al represen tati ons and to concrete situ atio nal details, respectively [26]. Vasopressin: a mammalia n neu rop ept ide similar in str ucture to oxyto cin tha t modulates social behavior by bindin g to vas opres sin 1a rec eptor s [16].Vaso- pre ssi n pro motes monog amy -re lat ed beh avi ors (e. g. , sel ect ive mat e pre fer - e nc e, p at er n al car e, ma te g ua r di ng [15]) in pa ir- bo nd in g s peci e s. Bo th vasopr ess in and oxy toc in are nec ess ary for the formatio n of selective social bon ds in mammals [15,16]. 1364-6613/2015 El sev ier Ltd. All ri ght s res er ved. http://dx.doi.org/10.1016/j.tics.2015.01.003 Corre spondi ng author : Wh ea tl e y, T. ([email protected]). Keywords: socia l intel ligen ce; cultu ral recyc ling; neura l reuse ; evolu tion . * Twitter: @ThaliaWheatley . Trend s in Cog ni ti ve Sci ences , March 201 5, Vol. 19, No. 3 133
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
[aRTPJ; see [28] for a recent parcellation of the tempor-
oparietal
junction
(TPJ)]
encodes
egocentric
distances
in
space,
time,
and
social
ties
similarly
(Figure 1). Social
and
temporal egocentric distances may have come to be
encoded
using
mechanisms
already
involved
in
egocentric
spatial
distance
encoding
partly
because
the
characteris-tics
of
aRTPJ
connectivity
and
structure
afford
encoding
information in self-centered coordinates [29], as well as
switching
between
internally
directed
attention
to
empha-
size decontextualized
representations,
and
externally
di-
rected attention to emphasize current contextual details
[27,30].
Directing attention inward. Different posterior parietal
cortex (PPC)
regions
[e.g.,
aRTPJ,
superior
parietal
lobule
(SPL), intraparietal sulcus (IPS)] may contribute distinctly
to evolutionarily recent cognitive functions owing to differ-
ences in the affordances of their connectivity and internal
structure.
For
instance,
whereas
the
aRTPJ
appears
to
serve
as
part
of
a
ventral
attention
network
that
switches
between internally and externally directed activities [30],
depending
in
part
on
the
relative
self-relevance,
or
psycho-
logical
distance,
of
the
information
at
hand
[27], SPL
circuitry originally evolved for allocating attention within
the
external
environment
may
have
been
repurposed
over
the
course
of
evolution
to
perform
similar
operations
with-
in internal space. Consistent with this possibility, the
human
SPL
supports
flexible,
top-down
attentional
alloca-
tion
within
external
space
and
within
internal
representa-
tions
[31–34].
Moreover,
cross-domain
multivariate
pattern classification analyses suggest that voluntary
shifts
of
attention
in
external
space
(e.g.,
when
performing
saccades)
and
in
internal
representations
(e.g.,
when
per-
forming
mental
arithmetic,
which
can
be
thought
of
as
a
shift of internal attention along a culturally learned num-
ber
line
[35]), involve
shared
encoding
mechanisms
in
the
SPL
[36]. Similarly,
when
greeting
an
old
friend,
common
SPL circuitry may support efforts to orient externally (e.g.,
to
her
voice
in
a
noisy
room)
as
well
as
internally
(e.g.,
to
relevant
personal
knowledge)
to
facilitate
conversation[31,33,37].
Thus, areas of human PPC involved in encoding and
operating
on
external
environmental
space
in
our
distant
mammalian
relatives
[38]
may
perform
similar
operations
on internal representations, including internal represen-
tations
of
relatively
abstract
contents
(e.g.,
social
and
temporal
distances).
Throughout
primate
evolution,
PPC
circuitry for encoding, tracking, and manipulating objects
in the
environment
may
have
been
repurposed
to
perform
similar operations on objects and abstract concepts in
working memory [7,8,31].
Mechanisms
of
evolutionary
repurposing
Theories
involving
evolutionary
repurposing
are
increas-
ingly widespread in cognitive neuroscience [3,6–10]. Even
so, the
mechanisms
underlying
purported
examples
of
neural
reuse
remain
underspecified
[9,10]
and
are
likely
heterogeneous. For instance, brain areas can take on new
functions
through
changes
in
patterns
of
local
postsynaptic
receptor
expression,
connections
with
other
regions,
or
duplication and divergence of subdivisions
[10,18,39,40].
Future
work
should
integrate
our
emerging
understanding
of
the
mechanisms
that
produce
phenotypic
diversity
in neocortex
[41,42]
with
theories
about
how
evolutionarily recent aspects of human sociality relate to
ancestral
systems.
Voxel
(A) (B)
(C)
1 2 3 4 5 6 7 8
Closer
Farther
Sooner
Voxel 1 2 3 4 5 6 7 8
Voxel
1
2
3
4
5
6
7
8
LaterVoxel
1
2
3
4
5
6
7
8
More
familiar
Less
familiar
Voxel 1 2 3 4 5 6 7 8
Voxel 1 2 3 4 5 6 7 8
Voxel 1 Voxel 1 Voxel 1
V o x e l 2
V o x e l 2
V o x e l 2
Closer
Key:
Farther
Key: Key:
Sooner Later More
familiar
Less
familiar
Spat
decodes
soc and temp
Soc
decodes
spat and
temp
Temp
decodes
soc and spat
All
TRENDS in Cognitive Sciences
Figure 1. Sharedmechanisms forencoding distance in space, time, and socialties.Multivoxel pattern analysis (MVPA) providesoneway to probe the information content
of particular brain regions. We recently applied MVPA to test for a common representation of spatial, temporal, and socialdistance from the self [27]. (A) Distributed local
response patterns were extracted throughout the brain corresponding to viewing increasing and decreasing spatial, temporal, and social distances for each fMRI run for
each participant (for clarity of visualization, eight- and two-voxel response patterns are depicted in (A) and (B), respectively). (B) A linear support vector machine learning
algorithm was trained to distinguish trials from one domain according to direction of distance change, then tested on each of the remaining two distance domains. Dots
represent two-voxel responsepatterns to examplesof each condition for a given participant.(C)
In the aRTPJ,it waspossible to decode relative egocentric distance acrossall distance domains. Convergent results were obtained using representational similarity analysis. These results are consistent with mechanisms for computing self-
relevance in terms of peripersonal space being repurposed to operate analogously on increasingly abstract contents as this region expanded over the course of human
evolution [7,8]. Abbreviations: aRTPJ, anterior right temporoparietal junction; Soc, social distance; Spat, spatial distance; Temp, temporal distance. Figure adapted and
reproduced from [27],
with permission from the Society for Neuroscience.
lifetimes is far rarer, relying, at least in part, on language
and
mentalizing.
The
distinctive
social-cognitive
capacities
of humans,
which
appear
to
set
us
apart
from
our
nearest
primate relatives – such as our capacities for social learn-
ing,
communication,
and
theory
of
mind
[43]
–
allow
usto create and transmit cultural inventions [43] and to
manipulate
one
another’s
minds
with
considerable
flexi-
bility.
We
describe
here
two
forms
of
repurposing
that
operate within the lifetime of an individual: the borrowing
or
modifying
of
pre-existing
brain
function
to
acquire
cultural inventions (cultural repurposing) and to exert
social
influence
(instrumental
repurposing).
Cultural repurposing
In the
same
way
as
evolutionarily
novel
ways
of thinking,
perceiving,
and
behaving
tend
to
emerge
through
borrow-
ing and modifying pre-existing structures [3,9], culturally
invented
ways
of
thinking,
perceiving,
and
behaving
must also borrow and modify pre-existing structures
[11]. However,
unlike
evolutionary
repurposing,
which
takes
place
through
genetic
changes
over
generations,
cultural
repurposing
must
occur
within
the
lifetime
of
an individual through neural plasticity [11]. This reuse
of
brain
systems
already
shaped
by
evolution
substantially
constrains
how
we
process
information
peculiar
to
today’s
world and the associated possibility space of our cultural
inventions.
That
is,
co-opting
circuitry
with
sufficiently related antecedent functionality and adequate plasticity
to
be
repurposed,
or
neuronal
niches,
necessarily
entails
inheriting
the
quirks
and
constraints
of the
pre-existing
circuitry [11]. Sociality contributes to both sides of this
equation.
First,
cultural
inventions
often
serve
social
pur-
poses: language facilitates long-term reciprocal exchanges
and
collaborations;
literacy
and
internet
use
extend
iden-
tity/reputation-tracking
beyond
the
limits
of
individual
cognition and provide access to others’ perspectives [44–
48].
Second,
because
living
in
complexly
bonded
groups
likely
drove
human
brain
evolution
[24], many
neuronal
niches likely have pre-existing social functions. For
example,
reading
co-opts
circuitry
for
face
and
object
rec-
ognition [49] (Box 2), and deducing why a computer fails to
Box
2.
How
learning
to
read
affects
the
neural
and
behavioral
signatures
of
face
perception.
Literacy likely comprises the best-studiedandmost dramatic example
of how acquiring a cultural invention can co-opt, and sometimes
reshape, an invaded neuronal niche. Reading recruits a common
neural substrate in the left fusiform gyrus (FG; the visual word form
area–VWFA) across cultures,
but emerged
too
recently for
this
regularity of functional brain organization to be attributable to natural
selection [11]. Compelling recent data suggest that literacy competes
with face and object processing for cortical real estate, dampening
and displacing face
processing
in the
left
fusiform gyrus,
and
promoting right-lateralization of face processing [49,80] (Figure I).
Consistent with literacy being constrained by pre-existing circuitry
for visual face andobject recognition, distributions of contours across
writing systems correspond closely with what this circuitry has been
tuned to detect in the natural world [11,81]. More specifically, line
junctions observable from most viewpoints (e.g., the points at which
an animal’s legs wouldmeet its torso; thepointswhere separate lines
meet in letters such as L or T) are especially important for both object
recognition and reading [82], and comprise the fundamental building
blocks of diverse wri ting systems [81]. In fact, the frequency
distributions of particular configurations of vertices are remarkably
similar across the world’s writing systems, and bear a striking
resemblance to the frequencydistribution of particular configurations
of vertices found in environmental images [81]. Thus, across cultures,
writing systems appear to be constrained by the pre-existing
competencies of their neuronal niche [11].
Interestingly, learning to read also changes aspects of face and
object processing, breaking mirror-image generalization and decreas-
ing holistic processing of faces and objects [83,84]. Mirror invariance,
present from infancy and useful for recognizing the same faces,
objects, and potential predators from different angles, is disadvanta-
geous for differentiating letters in many alphabets (e.g., discriminat-
ing ‘b’ from ‘d’), and is unlearned as we learn to read (Figure I).
Literates fare worse than illiterate adults at recognizing faces and
objects from their mirror images [83]. Thus, intensive engagement
with cultural inventions canalso reshape thepre-existing functions of
neuronal niches.
2.0
(A) (B) (C)
1.5
L e F G p a r a m e t e r e s m a t e s
1.0
0.5
Illiterates Literates
Faces
Key:
Leer
strings
Idencal images
of the same face
Mirror images
of the same face
Images of
different faces
Respond ‘same’
Respond ‘different’
L o g R T s f o r c o r r e
c t ‘ s a m e ’ r e s p o n s e s
7.00
6.75
6.50
6.25
6.00Illiterates Literates
Idencal images
of the same face
Key:
Mirror images
of the same face
TRENDS in Cognitive Sciences
Figure I.
Learning to read impacts brain systems involved in social perception. (A) Learning to read increases responses to letter strings and attenuates responses to
faces in the left FG (i.e., the VWFA), demonstrating that cultural inventions sometimes compete with representations corresponding to neuronal niches’ pre-existing
functions for cortical space [49] (in other cases, cultural inventions may more subtly extend or enhance the representational capacity of a niche [11]).
(B,C) When
instructed to judgeif face image pairs depict thesame individual (regardlessof orientation) or twodifferent individuals,literateadults areslower to recognizethe same
individual from mirror images than from identical images of his face, whereas illiterate adults are not [83].
Mirror-image generalization can be advantageous for
learning and survival because it facilitates the immediate recognition of a given individual, object, or predator from different viewpoints. However, learning to read
requires ‘unlearning’ mirror invariance. These effects are observed in individuals who learn to read as adults or as children [49,83],
suggesting that even relatively
modest experience with cultural inventions can influence the neural and behavioral signatures of neuronal niches [49]. Data and stimuli adapted from [49,83], with
permission from the American Association for the Advancement of Science and the American Psychological Association.
Opinion Trends in Cognitive Sciences March 2015, Vol. 19, No. 3
markets), with sometimes advantageous and sometimes
deleterious
consequences
[66,67]. Investigating
how
neuronal niches
shape,
andare reshaped
by, sociocultural
Rate(A)
(B)
(C)
Jier
Direcon
Step size
Dissonance/spikiness
Sad
SadHappy Happy
Angry Angry
Scared Scared
Peaceful Peaceful
TRENDS in Cognitive Sciences
Figure 2. Cultural repurposing of biological motion processing to create music. (A) Participants manipulated five slider bars corresponding to five dynamic features to
create either animations (bouncing ball) or musical clips (pianomelodies) thatexpresseddifferent emotions. (B) A clustering analysisof slider bar configurations revealeda
cross-modal, emotion-based structure such that analogous parameters were usednearly equivalently to express emotion in music andmovement. (C) These results were
replicated in a remote tribal village in Cambodia. Together, theseresults suggest that the expression ofmusic involvesco-opting circuitry for processing biological motion,
and that the musical forms that cultures use to convey emotional meaning are constrained by the neuronal niche that music co-opts. Figures adapted from [59].