Human infancy and parenting in global

rspb.royalsocietypublishing.org perspective: specificity

Marc H. Bornstein, Diane L. Putnick, Yoonjung Park, Joan T. D. Suwalsky
and O. Maurice Haynes
Research Child and Family Research, Eunice Kennedy Shriver National Institute of Child Health and Human Development,
6555 Rock Spring Drive, Suite 220, Bethesda, MD 20817, USA
Cite this article: Bornstein MH, Putnick DL, MHB, 0000-0002-6810-8427; DLP, 0000-0002-6323-749X
Park Y, Suwalsky JTD, Haynes OM. 2017
Human infancy and parenting in global We address three long-standing fundamental questions about early human
perspective: specificity. Proc. R. Soc. B 284: development and parental caregiving within a specificity framework using
20172168. data from 796 infant–mother dyads from 11 societies worldwide. Adopting
Downloaded from https://royalsocietypublishing.org/ on 18 November 2023

a cross-society view opens a vista on universal biological origins of, and con-
http://dx.doi.org/10.1098/rspb.2017.2168
textual influences on, infant behaviours and parenting practices. We asked:
how do infant behaviours and parenting practices vary across societies?
How do infant behaviours relate to other infant behaviours, and how do
parent practices relate to other parent practices? Are infant behaviours and
Received: 27 September 2017
parent practices related to one another? Behaviours of firstborn five-month
Accepted: 10 November 2017 infants and parenting practices of their mothers were microanalysed from
videorecords of extensive naturally occurring interactions in the home.
In accord with behavioural specificity, biological expectations and cultural
influences, we find that infants and mothers from diverse societies exhibit
Subject Category: mean-level society differences in their behaviours and practices; domains of
infant behaviours generally do not cohere, nor do domains of maternal prac-
Behaviour
tices; and only specific infant behaviours and mother practices correspond.
Few relations were moderated by society.
Subject Areas:
behaviour

Keywords:
mother – infant interaction, cross-society,
1. Introduction
infancy, parenting (a) Cross-society1 developmental science
Human infancy, infant caregiving, and infant–mother interactions represent our
earliest beginnings, and can have far-reaching effects across development. In con-
sequence, they are topics of perennial and compelling interest in science and to
Author for correspondence: society. However, our contemporary understanding of each is constrained
Marc H. Bornstein because most human biological, social and behavioural science is justly criticized
e-mail: [email protected] for drawing on too narrow a database. A central limitation of research in infancy
and parenting to date is that it has described constructs, structures, functions and
processes largely deriving from Western, educated, industrialized, rich and
democratic (so-called WEIRD) societies [1]. Reviews of the literature consistently
reveal that perhaps 80–90% of published science in these fields emanates from
Western Europe and North America (the minority ‘developed’ world), whereas
only perhaps 10–20% of the world’s literature represents humanity from the
majority (‘developing’) world [2]. Making broad theoretical generalizations
about infant behaviour and parental caregiving based largely on WEIRD societies
likely leads to errors of omission, and norms developed on limited samples risk
advancing ethnocentrism and its untoward ramifications.
Although the identification of psychological universals in humans constitutes
a central goal of the biological, social and behavioural sciences [3,4], contempor-
ary theory posits pervasive ecological moderation by physical and social
circumstances as influential in the development and expression of behaviour
Electronic supplementary material is available [5]. The absence of broad sampling or assessments of contextual moderation
have impeded a comprehensive appreciation of human development and caregiv-
online at https://dx.doi.org/10.6084/m9.
ing, led to repeated critiques of perspectives on vital topics in infancy and
figshare.c.3938296. caregiving that are built on inductions from single-society studies, and motivated

& 2017 The Author(s) Published by the Royal Society. All rights reserved.
 persistent calls for enlarged comparative studies across mul- (i) Question 1: society similarities and differences 2
tiple societies. Such comparisons are especially valuable Our first question concerns similarities and differences in

rspb.royalsocietypublishing.org
because they expose and deepen our understanding of bio- activities of infants and mothers across societies. Do infants’
logical, social and behavioural processes that may generalize behaviours and mothers’ parenting practices appear at the
across disparate populations, and provide the opportunity to same frequency and for the same duration? It could be that a
formally test society moderation of observed effects. tendency to arrive at the same phenotype, regardless of varia-
The remedy for these limitations—the lack of broad repre- bility in the genotype or environment [7,8], or universal
sentativeness and the complementary need for proper tests of demands on babies and caregivers, call forth highly similar
generalizable theory—is to identify effects that are society infant behaviours and parenting practices. Alternatively, how-
common (i.e. that obtain across societies and generalize ever, context forms a powerful basis for differentiating and
widely) versus society specific (i.e. that obtain only in the shaping activity, and it has been observed that the most forma-
place they are studied) through the close observation and tive source of influence operating on human development is

Proc. R. Soc. B 284: 20172168

comparison of multiple diverse wider-world samples. In a the place a child is born. Of course, infants and parents could
nutshell, only similarities and differences in infancy and show similar patterns of activity but at different absolute
parenting across disparate societies can supply unique evidence levels across different societies. We considered the behaviours
of society-common versus society-specific biological, social and of infants and the practices of mothers we studied might be
behavioural processes. The present investigation aimed to fill universal in the sense that infants and mothers everywhere
Downloaded from https://royalsocietypublishing.org/ on 18 November 2023

gaps in fundamental knowledge about human infancy, infant engage in similar activities, but we expected that their beha-
caregiving and infant–mother relationships in the vital develop- viours and practices would be engaged in at different levels
mental period of the first half-year of the human infant’s life across societies. We studied 13 infant behaviour indicators
by submitting standardized and detailed observations of that represent key developmental tasks and performance com-
infant behaviours and maternal parenting practices in 11 dis- petencies that are critical to successful adaptation of human
tinct societies around the world to quantitative microanalysis; infants around the middle of their first year of life, and 12
samples came from Argentina, Belgium, Brazil, Cameroon, maternal practice and context indicators that assess the pri-
France, Israel, Italy, Japan, Kenya, South Korea and the United mary caregiving tasks and performance competencies
States. Science can only benefit from enlarged empirical repre- required of the parent of a young infant.
sentativeness, and this cross-society approach, deriving from The derivation of our activity foci might best be understood
the world’s inhabited continents, afforded a rare opportunity with reference to the emic/etic distinction [9,10]. Etic constructs
to assess the specificity versus the generality of findings in the consist of accounts, descriptions and analyses of activities
youngest children and their caregivers across divergent contexts. expressed in terms of conceptual schemes and categories that
Several considerations guided our selection of the 11 are regarded as meaningful and appropriate by the broad com-
societies. The specific samples we studied all consisted of pri- munity of scientific observers. (By contrast, emic constructs
miparous mothers and their healthy young babies, with consist of accounts, descriptions and analyses expressed in
samples balanced for infant gender, but the 11 differed in terms of conceptual schemes and categories that are regarded
national origin, geographical location, history and ideology, as meaningful and appropriate by members of a particular
social and economic conditions, values and other sociodemo- community.) Etic classifications are available in advance of,
graphic characteristics thought to shape early development rather than determined during, analysis and are based on
and caregiving. Drawing on a wide representation of infants prior encompassing sampling or surveys (emic units are
and their mothers, this research begins to illuminate how indigenous and discovered, not predicted). Thus, the etic
comprehensive are presumed universals of human infancy organization of a cross-society scheme of activities is created
and parenting. This is not a report about specific societies, by the analyst, as we have done here. Etic knowledge is essen-
but one that uses information from many societies to address tial for cross-society comparison because such comparison
three general theoretical issues: similarities and differences in necessarily demands standard units and categories (emic
base rates of infant and maternal activities, coherence knowledge is relevant for an intuitive and empathic under-
in infant and in maternal activities and correspondence in standing of an activity). All behaviours can be described
infant –mother activities. from these two vantages. The etic approach validly treats
diverse samples together at one time (the emic approach is
sample specific and applies to one sample at a time).
(b) Three questions about infancy and parenting
Guided by theory, we re-visited three sets of basic and endur-
ing questions about infants, mothers and infant–mother (ii) Question 2: behavioural coherence
relationships in the context of this multi-society study, and Our second question pertains to coherence in infants’ and
we developed specific hypotheses about each. As detailed coherence in mothers’ activities. Do infants’ behaviours, and
below, contemporary scientific inquiry offers compelling but do mothers’ parenting practices, each exhibit cohering pat-
contrasting answers to each of the three basic questions. terns, or are the respective activities of each unrelated?
However, our expectations about society similarities and Coherence is a major theme in the biological, social and behav-
differences, covariation and correspondence in infant beha- ioural sciences and describes organized patterns among
viours and maternal parenting practices are tied together systems or activities. We asked if the behaviours of infants
by the specificity principle, which holds that individual infant and the parenting practices of mothers in different societies
behaviours are distinct and follow independent courses of are each organized in any systematic pattern as evidenced by
emergence and ontogeny, and that parenting practices are mul- their covariation, a standard means to evaluate associations.
tidimensional and characterized by specific forms that relate in Opposing perspectives on infancy and on parenting raise
specific ways to specific aspects of child development [6]. central theoretical questions about coherence.
 (1) Coherence in infancy maternal practices are largely uncoordinated, especially 3
With respect to infant coherence, historical and contempo- during the first fetus-ex-utero months of life. Infant and

rspb.royalsocietypublishing.org
rary theories disagree. Many classical observers of human mother are not always on the same schedules. The two have
infancy—Freud [11], Piaget [12], Erikson [13], and Werner different priorities and are sometimes simply out of step in
[14]—posited that early emerging behaviours might covary early dyadic development. Alternatively, parents’ overall
[see also 15–21] because interrelated behaviours come ‘online’ level of stimulation or engagement might relate to infants’
together and develop or manifest similarly. However, other overall level of development generally. From this perspective,
theorists have questioned the utility and even existence of parenting practices in any or all domains might coordinate
coherence, noting that such ‘stage’ theories are descriptive with infant behaviours in any or all domains. The overall sen-
rather than explanatory, downplay individual differences and sitive or authoritative mother may have the secure or behaved
fail to contribute to understanding individual development, dis- or verbal child. A third possibility is that specific (rather than
count the sometimes nonlinear progression of development, and general or ad hoc) practices of parents relate to specific (rather

Proc. R. Soc. B 284: 20172168

are idealistic when life is replete with unevenness, variation and than general or ad hoc) behaviours of infants. Real-time encoun-
inconsistency [22–24]. In accord, a general principle of biological ters between parents and infants are constructed of specific
development—the independence of systems—postulates that functional connections that shape broader behavioural
different organs and physiological structures develop at differ- tendencies in child development.
ent times and grow along different trajectories. For example, Correspondences begin to define mutual influences
Downloaded from https://royalsocietypublishing.org/ on 18 November 2023

different brain regions and neural connections between them that child and parent exert on one another. Operationally,
mature on different schedules in infancy, and different organ correspondence refers to a pattern of covariation between indi-
and sensory systems develop at different times and mature viduals. From specificity, we expected that infants and mothers
along different (staggered) trajectories [25–32]. would show positive correspondences but only between
conceptually related domains of activity. For example, signifi-
cant correspondence between maternal provisions in the
(2) Coherence in parenting
environment and infant exploration (but not other domains)
Some theories—psychoanalysis, personality, ethology and
points to mother–infant cognitive attunement, and correspon-
attachment—also conceptualize parenting (usually maternal)
dence between mother and infant social behaviours points to
practices as more or less cohering and unidimensional—often
their socioemotional attunement. However, there is no saying
denoted as ‘good’, ‘good enough’, ‘sensitive’, ‘warm’ and ‘ade-
that mothers’ promotion of infant verbal development, for
quate’ [33–38]. The assumptions of such monistic parenting
instance, is related to infant motor development. To detect
theories are that caregiving is characterized by one or a small
regular relations between caregiving or the environment on
number of bipolar dimensions and that caregiving is ‘trait-
the one hand and child characteristics or development on the
like’ and packaged in all-or-nothing ‘styles’ [39,40]. That is,
other necessitates seeking and finding specific and matching
parents’ practices cohere to manifest in similar ways across
combinations of domains.
interaction, time and context. Thus, for example, mothers are
To address these three questions and test our hypotheses,
authoritative or authoritarian or permissive. A position
we first identified, recorded, micro-coded, and analysed 13
alternative to the trait and style conceptualization is that
specific behaviours of infants and 12 specific parenting prac-
diverse parenting practices are not necessarily linked, parents
tices of mothers in the 11 worldwide societies; we then
do not necessarily or always behave in uniform or consistent
organized the behaviours and practices into thematic and
ways, and parents vary in the constellation and patterning of
theoretically derived overarching domains and finally, we sub-
their caregiving practices according to situations, goals and
mitted the infant and maternal domains to three sets of
child characteristics and behaviours [6]. That is, parents
analyses. The three theoretical questions about young human
emphasize different activities with their infants at different
infants and their mothers include one about mean levels of
times in different contexts for different ends, so diverse
infant and mother activity (the society similarity/difference
domains of parenting constitute distinctive constructs.
question), a second about activity organization in infants and
Operationally, coherence in infancy and in parenting
in mothers (the coherence question), and a third about activity
would be characterized by systematic interrelations among
associations between infants and mothers (the correspondence
diverse behaviours and practices, respectively. To test coher-
question). We also explored moderation of activity, coherence
ence, we correlated infant behaviours and correlated mother
and correspondence by society.
parenting practices. Because we subscribe to the specificity
principle, we expected infants and mothers, even from con-
trasting societies, to show independence across most
behaviours and practices (but not all because some parenting
practices, e.g. engaging in didactic activities and outfitting
2. Results
the material environment, go together). (a) Society similarities and differences
(i) Cross-society comparisons of infant behaviours
(iii) Question 3: correspondence Although infants in all societies engaged in etic physical,
Are infant behaviours and parent practices related to one social, exploration, nondistress vocalization and distress beha-
another in unsystematic, in omnibus, or in specific ways? Par- viours (see electronic supplementary material, figure S1), we
enting is an interpersonal activity that reflects the constant compared the 11 societies to provide an inferential understand-
interplay and coordination of two parties [41], and correspon- ing of mean levels (and ranges) of specific infant behaviours.
dences between the two parties index the nature of mutual We expected society differences in infant behaviours, and
influences that mother and infant exert on one another. they emerged for four of five domains: physical, social, explora-
It could be (however unlikely) that infant behaviours and tion and nondistress vocalization. Figure 1 plots society effects
 infant physical infant social infant exploration 4
1.0 1.0 1.0

rspb.royalsocietypublishing.org
0.5 0.5 0.5

0 0 0

–0.5 –0.5 –0.5

–1.0 –1.0 –1.0

lg a
Br m
m l

Fr n
Is e
el
ly
n

h ya
te rea

es

lg a
Br m
m l
Fr n
Is e
el
ly
n
a
a
es

lg a
Br m
m l

Fr n
Is e
el
ly
n
a
te rea

es
Ca azi

Ca azi

Ca azi
Be tin

Be tin

ut ny

Be tin

ut ny
oo

pa

oo

pa

oo

pa
ra

ra

ra
iu

Ita

iu

Ita

iu

Ita
at

at

at
an

an

ni Kor

an
So en

o
Ja

Ja
So K e

Ja
So K e
en

er

en

er

en

er
St

St

St
K

K
K
rg

rg

rg
d

d
h

h
te
ut
A

A
ni

ni
U

Proc. R. Soc. B 284: 20172168

infant nondistress infant distress
1.5 1.0

1.0
0.5
0.5
Downloaded from https://royalsocietypublishing.org/ on 18 November 2023

0 0

–0.5
–0.5
–1.0

–1.5 –1.0
lg a
Br m
m l

Fr n
Is e
el
ly
n
a
te rea

es

lg a
Br m
m l

Fr n
Is e
el
ly
n
ut nya

d ea
es
Ca azi

Ca azi
Be tin

Be tin

c
oo

pa

oo

pa
ra

ra
iu

Ita

iu

Ita
at

at
an

an

ni Kor
So en

o
Ja

Ja

So e
en

er

en

er
St

St
K
K

K
rg

rg
d
h

h
te
ut
A

A
ni
U

U
Figure 1. Infant behaviour domain comparisons by society. The effects of society, controlling for nuisance covariate(s) and their 95% confidence intervals for each
infant domain score. The zero-line represents the generalized grand mean. Where the confidence intervals do not overlap the zero-line, societies differ from the
generalized grand mean; where confidence intervals from different societies do not overlap, the generalized means for those societies differ.

and their 95% confidence intervals as differences from the domains tested coherence of infant domains across societies;
generalized grand mean across societies for each infant no differences by society emerged.
domain. (Generalized means, standard errors, 95% confidence We also expected that most observed correlations among
intervals, F-tests of society differences and Holm–Bonferroni domains of mother parenting practices would be small, but
pairwise tests for infant behaviours are reported in the conceptually related domains would correlate. Nine out of
electronic supplementary material.) 15 cross-domain correlations were statistically significant
(table 1B bottom), again powered by the large sample size.
However, the mean absolute coherence ¼ 0.18 or 3% shared
(ii) Cross-society comparisons for mother parenting practices variance, range jrj ¼ 0.01 –0.47, and effect sizes of significant
Although mothers in all societies engaged in etic nurturing, correlations were generally small to medium (table 1B top).
physical, social, didactic, material and language practices Five omnibus society differences—out of 15—are described
with their infants, society differences emerged in levels for in the electronic supplementary material.
all six maternal parenting practice domains. Figure 2 plots
the society effects and their 95% confidence intervals as
differences from the generalized grand mean for each (c) Correspondences between behaviours of infants
mother practice domain score. (Generalized means, standard
errors, 95% confidence intervals, F-tests of society differences
and parenting practices of mothers
In accord with specificity, we expected positive relations in six
and Holm –Bonferroni pairwise tests for maternal parenting
conceptually corresponding infant and maternal domains:
practices are reported in the electronic supplementary
infant physical and mother physical, infant social and
material.)
mother social, infant social and mother language, infant
exploration and mother didactic, infant exploration and
mother material, and infant nondistress vocalization and
(b) Coherence among behaviours of infants and among mother language. Table 2 shows correlations between infant
parenting practices of mothers and maternal domain scores. Significant positive relations
We expected that observed coherence correlations of infant emerged in all conceptually corresponding domains except
behaviour domains would be small, and they were. Seven one pair (infant nondistress vocalization and mother
out of 10 cross-domain correlations were statistically signifi- language). The average correlation using the six correspond-
cant (table 1A bottom); however, these statistical ing infant and maternal domain scores, r ¼ 0.20,
significances were powered by the large sample size (N ¼ range ¼ 20.06 to 0.40, was significantly larger than the aver-
796), and the mean absolute correlation ¼ 0.09 or less than age absolute correlation among the other domains, jrj ¼
1% of variance shared (table 1A top); effect sizes were 0.06, range ¼ 20.44 to 0.14, z ¼ 2.84, p ¼ 0.005. Only one
small, range jrj ¼ 0.01 to 0.17. Omnibus tests of 11 correlations omnibus society difference—out of 30—is described in the
and pairwise analyses by society for each pair of infant electronic supplementary material.
 mother nurture mother physical mother social 5
1.0 1.0 1.0

rspb.royalsocietypublishing.org
0.5 0.5 0.5

0 0 0

–0.5 –0.5 –0.5

–1.0 –1.0 –1.0

lg a
Br m
m l

Fr n
Is e
el
ly

K n
ya

te rea

es

lg a
Br m
Ca azil

Fr n
Is e
el
ly
n
ut nya

ea

es

lg a
Br m
Ca azil

Fr n
Is e
el
ly
n
ut nya

ea

es
Ca azi
Be tin

Be tin

Be tin

c
oo

pa

oo

pa

oo

pa
ra

ra

ra
iu

Ita

iu

Ita

iu

Ita
at

at

at
an

an

or

an

or
So en

o
Ja

Ja

So K e

Ja

So K e
en

er

en

er

en

er
St

St

St
K

K
m

m
rg

rg

rg
d

d
h

h
te

te
ut
A

A
ni

ni

ni
U

Proc. R. Soc. B 284: 20172168

mother didactic mother material mother language
1.0 1.0 1.5
1.0
0.5 0.5 0.5
0
Downloaded from https://royalsocietypublishing.org/ on 18 November 2023

0 0
–0.5
–1.0
–0.5 –0.5
–1.5
–1.0 –1.0 –2.0
lg a
Br m
m l

Fr n
Is e
el
ly

K n
ya

te rea

es

lg a
Br m
Ca azil

Fr n
Is e
el
ly
n
ut nya

ea

es

lg a
Br m
Ca azil

Fr n
Is e
el
ly
n
ut nya

ea

es
Ca azi
Be tin

Be tin

Be tin

c
oo

pa

oo

pa

oo

pa
ra

ra

ra
iu

Ita

iu

Ita

iu

Ita
at

at

at
an

an

or

an

or
So en

o
Ja

Ja

So Ke

Ja

So Ke
en

er

en

er

en

er
St

St

St
K

K
m

m
rg

rg

rg
d

d
h

h
te

te
ut
A

A
ni

ni

ni
U

U
Figure 2. Mother parenting practice domain comparisons by society. The effects of society, controlling for nuisance covariate(s), and their 95% confidence intervals
for each mother domain score. The zero-line represents the generalized grand mean. Where the confidence intervals do not overlap the zero-line, societies differ
from the generalized grand mean; where confidence intervals from different societies do not overlap, the generalized means for those societies differ.

Table 1. Coherence correlations. Note, N ¼ 796. Below the diagonal: the correlation coefficient in each cell controlled for potential nuisance covariates. All
significances held when controlling for possible corresponding covariates. Above the diagonal: corresponding proportions of unshared variance.

(A) Correlations among infant behaviour domains and proportions of unshared variance
physical social exploration nondistress vocalization distress communication
physical — 0.99 0.98 0.99 0.99
social 20.10** — 0.99 0.98 0.99
exploration 0.15*** 0.11** — 0.99 0.99
nondistress vocalization 20.07 0.14** 20.03 — 0.97
distress communication 20.01 20.11** 20.08* 0.17*** —
(B) Correlations among mother parenting practice domains and proportions of unshared variance
nurture physical social didactic material language
nurture — 0.99 0.99 0.98 0.93 0.98
physical 20.01 — 0.96 0.95 0.99 0.99
social 0.05 0.19***a — 0.87 0.99 0.78
didactic 20.15*** 0.22***a 0.36***a — 0.92 0.84
material 20.27*** 20.06 20.04 0.29*** — 0.99
a a
language 0.13*** 0.05 0.47*** 0.40*** 0.04 —
a
Society was covaried when society effects emerged in the correlation.
*p , 0.05, **p , 0.01, ***p , 0.001 for two-tailed test.

limitations in the research literature constrain answers to these

3. Discussion fundamental questions about human infancy and early parental
Are infant behaviours and mother parenting practices patterned caregiving. Specifically, the narrow participant database in
across societies around the world? Does each vary across the research enterprise undermines global generalizability, as
societies? Do infants’ behaviours, and do mothers’ parenting biological, social and behavioural developmental science suffers
practices, each cohere? Do infants’ behaviours and mothers’ a dearth of multi-society data on conditions of early develop-
parenting practices correspond? Contemporary context-related ment [2]. Multi-society studies are requisite to identifying,
 Table 2. Correspondence correlations between infant behaviour and mother parenting practice domains. Note, N ¼ 796. The correlation coefficient in each cell 6
controlled for potential nuisance covariates.

rspb.royalsocietypublishing.org
infant domain

mother domain physical social exploration nondistress vocalization distress

nurture 20.10** 20.11** 20.44*** 20.02 0.11**

†††
physical 0.40 20.02 0.14*** 20.02 0.00
social 0.03 0.17††† 20.06 20.02 20.02
didactic 0.01 0.07 0.30††† 20.07 0.02

Proc. R. Soc. B 284: 20172168

†††
material 0.05 20.02 0.38 20.04 0.01
language 20.07 0.09††a 20.04 20.06 0.03
a
Society was covaried when society effects emerged in the correlation.
**p , 0.01, ***p , 0.001 for two-tailed test of non-corresponding domains.
Downloaded from https://royalsocietypublishing.org/ on 18 November 2023

††
p , 0.01, †††p , 0.001 for one-tailed test for corresponding domain.

distinguishing and appreciating general as well as specific correlations for mother parenting practices were significant, but
patterns in human development. Only multi-society inquiry in accord with specificity most were unrelated or negatively
provides adequate opportunity to explore uniformity and related. The largest coherences obtained between mother
diversity in human phenotypic expression. language and social and didactic caregiving, which is unsurpris-
With respect to our three research questions, first, very ing because the social and didactic domains of caregiving
young infants in different societies engaged in the five domains include verbal attempts to amuse the baby and to draw the
of behaviour we measured. Still, in accord with specificity, baby’s attention to objects, respectively. Given the study’s sub-
society differences in levels emerged in four of the five stantial sample size there was adequate power to detect even
domains. Only infant distress did not differ across societies, small effects, but the few society differences in mother coherence
which is not surprising as many contemporary theoreticians correlations that emerged were attributed to a single society.
have pointed to the conserved nature of infant cry across Thus, different domains of behaviours and of practices appear
species and the likely high survival value of infant cry to be largely separate in infants and in mothers, respectively, as
[42–44]. Mothers of young infants in different societies also specificity predicts. We studied behaviours and domains at
show etic behaviours, physically encouraging their infants’ micro-coded levels; it could be that behaviours of infants and
motor skills, socially engaging with them, encouraging their practices of mothers at more aggregated levels would cohere.
didactic activities and speaking to them; nonetheless, as speci- Nonetheless, these infant and parent findings raise questions
ficity predicts, society differences emerged in all six domains. about widely accepted stage, style and trait interdependent
That five-month olds in different societies already vary in views of development and of caregiving in the first years of life
some basic behaviours calls into question the validity of and, rather, favour biological independence perspectives.
widely accepted Western norms of early human development Finally, we found that mothers and their young infants
[45]. That mothers of five-month olds in different societies were moderately attuned, but mostly in conceptually corre-
vary in how they parent very young infants points to societal sponding domains, also as predicted by specificity: mothers
influences over basic human caregiving. Different societies who more often promoted their infants’ physical development
emphasize different specific practices. Notably, however, had infants who were more developed physically; mothers
infants and mothers in no one society exceeded or fell behind who more often encouraged their infants to engage in dyadic
all others; rather, the distributions of behaviours and practices social exchanges had infants who attended to and smiled at
varied in society-specific ways. Likely, these early idiosyncratic them more; and mothers who more often engaged their infants
society differences set the stage for how unique constellations didactically and provided more, and more responsive, investi-
of experiences inflect trajectories of society-specific ontogenetic gative materials for their infants had infants who explored their
development. Indeed, it has been argued that central to a con- environments more. These relations were also common across
cept of society is the fact that different peoples possess different societies, with only one society difference in mother–infant
ideas as well as behave in different ways with respect to their correspondence. Of course, these correlations do not measure
parenting and their children’s development. Some authorities direction of effects or imply causation between infants and
have even contended that the transmission of society-specific mothers; importantly, however, they document the pervasive-
values and characteristics largely transpires intergeneration- ness of specific and early attuned dyadic interactions. This
ally—from parent to child—within families. That is, that uniformity of correspondences further supports some univer-
societal values are conserved through family life. sal aspects of mother–infant interaction. The trait and style
Second, in accord with specificity, meaningful between- approach to parenting tends to reify typologies and to cast
domain correlations in infant behaviours are sparse, them as inherent to the individual; it is also unwelcoming to
unsystematic and small. Only meagre patterns of correlation developmental questions, to ontogenetic change and to bidir-
across domains were obtained in spite of the large sample size ectionality (that particular child characteristics or behaviours,
and shared method and source variance. Furthermore, no society situations or goals, times and interactions, evoke particular
differences in infant coherence emerged. A few between-domain features of parenting). Yet, child effects and transactional
 models implicate behavioural adjustments parents regularly philosophers and theorists through the ages, in childhood we 7
make relative to children’s age and gender, appearance and are open to expectations and influences we carry with us long

rspb.royalsocietypublishing.org
behaviour, temperament and activity. after we leave our family of origin. Reciprocally, virtually
In one exception to predicted correspondence, amounts of all of young infants’ worldly knowledge is acquired from
infant nondistress vocalization and mother language were interactions they have with their parents. Infant and parent
unrelated. However, the analysis of simple base rates of may be separate entities, but they are inextricably dyadic [49].
mother –infant correspondence in the vocal modality is inap- Establishing adaptive modes of mutual infant–mother inter-
propriate; turn-taking in conversation is more fundamental action soon after birth is a primary task of the dyad because
and valid. Sequential analysis of these data has borne out it is requisite to wholesome development in otherwise
this explanation and supports correspondence of this specific dependent young.
contingency in mother –infant vocalization [46]. (In addition, The normative patterns of expression and association of
two un-hypothesized, small, positive correlations are infant behaviours and maternal parenting practices, as they

Proc. R. Soc. B 284: 20172168

discussed in the electronic supplementary material.) occur in dyadic involvement in everyday situations, have far-
Contrasting theoretical positions have grown up to explain reaching implications for biological and developmental theory
basic features of infant development and parenting. Some and research, as well as clinical practice and social policy.
emic theoretic orientations focus on idiosyncratic expressions Because it embraces a wider selection of samples and analyses
of behaviour or caregiving in one or another societal infant and mother activities in a detailed way, this investigation
Downloaded from https://royalsocietypublishing.org/ on 18 November 2023

group, others have been advanced to account for dimensional- may advance theory, help guide fresh research about early
ity in infant behaviours and parenting practices. Overall, human development, and inform programmes aimed at
however, the three sets of answers to the societal similarity improving parental caregiving and child well-being [50]. On
and difference, coherence and correspondence questions in the basis of these findings, biological, behavioural and clinical
infancy and caregiving from this multi-society study accord scientists may formulate more accurate accounts of how parents
with the specificity principle in human development and and infants interact in everyday situations to promote or hinder
parenting [6]. children’s wholesome development. Despite the pervasive
In considering the findings, strengths of this study should be differences in mean levels of nearly all infant and mother activi-
balanced with its limitations. It is noteworthy that some samples ties across societies, coherence among infant and among mother
were relatively, although not wholly, restricted in terms of their activities and correspondence between infant and mother activi-
sociodemographic level, education and living conditions; ties were nearly universal across societies. Consistent patterns of
mothers were the primary caregivers to their infants; and infants interaction in parent–infant encounters anticipate stable
were all firstborn and healthy. This sampling homogeneity con- psychological tendencies in children [51] and forecast stable be-
trols for potential confounding factors that are often present, but havioural currents in other relationships and other situations
unaccounted for, in multi-society comparisons. It is possible, children encounter as they develop [19]. Social competence in
however, that different results might be found with other children, for example, has origins in specific socioemotional
samples, such as multiparous mothers or mothers with older characteristics of the parent–infant relationship [52].
or at-risk children or with fathers and other caregivers. It is Some psychological theorists seek developmental univer-
also the case that some society samples were small, which left sals, whereas other theorists focus on diversity in core
pairwise comparisons underpowered, and future studies processes. Comparative biological, social and behavioural
should include larger and more diverse samples. Whereas science concerns the common and the specific as well as the
most social and behavioural science studies compare general complementary conditions under which society-common
parental cognitions or broad parental styles, almost always in mechanisms may be expressed in society-specific ways. These
single-society settings, the present study observed, recorded, two mutually interacting forces shape ontogeny and caregiving
coded, analysed and compared specific infant and mother from the very start of each individual’s life.
micro-coded activities directly in almost 800 families from
about a dozen societies around the world.
Infants everywhere have the same biological needs and
must succeed at many of the same developmental tasks and 4. Material and methods
challenges. Infancy is the time when we forge our first social
bonds, first learn how to express and read human emotions,
(a) Participants
A total of 796 infant–mother dyads from Argentina, Belgium,
and first make sense of the physical world. Parents escort Brazil, Cameroon, France, Israel, Italy, Japan, Kenya, South Korea
infants through all those dramatic firsts. Adult caregiving and the United States participated. All infants were five months
responsibilities are substantial during infancy because the altri- of age when observed, firstborn children in their families, born
cial human infant’s ability to cope alone is minimal and infants within three weeks of their due date, and healthy at birth and at
depend on caregiving for survival. Parents bear responsibilities the time of the observation. These inclusion criteria were selected
to ensure offspring vitality and guide adjustment and adap- to control some of the natural variation in infant–mother relation-
tation to the physical and social environment in which the ships associated with infant age, birth order, term status and health.
child is born. The fit is neat to the task. The amount of inter- Approximately equal proportions of girls and boys were recruited
into each society sample (electronic supplementary material).
action between parents and offspring is greatest in infancy,
Electronic supplementary material table S1, provides
and early childhood is the time when human beings are
sociodemographic characteristics of the samples, and electronic
especially susceptible to the influences of experience [47]. supplementary material table S2, provides comparative socio-
Indeed, the opportunities afforded by enhanced parental influ- demographic information about the larger societies. Argentine
ence and prolonged childhood learning are thought to infant–mother dyads (n ¼ 134) were recruited from Buenos
constitute evolutionary reasons for neoteny, the extended Aires (66% of the sample) and its surrounding metropolitan area
duration of human infancy [47,48]. In the view of many and from Córdoba Province (34%), a rural community. Belgian
 dyads (n ¼ 115) were recruited from Ghent (47%) and Antwerp indicators of the five infant behaviour and six parenting practice 8
(53%). Brazilian dyads (n ¼ 59) were recruited from metropolitan domains appear in the electronic supplementary material.)
Rio de Janeiro. Cameroonian dyads (n ¼ 29) were Nso, a group This report focuses on domain-level analyses. Domain scores

rspb.royalsocietypublishing.org
indigenous to the northwestern part of Cameroon. French dyads were calculated as the mean of the (usually standardized) infant
(n ¼ 59) were recruited from Paris and its immediate banlieux. behaviour, maternal practice, and context indicators that related
Israeli dyads (n ¼ 31) were recruited from metropolitan Haifa. Ita- conceptually to the domain [54,55]. (Descriptive statistics and pro-
lian dyads (n ¼ 98) were recruited from metropolitan Padua (58%), files for domain indicators appear in electronic supplementary
a northern city, and from Ruoti (42%), a southern town. Japanese material, figures S1 and S2.)
dyads (n ¼ 74) were recruited from the greater metropolitan Tokyo
area. Kenyan dyads (n ¼ 25) were Kamba, Bantus living in eastern
parts of Kenya. South Korean dyads (n ¼ 57) were recruited (c) Analytic plan
from the greater Seoul metropolitan area. Dyads from the United Analyses of infant and mother domains followed three main
States (n ¼ 115) were randomly selected from a larger sample of paths developed to evaluate the three sets of substantive and

Proc. R. Soc. B 284: 20172168

European American participants (N ¼ 373); 83% came from the theoretical questions: (i) society similarity and difference, (ii)
Washington, DC, metropolitan area, including Maryland and infant coherence and mother coherence and (iii) infant– mother
Virginia, and 17% from rural West Virginia. All mothers gave correspondence. We collapsed data from the 11 societies and cal-
informed consent, and data collection procedures were approved culated rates and durations per hour and correlation coefficients.
by the NICHD Institutional Review Board (IRB) as well as local (In the electronic supplementary material, we also report omnibus
tests of society differences.) Analyses used robust and resistant
Downloaded from https://royalsocietypublishing.org/ on 18 November 2023

IRBs in the 11 societies. (See the electronic supplementary material

for details of sample recruitment, societal variation, etc.) statistics in R v. 3.1.1 (R. Development Core Team, 2014) calling
the rlm() function [56] with MM estimation to obtain weighted,
unbiased, consistent and efficient estimates of model coefficients
with high breakdown points. To focus on society-level variation
(b) Procedures per se, the analyses also applied multiple covariates: (i) all ana-
Infants and mothers were visited and audio/videorecorded in lyses controlled for nuisance covariates—cross-society variation
their homes for 1 h of naturalistic ongoing interaction (an optimal associated with sampling error (infant age and birth weight),
time for reliable and valid data collection [53]) when only infant, (ii) significant society differences were controlled for context cov-
mother and a female researcher, who was native to the society, ariates—alternative variance associated with society differences
were present. Home visits were scheduled at a convenient time (mother age, education, and the duration she was in view of
for the families and when the infants were expected to be awake her child), and (iii) when appropriate, analyses controlled for cor-
and alert. This research was conducted in the natural context responding covariates—variance associated with the dyadic
in which infant and mother interactions take place to achieve nature of data collection (mother for infant and infant for
maximal ecological validity. Records were then coded using a mother). (More details about the analytic plan, power analyses,
mutually exclusive and exhaustive, continuous and comprehen- preliminary analyses, covariates, and correlations of infant and
sive, coding system yielding unbiased estimates of the frequency maternal domains with covariates are available in the electronic
and duration of each activity. This microanalytic strategy allowed supplementary material, tables S6 and S7.)
us to examine infant and mother activities at the level of in-
the-moment lived experiences. Coders established initial training Ethics. This study was performed in compliance with the guidelines of
reliability and maintained reliability throughout coding by the Eunice Kennedy Shriver National Institute of Child Health and
having two coders independently score approximately 20% of Human Development Institutional Review Board.
the videorecords in their entirety. Inter-rater reliabilities follow Data accessibility. Informed consent forms signed by participants in this
the individual codes below. study do not allow public sharing of the data. Researchers interested
In infants, the five domains were operationalized in 13 behav- in collaborating on these data may contact the first author.
iour indicators. We focused on physical development (balance, Authors’ contributions. M.H.B. designed the study and drafted the manu-
ICC ¼ 0.89 and movement, Intraclass correlation coefficient script; D.L.P., Y.P. and O.M.H. performed the data analyses, drafted
the method and results, and provided critical revisions of the full
(ICC) ¼ 0.94), social interaction (looking at mother, k ¼ 0.66, 96%
manuscript; and J.T.D.S. helped to develop the coding systems, over-
agreement; smiling, k ¼ 0.50, 99% agreement; alert expression, saw the training and data coding across countries, and provided
k ¼ 0.70, 88% agreement), exploration (extent and efficiency of revisions of the manuscript. All authors approved the final version
looking, k ¼ 0.70, 86% agreement; touching; k ¼ 0.66, 89% agree- of the manuscript for submission.
ment and mouthing objects, k ¼ 0.75, 95% agreement), nondistress Competing interests. We have no competing interests.
vocalization (k ¼ 0.68, 94% agreement) and distress communication Funding. This research was supported by the Intramural Research
(by face and voice, k ¼ 0.72, 99% agreement). In mothers, the six par- Program of the NIH, NICHD.
enting practice domains were operationalized in 12 practice and Acknowledgements. We thank H. Azuma, S. Bali, E. Berti, A. De Houwer,
context indicators. We focused on nurturance (feeding, k ¼ 0.90, C. de Zingman Galperin, M. Kabiru, K. Kwak, S. Maital, M. L. de
98% agreement; bathing, k ¼ 0.87, 98% agreement; holding, k ¼ Moura de Siedel, A. Nsamenang, M.-G. Pêcheux, J. Ruel, S. Toda,
0.93, 97% agreement), promotion of physical development (balance, P. Venuti and A. Vyt.
ICC ¼ 0.91 and movement, ICC ¼ 0.89), social interaction (encoura-
ging attention to mother, k ¼ 0.70, 95% agreement; expressing
affection, k ¼ 0.68, 99% agreement; social play, k ¼ 0.74, 98% agree-
ment), didactics (encouraging attention to properties, objects or Endnote
1
events in the environment, k ¼ 0.73, 94% agreement), materially out- Here we use the term ‘society’ to mean a group of individuals living
fitting the infant’s surround with quantity (ICC ¼ 0.97) and quality as members of a community and to refer to our samples generically.
We distinguish society from (i) ‘culture’ which connotes the beliefs
(ICC ¼ 0.95) of objects, and speech to baby (k ¼ 0.69, 89% agree-
and behaviours characteristic of a particular social or ethnic group
ment). These purpose-made codes were theoretically derived and
and are transmitted from one generation to another, (ii) ‘country’
consisted of specific, observable, micro-coded etic behaviours that which connotes a political state, and (iii) ‘nation’ which connotes a
are commonly observed in infants and mothers in all societies. body of people, associated with a particular territory, that possesses
Using specific micro-coded indicators rather than subjective a government peculiarly its own. Although the samples came from
macro-coded ratings increased cross-society validity. (Additional different nations, we do not claim that they are representative of
procedures, operational definitions, and coding details for all their respective nations.
 References 9

rspb.royalsocietypublishing.org
1. Henrich J, Heine SJ, Norenzayan A. 2010 The 17. Emde RN, Gaensbauer TJ, Harmon RJ. 1976 developmental science (eds LS Liben, U Müller),
weirdest people in the world? Behav. Brain Sci. 33, Emotional expression in infancy: a biobehavioral pp. 9–62, 7th edn. Hoboken, NJ: Wiley.
61 –83. (doi:10.1017/S0140525X0999152X) study. Psychological issues monograph series (10th 31. Turkewitz G, Kenny PA. 1982 Limitations on input
2. Tomlinson M, Bornstein MH, Marlow M, Swartz L. vol., monograph 37). New York, NY: International as a basis for neural organization and perceptual
2014 Imbalances in the knowledge about infant University Press. development: a preliminary theoretical statement.
mental health in rich and poor countries: too little 18. Fischer KW. 1980 A theory of cognitive Dev. Psychobiol. 15, 357–368. (doi:10.1002/dev.
progress in bridging the gap. Infant Ment. Health J. development: the control and construction of 420150408)
35, 624–629. (doi:10.1002/imhj.21462) hierarchies of skills. Psychol. Rev. 87, 477– 531. 32. Yakovlev PI, Lecours AR. 1967 The myelogenetic
3. Cooper CR, Denner J. 1998 Theories linking culture (doi:10.1037/0033-295X.87.6.477) cycles of regional maturation of the brain. In

Proc. R. Soc. B 284: 20172168

and psychology: universal and community-specific 19. McCall RB. 1970 Qualitative transitions in behavioral Regional development of the brain in early life (ed. A
processes. Annu. Rev. Psychol. 49, 559 –584. development in the first two years of life. Minkowski), pp. 3– 70. Boston, MA: Blackwell
(doi:10.1146/annurev.psych.49.1.559) In Psychological development from infancy (eds MH Scientific Publications.
4. Norenzayan A, Heine SJ. 2005 Psychological Bornstein, W Kessen), pp. 183–224. Hillsdale, NJ: 33. Ainsworth MDS, Blehar M, Waters CE, Wall S. 1978
universals: what are they and how can we know? Erlbaum. Patterns of attachment: a psychological study of the
Downloaded from https://royalsocietypublishing.org/ on 18 November 2023

Psychol. Bull. 131, 763–784. (doi:10.1037/0033- 20. Sander LW. 1962 Issues in early mother–child strange situation. Hillsdale, NJ: Erlbaum.
2909.131.5.763) interaction. J. Am. Acad. Child Psychiat. 1, 141–166. 34. Bettelheim B. 1988 A good enough parent: a book
5. Bronfenbrenner U, Morris PA. 2007 The (doi:10.1016/S0002-7138(09)60013-3) on child-rearing. New York, NY: Vintage.
bioecological model of human development. In 21. Sroufe LA. 1997 Emotional development: the 35. Brody GH, Flor DL. 1997 Maternal psychological
Handbook of child psychology: theoretical models of organization of emotional life in the early years. functioning, family processes, and child adjustment
human development (W Damon, series ed.; RM Cambridge, UK: Cambridge University Press. in rural, single-parent, African American families.
Learner, volume ed.), vol. 1, pp. 793–828, 6th edn. 22. Ayer L, Bornstein MH. 2005 Stage theories of Dev. Psychol. 33, 1000–1011. (doi:10.1037/0012-
Hoboken, NJ: Wiley. human development. In Encyclopedia of 1649.33.6.1000)
6. Bornstein MH. 2015 Children’s parents. In Ecological applied developmental science (eds CB Fisher, RM 36. Rohner RP. 1986 The warmth dimension:
settings and processes in developmental systems. Lerner), vol. 2, pp. 1060–1061. Thousand Oaks, CA: foundations of parental acceptance-rejection theory.
Volume 4 of the handbook of child psychology and Sage. London, UK: Sage.
developmental science (RM Learner, series ed.; MH 23. Brainerd CJ. 1978 The stage question in cognitive- 37. Scarr S. 1992 Developmental theories for the 1990s:
Bornstein & T Leventhal, volume eds), pp. 55 –132, developmental theory. Behav. Brain Sci. 2, development and individual differences. Child Dev.
7th edn. Hoboken, NJ: Wiley. 173 –182. (doi:10.1017/S0140525X00073842) 63, 1 –19. (doi:10.2307/1130897)
7. Waddington CH. 1942 The canalization of 24. Lerner RM, Lewin-Bizan S, Warren AEA. 2011 38. Winnicott DW. 1957 The child and the family.
development and the inheritance of acquired Concepts and theories of human development. London, UK: Tavistock.
characteristics. Nature 150, 563– 565. (doi:10.1038/ In Developmental science (eds MH Bornstein, ME 39. Baumrind D. 1991 The influence of parenting style
150563a0) Lamb), pp. 3–49, 6th edn. New York, NY: on adolescent competence and substance use.
8. Waddington CH. 1957 The strategy of the genes: a Psychology Press. J. Early Adolescence 11, 56 –95. (doi:10.1177/
discussion of some aspects of theoretical biology. 25. Dubois J, Dehaene-Lambertz G, Perrin M, Mangin 0272431691111004)
London, UK: George Allen & Unwin, Ltd. J-F, Cointepas Y, Duchesnay E, Le Bihan D, Hertz- 40. Holden GW. 2009 Parenting: a dynamic perspective.
9. Harris M. 1976 History and significance of the emic/ Pannier L. 2007 Asynchrony of the early maturation Thousand Oaks, CA: Sage.
etic distinction. Annu. Rev. Anthropol. 5, 329–350. of white matter bundles in healthy infants: 41. Maccoby EE. 1992 The role of parents in the
(doi:10.1146/annurev.an.05.100176.001553) quantitative landmarks revealed noninvasively by socialization of children: an historical overview. Dev.
10. van de Vijver FJ. 2010 Emic–etic distinction. In diffusion tensor imaging. Hum. Brain Mapp. 29, Psychol. 28, 1006 –1017. (doi:10.1037/0012-1649.
Encyclopedia of cross-cultural school psychology 14 –27. (doi:10.1002/hbm.20363) 28.6.1006)
(ed. CS Clauss-Ehlers), pp. 422–423. New York, NY: 26. Gedd JN et al. 1999 Development of the human 42. Lingle S, Wyman MT, Kotrba R, Teichroeb LJ,
Springer. corpus callosum during childhood and adolescence: a Romanow CA. 2012 What makes a cry a cry? A
11. Freud S. 1949 An outline of psycho-analysis. longitudinal MRI study. Prog. NeuroPsychopharmacol. review of infant distress vocalizations. Curr. Zool.
New York, NY: Norton. Biol. Psychiatry 23, 571–588. (doi:10.1016/S0278- 58, 698 –725. (doi:10.1093/czoolo/58.5.698)
12. Piaget J. 1952 The origins of intelligence in children. 5846(99)00017-2) 43. Newman JD. 2007 Neural circuits underlying crying
Oxford, UK: International University Press. 27. Gottlieb G. 1971 Ontogenesis of sensory function in and cry responding in mammals. Behav. Brain Res.
13. Erikson EH. 1950 Childhood and society. New York, birds and mammals. In Biopsychology of development 182, 155–165. (doi:10.1016/j.bbr.2007.02.011)
NY: Norton. (eds E Tabach, LR Aronson, E Shaw), pp. 67–128. 44. Zeifman DM. 2001 An ethological analysis of human
14. Werner H. 1957 The concept of development from a New York, NY: Academic Press. infant crying: answering Tinbergen’s four questions.
comparative and organismic point of view. In The 28. Huttenlocher PR, Dabholkar AS. 1997 Regional Dev. Psychobiol. 39, 265–285. (doi:10.1002/dev.
concept of development: an issue in the study of differences in synaptogenesis in human cerebral 1005)
human behavior (ed. D Harris), pp. 125–148. cortex. J. Comp. Neurol. 387, 167–178. (doi:10. 45. Karasik LB, Tamis-LeMonda CS, Adolph KE,
Minneapolis, MN: University of Minnesota Press. 1002/(SICI)1096-9861(19971020)387:2,167::AID- Bornstein MH. 2015 Places and postures: a cross-
15. Bowlby J. 1969 Attachment and loss: vol. 1 CNE1.3.0.CO;2-Z) cultural comparison of sitting in 5-month-olds.
attachment, 2nd edn. New York, NY: Basic Books. 29. Rubenstein JL, Rakic P. 1999 Genetic control of J. Cross Cult. Psychol. 46, 1023–1038. (doi:10.1177/
16. de Rosnay M, Murray L. 2012 Maternal care as the cortical development. Cereb. Cortex 9, 521 –523. 0022022115593803)
central environmental variable. In The Cambridge (doi:10.1093/cercor/9.6.521) 46. Bornstein MH, Putnick DL, Cote LR, Haynes OM,
handbook of environment in human development 30. Stiles J, Brown TT, Haist F, Jernigan TL. 2015 Brain Suwalsky JTD. 2015 Mother-infant contingent
(eds LC Mayes, M Lewis), pp. 58 –81. New York, and cognitive development. In Cognitive processes: vocalizations in 11 countries. Psychol. Sci. 26,
NY: Cambridge University Press. vol. 2 handbook of child psychology and 1272– 1284. (doi:10.1177/0956797615586796)
 47. Bjorklund DF, Myers A. In press. The evolution of middle childhood. Parent. Sci. Pract. 15, 295– 320. rearing. Psychol. Bull. 125, 223–254. (doi:10.1037/ 10
parenting and evolutionary approaches to (doi:10.1080/15295192.2015.1065120) 0033-2909.125.2.223)

rspb.royalsocietypublishing.org
childrearing. In Handbook of parenting 51. Joussemet M, Landry R, Koestner R. 2008 54. Bradley RH. 2004 Chaos, culture, and covariance
(ed. MH Bornstein), 3rd edn. New York, NY: A self-determination theory perspective on structures: a dynamic systems view of children’s
Routledge. parenting. Can. Psychol. 49, 194–200. (doi:10. experiences at home. Parent. Sci. Pract. 4,
48. Gould SJ. 1985 Ontogeny and phylogeny. New York, 1037/a0012754) 243–257. (doi:10.1080/15295192.2004.9681272)
NY: Belknap Press. 52. Attili G, Vermigli P, Roazzi A. 2010 Children’s social 55. Streiner DL. 2003 Being inconsistent about
49. Winnicott DW. 1956 Primary maternal preoccupation. competence, peer status, and the quality of mother- consistency: when coefficient alpha does and
In The maternal lineage: identification, desire, and child and father –child relationships. Eur. doesn’t matter. J. Pers. Assess. 80, 217–222.
transgenerational issues (ed. P Mariotti), pp. 300– Psychol.15, 23 –33. (doi:10.1027/1016-9040/ (doi:10.1207/S15327752JPA8003_01)
305. London, UK: Tavistock. a000002) 56. Venables WN, Ripley BD. 2002 Modern
50. Bradley RH, Pennar A, Iida M. 2015 Ebb and flow in 53. Holden GW, Miller PC. 1999 Enduring and different: applied statistics with S, 4th edn. New York, NY:

Proc. R. Soc. B 284: 20172168

parent–child interactions: shifts from early through a meta-analysis of the similarity in parents’ child Springer.
Downloaded from https://royalsocietypublishing.org/ on 18 November 2023