Activities and Interaetlens of Mothers and Their
Firstborn Infants in the First Six Months of
Life: Covariation, StabtUty, Continuity,
Correspondenee, and Prediction
Marc H. Bonutein and Catherine S. Tamis-LeMonda
covariation, stability, continuity, correspondence, and premction. Generally, mothers' activities did
this time period; some developmentally increased, and some developmentally decreased. Infants'
activities were unstable, but most increased over time. Specific mother and infant activities corresponded, and over time mothers and infants influenced one another in specific ways. In die critical
period of thefirsthalf year, infants appear to beflexibleand plastic in their behavioral repertoires
and are influenced by their mothers; mothers are somewhat consistent, but they also adapt to the
.. Jis with an endures, independent of later interactions bea period of tween the two and independent of stability in
, .or the infant, the infent In a second model, mother exerts a
the first
st months
months
are are
uniqueunique
influenc influence
h
ove
if
he l
ny
criticall experiences
i
in
i
later
l t point
i t iin
poin
n ddevelopment and independent
d aimedd to off stability
bili in
i h i f t In a third model,
study
H
auuu. la.o nature of mother-infant both early and late matemal inHuences
exert
and its consequences in the period independent ef&ets. In this study, we examof the dyad's initial accommodation during ined these three models of unique influences
thefirsthalf year of the infant's life by ex- between mothers and infents in the period of
amining unique and specific influences of thefirst6 months of life. Weflrstdefined a set
mother on infant and of infant on mother.
of salient interactive activities in mother-andseveral generic m o d e , of mutual effect,
^^1^^^^^^^
|££H!?Stra^;
of the mother, for example, uniquely affects
Domains of Mother and Infant
the infant at an early time point, and the consequent change in the infent thenceforward
Interaction
Everyday interactions between mothers
and babies are characterized by a melange
M.H.B. was supported by research grants (HD20559 and HD20807) and by a Research Career
Development Award (HD00521)fomthe National Institute of Child Health and Human Development; this paper was partially completed during M.H.B.'s tenure as a Guggenheim Foundation
Fellow and Professeur Invite at the l^boratoire de Psychologic du D6veloppement et de 1 EducaH°e'lbetn Scholarshfe and National Institute of Child Health and Human Development IHTA Fellowship. We thank G. Fitzmaurice, J. Suwalsky, J. Tal, and B. Wtight BefluestreprintsfomMarc H.
Bomstein, ChUd and Family Research, National Institute of Child Health and Human Development Building 31—Room B2B15, 9000 BockviUe Pike, Bethesda, MD 20892.
Bornstein and Tamis-LeMonda 1207
of activities. This study focused on prominent domains concerned with different orientations- of visual and vocal exchange. A
social domain describes interactions that, for
mother, encompass physical and verbal strate-
therefore that parents behave in consistent
gies used in engaging the infant interpersonally and, for the infant, have a corresponding
focus ofattention on mother. Many published
studies of mother-infant interaction have
concentrated attention on the social domain. Even within thefirsthalf year of life,
however, mothers and babies alike are increasingly concerned to distinguish and
incorporate into their interactions the world
outside the dyad (see, e.g., Belsky, Gilstrap, &
Kovine, 1984). A didactic domain of activities
describes interactions that tum outward from
and verbal ^toate'Jes*' u
°
f
stimulating the infant to engage the en
ment and, for the infant, they imply a
spKm mg ocus o a t t t i
_
,
1956; MacPhee, Ramey, & Yeates, 1984;
Rohner, 1985; Rothbaum, 1986). This view
builds on the infrequently tested assumption
that parenting reflects personality traits, and
ways across domains of interaction, time, and
context. The term covariation can be used to
describe consistency in rank-order status of
different parenting activities within individuals; covarying activities in parents would
therefore include those that parents perform
frequently relative to one another. Alternative
positions are that frequently performed activities are not psychologically linked, and that
individuals vary in the constellation and pattern of their activities such that there is no
monistic organization of parenting.
"
it has frequent
been suggested that the prosodic contou
sueh "motherese" (as opposed to adult conversational tones) might provoke or potentiate
infant attention, thereby promoting infant information processing (PapouSek, Papouiek, &
Bomstein, 1985). Infants, too, vocalize non-
meanin
ways.
;, the activities of mothers and
• •
'.meshed, so that differlways or
one an' investigators have operation_
3d the social and didactic domains as separable and significant, whether
they are called animate versus inanimate, affective versus informational, or social versus
object-centered (e.g.. Field, 1981; Goldfield,
1987; Sherrod, 1981; Stem, 1985). In this paper, we set out to assess several characters-
enting' activity, for example, would be „..„
that some parents perform relatively frequently when their infants are young and
again perform relatively frequendy when
their infants are older. The term continuity
can be used to describe consistency in the
absolute level of group performance of an activity over time. A continuous parenting activity would be one performed by a group of
parents with approximately the same frequency when their infants are young and old.
Stability of individuals and continuity in the
group are independent (McCall, 1981).
Characteristics of Mother and Infant
monistic view of parenting is that the overall
Despite the dynamic range of individual
activities that mothers naturally engage in
with their infants, many authorities, including
psychoanalysts and ethologists, have conceptualized parental (read: matemal) caretaking
as adhering to only one or a small number of
bipolar^ dimensions,^ cornmonly described as
"good," "sensitive," or "warm" (e.g., Brody,
overall level of development (see Maccoby &
Martin, 1983). However, increasing evidence
suggests that specific parental activities relate
concurrently and predictively to specific aspects of child performance (Bomstein, 1989b;
Bradley et al., 1988; Vibbert & Bomstein,
1989; Wachs & Chan, 1986). The term correspondence can be used to describe con-
1208
ChiM Deveiopment
sistency in the rank-order status of specific
parent and infant activities. In achieving
Home Observation Procedure
Home observationsfollowedthe same
kind of activity' relatively more often, more
appropriately, etc., can be expected to have
infants who perfomi a specific corresponding
activity relatively more frequently or who
possess a relatively higher ability of a particular kind.
identical to those we have used previously,
Briefly, mothers were asked tx) behave in their
usual manner and to disregard the observer's
presence insofar as possible; beside the observer, only mother and baby were present at
home; and observations took place at optimal
f the day for individual babies '
1985; Tamis-LeM(
Studies of parent-cl
sampling technique in which a 0»-^^^ »^^^.
Present Study
vation period was followed by a 30-sec re-
The present study was designed to assess
--'ation, stability and continuity, and coni;m,c.it and predictive correspondences in so-
cording period (Seitz, 1988). The bounds of
each period were signaled to the observer by
a covert automatic timer,
cial and didactic domains of activity and in
different modes of vocalization in mothers
and infents,first,as independent issues in
The 2-5-month data reported here were
excerpted itora a larger ongoing longitudinal
early development and, second, for the critical role they play in differentiating unique
and specific models of early interaction The
study began when infants were 2 months of
study. Four matemal activities and five infant
activities were subjected to analysis. Two
codes recorded the mother's engagement of
her infant and organization of infant attention
age because ofthe intentionality and flexibil- socially (i.e., to the mother herself) and di-
ity in behavioral organization thatfirstemerge
at about thistime(e.g., Emde, Gaensbauer, &
Harmon, 1976). Mothers and infents were revisited at 5 months because by thistimethe
baby's scope of apperception has considerably broadened beyond the dyad, infants look
It and reach out and grasp,
•
dactically (i.e., to some property, object, or
event in the environment). Mothers could do
these by physical or verbal means and by
inti-oducing a new topic or elaborating on
a topic already introduced. Thus, a mother
might touch, gestiire toward, or position her
infent with the expUcit purpose of engaging
'1-baby to herself, or she might demonstrate,
ir^fant's visual and/or tactiial exploration of
codes assessed speech to the
s infentregister(characterized
two home observations. Dyads werefirstseen
jhree of thefiveinfknt activities coded
when the infants were 2 months of age (M =
orientation of visual exploration to mother or
66 days) and again when 5 months of age (M to a property, object, or event in the environ= 161 days). Infents were term at birth (M ^ent and active hand contact with an object,
weight = 3.39 kg; M length = 51.2 cm) and xhefinaltwo codes a d
ifet
li
were healthy throughout the course of the uon a i h
di
study. Households were middle- to upper- initiat
SES (M
59 n the Hollingshead Four the i
5 G f r d 1985) D d
fton,
During the course of the study, 25% ol
the visits were independently coded by two
observers. Reliability (Pearson r) on the nine
Bomstein and Tamis-LeMonda 1209
categories of mother
spoke to their 2-month-olds in the infant register in as few as 6% of the intervals of the
observation period, and some in as many as
ic mothers never spoke to their 5-
le 5-month-olds vocalize
Hollingshead SES) was systematically related
to any ofthe measured variables, even though
each showed variation; moreover, obtained
relations held equivalently for boys and girls.
Statistical analyses collapsed across these fac-
hers engaged th
at 2 and at 5 months, and the comparison of
pattems of covariation between the two ages;
infants rather equally
stability and continuity of mothers' and of in-
tween 2 and 5 months, therf
changed the emphasis of thei
correspondence between mothers' and in-
significantly reducing interpersonal
son"ofp^attem"s''of''these^concurrent''c"mla-
tions between the two ages. These analyses
he evaluation of unique 2- to 5•2-and5-r
'
Mi
M: M
iJlili
STABILITY AND CONTINUITY: FREQUENCY OF 1VlOTHEBS ANDINFA
NTS ACTIVITIES AT 2 AND 5 MONTHS
CONTINUITY
2 Months
r
M
(Range)
M
(Range)
1
1212
ChUd Development
portional emphasis on
In infants.—Even though infants tended
to be unstable in their individual ranks over
thefirst6 months. Table 2B shows that infants
increased significantly in their orientation to
the environment between 2 and 5 months. At
2 months, infants explored equivalendy in so.,
cial and didactic orientations; al '
'
however, theyfevoreder
At 2 months, infants vocalized nondis-
daetic plus social foci
months, r = .48, p = .0
= .53, p = .0"••
couragement c. . . . . —
—
daetic (as opposed to the soeial) mode re:
positively to infent tactual exploration al'
' - " " - "" " — " - --• • "
tenuating so
tress and distress equivalently during the
distress
vocalization
had
significantly
in-
creased. An analysis of this divergence failed
to show a SigniBcant difference across age.
^^^^ - 10 < r s < 09 However mothers-
Table 3 shows concurrent relations be
tween organization-of-stimulation variables in
mothers and orientation-of-exploration vari-
appropriately in developmental
their in&nts' changing fbcus of ex
with attention to mother at 2 moi
using the infent register tended to correlate
ables in infants. Mother-infant interacti
Bornstein and Tamis-LeMonda 1213
r,
d
at
<
eial exploration at 5 months (Set la), matemal
didactic encouragement at 5 months alone
shared unique variance with infant didactic
exploration (Set 2a), and matemal vocaliza•
in the infant register a' "
-
'
second line provides evidence for the unique cipal forms of
o
,
predictive association between members of and didactic, as well as for matemal speech in
the dyadfrom2 to 5 months. These equations the infant register at 5 months. With respect to
provide solutions after both 2- to 5-month sta- Model 1, no infant activity uniquely and
bility in the individual and any 5-month con- • alone infiuenced matemal activitiesfrom2 to
social exploration alone shared unique variance with mothers' social encouragement at 5
months (Set la), and infant didactic explora-
tion and nondistress vocalization each alone
J ... —
shared unique variance with matemal dinth predictive influence daetic encouragement at 5 months (Sets 2a,
IS, Model 1 is supported d). Finally, withrespectto Model 3, in"
'
_, - . ,
..__ ... which line 2 alone nondistress vocalization, ta
(unique early effect) is significant (e.g.. Set 3a
in Table 44); Model 2 is supported by equation sets in which line 3 alone (unique con....
current effect) is significant (e.g.. Set la); and, (Sets 3c
Model 3 is supported by equation sets in
which both lines 2 and 3 are signiBcant (e.g..
Set 3b) In summary the last two lines in each
setrefiectsolutions for separate and indepen-
Few behaviors of mothers or of infants
dent families of predictive equations each
with three predictor variables.
covaried positively, suggesting that mothers
and infants alike tend to specialize in particular kinds of activities. Notably, social and di-
ports*'moAet°to"n'fant''reTults""for the thr'ec
principal forms of infant exploration (social, didactic, and
tactual) and for
infant
purposes, several (but not all)
re'utedTt''ti
social are ni
daetic, and
ss vocalization at 5 months. For
-,-,—jve
pendence
necessarily
i
matemal predictors are presented; matemal
vocalization in adult conversational tones was
doubt on a :
example, br
these infant activities at the zero-order level,
for example, and so it was not included. With
respect to Model 1, only matemal didactic encouragement at 2 months alone predicted
unique variance in infant tactual exploration
hs (Set 3a). With respect to Model 2,
.
... .^
^
over thefirsthalf year of their in.
(see, too, Belsky et al., 1984; Clarke-Stewart &
Tk
Richards'", "wTJ^ Gottfted,' 1984^
Bornstein and Tamis-LeMonda 1215
Wachs, 1987). Mothers' aetivities varied in
continuity, however. Notably, matemal social
engagement decreased, while didactic engagement increased (see, too, Belsky et al.,
predictiverelationsfollowed. Only infent tactual exploration and nondistress vocalization,
and only matemal vocalization in the infent
register, were uniquely predictedfiram2 to 5
1984). Infents, by contrast, were unstable, although theyregularlyincreased in exploration and in nondistress vocalization (Belsky et
al., 1984; Moss, 1967).
months. Mother and infent activities shared
unique variance more at 5 than at 2 months,
and corresponding relations between partners
predominated, such as mother and infent
social and didactic activities and mothers'
speaking in the infent register and infents'
nondistress vocalizing. With respect to presults overshadowed Model 1, and significant
in tne intantreinsterland ones '^° * ^ predictive; Model 3 relations are
ot (social enea^merit) exerted common, and if an association is predictive it
fater direiopmental effects over infants, will be concurrent as well.
Thus, matemal activities thatfluctuatein
time
Discontinuities characterize infant devel(those that are neither stable nor continuous) opment in thefirstyear of life (e.g., Emde et
are not automatically random or develop-
al., 1976; Fischer, 1980); presumably, if our
mentally ineffectual; rather, they may actually
display systematic change in level and in pattem in relation to developmental changes in
the infent, and on this basis harbor significant
consequences for infknt development.
Mothers and infants specialize, and their
specializations match; that is, mother-infant
interactions even in thefirstmonths can be
observationtimeshad straddled periods of
reorganization in infency, other pattems of resuits might have emerged. Also, the results
reported might apply uniquely to the open
style of interaction we studied but not to other
'Vf'"' unstrained situations. The families in
*'« ^f^^y wererelativelyrestricted in terms
"f socioeconomic status and educational his-
described as mutually corresponding. More-
pattems
^'V-
different
over, the domains of stimulation in mothers
and of exploration in infents relate to one another in increasingly specific ways. In the age
period studied, niatemal didactic stimulation
of
resulte
could
^"^S^ i" mothers (and conceivably in in™ ) '^°'™m8fromotherregionsof the SES
f^f^hus, the ~«»™1i.«Ki1iK, ^f M,...
fndings might b.
pnaie to lnianis aeveiopmenc. t^orrespon
dences in the longitudinalfindingslend additional weight to the specificity hypothesis.
mothers with their infknts. Moreover, n
" 1 « " i l ! ' ' ' / ' • ^ ' T ' / ' * " " ^ ' '1^^"^^'°
open to one anothers influencefroman early
period in the infant's life. The instability of
their individual behaviors over the same time
period may indicate additionally that mothers
and their new babies are to a certain degree
flexible, plastie, and adaptable. The predictiveresultsbear out both of these inferences.
Longitudinal data even in thefirst6 months
provide some evidence for unique early,
unique later, and combined early and late ex-
Theoreticians andresearchershave long
supposed that the child's earliest interactive
experiences may affect the course of later development (Plato, ca. 350 B.C.), and the importance and pervasiveness of infant effects on
caretakers are generally wellrecognized(e.g.,
Bell & Harper, 1977). Our analyses converge
to support such a mulUvariate and multimodel view of specificity in mother-infent
mutual interactions in theflrsthalf year ofthe
perience effects between mothers and infants.
infent's life.
The results ofthe predictive models can
which behaviors of infents and mothers were
predicted and, second, what pattems existing
Pennsylvania In&nt and Family Development
guage. Applied Psycholinguistics, S, 267-2J
Dttfiied, A. W. (Ed.). (1984). Home environm,
and early cognitive developtnent. Orlando, I
Development!55, 692-705.
jointly contribute to developing cognitive comtionalAcaderrw of Sciences, 82, 7470-7473.
lomstein, M. H. (1989a). Sensitive periods in
of SocicLl Status. UnpubUshed manuscript!
Yale University, New Haven, CT.
Psychological Bulletin,
mothers and infents. Decelopmental Psychol-
tion in the context of the family; Parent-child
interaction. In E. M. Hetherington (Ed.), P. H.
Mussen (Series Ed.), Handbook of child psychology: Vol. 4. Socialization, personality, and
social development (pp. 1-101). New York:
Wiley.
MacPhee, D., Ramey, C, & Yeates, K. (1984). The
and early mental development (pp. 343-370).
realms of development: A proposed integra-
ChUd Devehtmient, 52,1-12.
Developmental Psychology, 17, 127-14S.
;ohen, J., & Cohen, P. (1983). Armlied multiple
ty and variability in parental speech. Ir
tions on the developing relationship hetween
mother and hahy in the neonatal period. In
H. H. Schaffer (Ed.), Studies in mother-infant
MonograpAs, 112,435-458.
hiohehavioral study. Psychological h
s.). Developmental psychol-
during early interactions. In L. P. Upsitt (Ed.
Advances in infancy research (Vol. 1, pp. 57
100). Norwood, NJ: Ablex.
tual development: The special ea
stimuli. In M. E. Lam' ' '
Bornstein and Tamis-LeMonda
Habituation and maternal encouragement of at-
gregated and nonaggregated measures of pa-
tendon in infancy as predictors of toddler Ian-
rental behavior. Child Development, 58, 796-
guage, play, and representational competence.
Child Development, 60, 738-751.
797.
Wachs, T. D., & Chan, A. (1986). Specificity of en-
Menlo Park, CA: Addison-Wesley.
'ibbert, M., & Bomstein, M. H. (1989). Specific
"
correlates of infarits' communication performance. Child Development, 57, 1464-1474.
ween domains of mother-child
toddler referential language
iy. Infant Behavior and Devel-
1217
Wachs, T. D., & Cruen, C. E. (1982). Early expe-
rience
Plenum.
and human development. New York: