The Role of Culture in Moderating the Links Between Early Ecological Risk and Young Children’S Adaptation

The role of culture in moderating the links between early ecological risk and young children’s adaptation

RUTH FELDMANa and SHAFIQ MASALHAb aBar-Ilan University, Israel; and bTel-Aviv University

Abstract To examine the effects of risk on infant development within cultural contexts, 141 dual-earner Israeli and Palestinian couples and their first-born child were observed at 5 months and again at 34 months. Eight ecological determinants were examined as potential risk factors, including the infant’s observed and parent-reported difficult temperament; the mother’s depressive symptoms, work–family interference, and experience of childbirth; the parents’ marital satisfaction and social support; and observed maternal and paternal sensitivity. Symbolic play and behavior problems were assessed at 34 months. Culture-specific effects of risk and protective factors were found. Parent sensitivity facilitated symbolic competence to a greater extent in the Israeli group. Culture moderated the effects of maternal depression and family social support on toddlers’ behavior problems. Maternal depressive symptoms had a negative impact on the behavior adaptation of Israeli children and social support buffered against behavior problems in the Arab group. Implications for research on risk and resilience and the role of culture in moderating the effects of ecological risk are discussed.

Among the central questions in the study of the role of culture in shaping child adaptation developmental psychopathology is how risk has received surprisingly little attention. Cul- conditions, stemming from child, parent, con- ture, as a set of beliefs, attitudes, practices, text, or culture, coalesce into a significant and behaviors pertaining to child rearing and enough risk that alters the child’s develop- the family, exerts the most significant impact ment to a maladaptive trajectory. The specific on the infant’s rearing environment ~Kagan, parameters that add up to a substantial risk, 2001; Keller, 2003!. Yet, the contribution of the way contexts exacerbate or attenuate the culture to the child’s propensity for psychopa- effects of risk on child outcomes, and the level thology remains largely unknown. at which independent risk factors cross a crit- This study examines the role of culture in ical cutoff are issues of theoretical and clini- moderating the relations between eight indi- cal importance ~Belsky, 1998; Cicchetti & ces of risk in the infant’s ecology and two Lynch, 1993; Rutter & Sroufe, 2000!. Al- types of developmental outcomes in young though research has addressed the cumulative children: symbolic competence and behavior effects of parent, child, and contextual risk, adaptation. Specifically, we examined two sets of culture-related hypotheses with regard to The Israeli–Palestinian Project was supported by the New- the relations of risk and development. The Land Foundation. The study is dedicated to the memory first set considered culture-specific links be- of the late professor Donald Cohen, MD, whose vision tween infant, parent, and contextual deter- and support enabled this project. minants and child outcomes. Theoretical Address correspondence and reprint requests to: Ruth Feldman, Department of Psychology and the Gonda Brain perspectives on development within cultural Sciences Center, Bar-Ilan University, Ramat-Gan, Israel ecologies suggest distinct pathways to the 52900; E-mail: [email protected]. emergence of competence and adaptation in

1 2 R. Feldman and S. Masalha cultures guided by different orientations Culture is perhaps the most neglected topic ~Greenfield, Keller, Fuligni, & Maynard, 2003; in the study of risk and resilience. To date, Kagˇitçisi,, 1996; Rogoff, 2003; Shweder, 2003!. few studies examined the differential effects The second set of hypotheses was informed of risk conditions on developmental out- by the concept of “cumulative risk.” Models comes as they emerge within cultural contexts on risk and resilience theorize that the cumu- ~Shweder et al., 1998!. A recent survey of the lative number of risk conditions in the infant’s literature on infant mental health indicates that ecology is the determining factor in shaping 94% of the studies come from North America child maladaptation ~Cicchetti & Toth, 1998; or Europe ~Tomlinson & Swartz, 2003!, a sit- Rutter, 1987; Sameroff & Fiese, 2000!. This uation that not only prevents the implementa- “cumulative risk” effect is suggested to be tion of culturally sensitive interventions but universal and has been observed in studies of also seriously skews our understanding of hu- Western and non-Western cultures ~Sameroff, man development. Because culture provides Seifer, Barocas, Zax, & Greenspan, 1987; the organizing framework for the child’s ecol- Werner & Smith, 1979!. We thus made an ogy, the meaning, saliency, and impact of spe- attempt to address both the culture-specific cific risk conditions are likely to vary across and universal contributions of risk factors in cultural communities ~Celia, 2004!. infancy to domains of young children’s com- The effects of culture on the architecture of petence and adaptation in a community sam- risk and protective factors may stem from sev- ple of Israeli and Palestinian couples and their eral sources. First, if one considers the central first-born child. dichotomy between cultures, phrased in terms Ecological and transactional perspectives of individualism versus collectivism ~Trian- suggest that infant development occurs within dis, 1989! or independence versus interdepen- a growing matrix of proximal and distal influ- dence ~Markus & Kitayama, 1991!, it is clear ences ~Belsky, 1984; Bronfenbrenner, 1977!. that these dichotomous worldviews and life- These include the child’s biological disposi- styles determine the centrality of the social tions; the parent’s personality, experiences, and context to the growing child. Social support the parent–infant relationships; the larger cir- has been shown to exert a greater impact on cle of social networks; and the more distal maternal functioning and child adaptation in effects of cultural attitudes and philosophies, cultures guided by a more collectivistic orien- which in turn, define parental attitudes, living tation ~Cutrona, Russell, Hessling, Brown, & conditions, and interpersonal patterns. Each Murry, 2000!. The effects of social support on ecological factor influences and is influenced child adaptation may thus vary according the by other factors at both the same ~e.g., moth- culture’s collectivistic philosophies and living ering on fathering! and different ~e.g., infant conditions. Second, cultures define the param- dispositions on parenting! levels of the eco- eters of health and pathology, delineate ac- logical ring. In combination, this network of cepted from nonaccepted behavior, and offer bidirectional influences forms a mutually reg- living conditions that may intensify or miti- ulating dynamic system that continuously gate the effects of parental psychopathology shapes developmental outcomes ~Lynch & Cic- on the growing child ~Shweder, 2003!. For chetti, 1998; Sameroff, 1997!. Each ecologi- instance, maternal depression, which is closely cal determinant at the level of child, parent, or related to social support, particularly among context may, at some point or in certain com- Arab women ~Stuchbery, Matthey, & Barnett, binations, turn into a risk condition. Consis- 1998!, may have a different effect on child tent with the formulations of dynamic systems’ adaptation in the context of nuclear versus theory, when a risk condition occurs at critical extended living arrangements. In extended fam- nodes or within specific contexts its effect may ily dwelling, where several women raise chil- be magnified, and even small differences in dren together ~Sharma & Fischer, 1998!, the levels of risk can make a substantial impact effects of maternal depressive symptoms may with repeated iterations ~Fogel, 1993; Fogel be attenuated by the infant’s daily interactions & Thelen, 1987!. with other women of a kin relationship. Third, Risk, adaptation, and culture 3 collectivistic societies, which view deference 2003! and predicts lower cognitive and sym- to authority and the reduction of aggression as bolic skills ~Feldman, Eidelman, & Roten- central parenting goals ~Briggs, 1972; Hof- berg, 2004; van Bakel & Riksen-Walraven, stede, 2001!, may consider the infant’s tem- 2002!. As to the controversy of whether peramental disposition to negative emotionality maternal report is a valid measure of infant as more disruptive. Fourth, cultural beliefs and temperament ~Seifer, Sameroff, Barrett, & attitudes shape the parents’ interactive behav- Krafchuk, 1994!, observed and self-reported ior in ways that are consistent with the culture’s difficult temperament are likely to exert sep- overarching goals and meaning systems ~Le- arate and cumulative effects on infant devel- Vine, 2002; Super & Harkness, 1986; Whiting opment. Inborn dysregulation limits the infant’s & Whiting, 1975!. In individualistic societies, capacity to extract formative experiences from parent–infant interactions are built on a face- the environment during critical periods for the to-face position and involve patterns of covo- development of self-regulation ~Tucker, 1992! calization, mutual gazing, and object whereas maternal perceptions create the emo- presentation, behaviors that mark the dialogue tional context in which infants form internal- between separate individuals who coordinate ized models of self, other, and relationships their interpersonal exchange. In collectivistic ~Stern, 1995!. societies, interactions are framed by continu- Maternal postpartum depression is a prev- ous contact ~Tronick, 1995!, which reflects alent mother-related risk, affecting 10–12% the inseparateness of mother and child ~Le- of women of childbearing age ~Burt & Stein, Vine, 2002!. Patterns of visual and vocal co- 2002!, and its negative effects on children’s ordination have been shown to promote self- cognitive skills and behavior adaptation are regulation and competence in children growing well known ~Goodman & Gotlieb, 1999; Wein- in individualistic societies, not among those berg & Tronick, 1998!. Maternal depression reared in collectivistic contexts ~Keller et al., decreases the quality of child play ~Easter- 2004!. As such, the study of risk and resil- brooks, Biesecker, & Lyons-Ruth, 2000; ience may benefit from including culture as a Murray et al., 1999! and predicts more behav- research parameter and from assessing Risk ϫ ior problems ~Burt et al., 2005; Carter, Garrity- Culture interactions. Interaction effects that Rokous, Chazan-Cohen, Little, & Briggs- may be important to the continuity from in- Gowan, 2001; Feldman & Eidelman, 2004!. fancy to the preschool years are the inter- In the present study, we included two addi- action of culture and social support, culture tional mother-related risks that shape the and maternal depression, culture and the mother’s world in the postpartum period and infant’s negative emotionality, and culture and are related to depressive symptoms: work– the parent–infant behavioral repertoire. family interference and the experience of childbirth. Because of the large rise in rates of maternal employment in recent years, most Ecological Risk at the Transition mothers of young infants are currently em- to Parenthood ployed ~Han & Waldfogel, 2003!, and maternal work–family interference is an important Risk conditions in infancy are generally feature of the infant’s ecology ~Voydanoff, thought to stem from three sources: child, 1989!. Mother work–family interference is as- parent, and context ~Belsky, 1984!.Ofthe sociated with higher depression, lower adap- infant-related risks, the infant’s dysregulated tation, and less marital support, particularly at temperament is among the most frequently the transition to parenthood ~Feldman, Suss- studied biological risk. In relation to the de- man, & Zigler, 2004; Hyde, Klein, Essex, & velopmental outcomes studied here, infant dys- Clark, 1995; Schwartzberg & Dytell, 1996!. regulation plays a role in the consolidation of An additional mother-related risk is the expe- behavior problems from toddlerhood to adult rience of childbirth and the rate of recovery, life ~Belsky, Hsieh, & Crnic, 1998; Caspi, especially among first-time mothers. A nega- 2000; Rubin, Burgess, Dwyer, & Hastings, tive childbirth experience correlates with lower 4 R. Feldman and S. Masalha maternal adaptation, slower entrance into the petencies develop along the line of increasing maternal role, and less optimal mother–infant symbolic distance. Symbolic play has been relationship ~Feldman, Sussman, et al., 2004; studied as a developmental function related Goldstein-Ferber & Feldman, 2005!. to, but distinct from, both general cogni- Contextual risk and support are typically tive skills and language competence ~Bates, examined in relation two sources: the marital O’Connell, & Shore, 1987; Tamis-LeMonda relationship and the larger social context. The & Bornstein, 1994! and its maturation follows quality of the marital relationship and the de- a similar line from concrete to decontextual- gree of social support are important contex- ized expression ~Fenson & Ramsay, 1980; Mc- tual assets, associated with higher maternal Cune, 1995!. During the third year, symbolic sensitivity, lower maternal depression, and play becomes more complex, play sequences lower child negative emotionality ~Cowan & grow more elaborate, and individual variabil- Cowan, 1992; Cutrona & Troutman, 1986; ity in symbolic complexity reflects both cog- Pauli-Pott, Mertesacker, & Beckmann, 2004!. nitive growth and emotional well-being ~Noll In terms of the link to symbolic competence & Harding, 2003; Slade & Wolf, 1994!. and behavior adaptation, marital satisfaction In addition to its links to maternal depres- was found to predict the complexity of sym- sion, infant dysregulation, and social support, bolic play in preschoolers ~Keren, Feldman, children’s symbolic expression is shaped by Namdari-Weinbaum, Spitzer, & Tyano, 2005!, the cultural context. It has been suggested that and is associated with less behavior problems every society must prepare its children to com- in the childhood years ~Seiffge-Krenke & Koll- prehend and use symbols, but cultures employ mar, 1998!. Similarly, social support buffers different methods to reach that goal ~Rogoff, against the development of child behavior 2003; Vygotsky, 1978!. In more traditional so- problems ~Franz, Lensche, & Schmitz, 2003; cieties, children learn tool use through guided Kroes et al., 2002!, and the role of contextual participation and copractice with more profi- support is especially strong for mothers in col- cient adults, whereas parents in individualis- lectivistic societies, where child care by a kin tic societies facilitate symbolic growth through is a common practice ~Sharma & Fischer, play that focuses on verbal and affective ex- 1998!. The effects of risk on family adap- change ~Rogoff, Mistry, Goncu, & Mosier, tation increases during the transition to par- 1993!. Thus, with regard to the culture-specific enthood, a period when bidirectional links pathways to symbolization, parents in indi- between risk factors are tighter and their im- vidualistic societies may support symbolic pact on family functioning is more salient skills through behavioral coordination with the ~Belsky & Pensky, 1988; Heinicke, 1984; infant communicative signals ~Feldman & Levy-Shiff, 1994!. Furthermore, because the Greenbaum, 1997; Slade, 1987!. Symbolic and transition to parenthood is a significant period linguistic capacities in traditional societies, par- from a cultural perspective, marking the trans- ticularly in Arab societies, are more closely mission of the cultural heritage to a new gen- linked to the mother’s social support networks eration, the effects of cultural practices, support ~Eapen, Zoubeidi, & Yunis, 2004!. Still, the networks, and relational patterns on the fam- correlations found between maternal verbal ily ecology may be more pronounced. responsiveness and child play for Egyptian toddlers ~Wachs et al., 1993! suggest both main and moderating effects in the relations of par- Developmental Outcomes ent sensitivity and symbolic complexity. As to in Young Children the universal hypothesis, because symbol use Between the age of 2 and 3 years, the infant’s is among the defining features of the Homo world expands as language develops, mastery sapiens, it is likely that under normative rear- of symbol use grows, and the self emerges as ing conditions all children would develop ad- the locus of internal thoughts and feelings ~Cic- equate symbolic skills and play would be chetti & Beeghley, 1990; Kagan, 1981; Piaget, organized by a similar process ~Bornstein, 1962!, abilities that indicate that cognitive com- Haynes, Pascual, Painter, & Galperin, 1999!. Risk, adaptation, and culture 5

Similarly, lower levels of symbolic play are lations of risk conditions in infancy and the likely to be associated with higher levels of development of symbolic play and behavior cumulative risk. adaptation. Participants were dual-earner Is- In parallel to the maturation of symbolic raeli and Palestinian couples and their first- capacities, children begin at this age to dis- born child observed at 5 and 34 months. The play behavior problems that clearly deviate age of 5 months was selected because infants from the culturally accepted norms and pre- at this age are already familiar with the culture- dict lower adaptation in later childhood and specific patterns of parent–infant interaction adolescence ~Caspi, 2000!. Toddlers’ behav- and the family has typically recovered from ior problems are related to risk conditions stem- the experience of childbirth. The age of 34 ming from each level of the ecology, including months, the last quarter of the third year, is a difficult infant temperament, maternal depres- point when the child’s symbolic play reaches sion, low social support, and less optimal a certain complexity and behavior problems parent–child relations ~Dawson et al., 2003; are beginning to emerge in the externalizing Holden & Ritchie, 1991; Mahoney, Jouriles, and internalizing domains. Eight ecological & Scavone, 1997!. Studies of children’s be- determinants were examined as potential risk havior problems in cultures guided by individ- conditions and independent, cumulative, and ualistic and collectivistic philosophies indicate interactive effects were tested. Consistent with that the underlying structure of externalizing the ecological perspective ~Belsky, 1984; Bron- and internalizing symptoms, their associa- fenbrenner, 1977!, we hypothesized that risk tions with ecological risk factors, and the va- originating in the infant, parent, and context lidity of the instrument are retained across a would predict less optimal outcomes in the wide range of cultures ~Bengi-Arslan, Ver- toddler stage, that culture would moderate the hulst, van der Ende, & Erol, 1997; Chang, effects of risk on development, and that risk Morrissey, & Koplewicz, 1995; Heubeck, factors would exert a cumulative impact on 2000; Javo, Heyerdahl, & Ronning, 2000; Tra- maladjustment. montina, Schmitz, Polanczyk, & Rohde, 2003!, Eight determinants of the infant’s ecology including the Israeli and the Palestinian soci- were examined as predictors of toddlers’ sym- eties ~Auerbach, Yirmiya, & Kamel, 1996!. bolic skills and behavior adaptation. These in- Among Arab toddlers, the family’s social sup- cluded infant observed and parent-reported port is an important determinant in the con- difficult temperament as the child’s factors; solidation of behavior problems ~Eapen, Yunis, maternal depression, work–family interfer- Zoubeidi, & Sabri, 2004!. Thus, with regard ence, and the experience of childbirth as the to the culture-specific pathway, it is possible mother’s factors; mother–infant and father– that social support may have a stronger buff- infant interactions as the observed indices of ering effect on child adaptation in cultures the microsystem; and the parents’ marital sat- organized by multigenerational living condi- isfaction and social support as the contextual tions where support availability is high. In variables. Children’s symbolic capacities and addition, infant dysregulated temperament may behavior problems were selected as the age- have a more negative effect on the emergence appropriate cognitive and social–emotional of behavior problems in societies that stress outcomes because of their close dependence deference to authority ~Hofstede, 2001!.In on environmental conditions. It was expected terms of the universal hypothesis, it is likely that difficult temperament, maternal depres- that higher levels of behavior problems would sion, work overload, and negative childbirth correlate with more cumulative risk across experience, low parent sensitivity, and less mar- cultures. ital and social support would predict low symbolic competence and more behavior problems. The “cultural pathways” perspective The Present Study ~Greenfield et al., 2003! was adapted to the In light of the above, the present study exam- study of developmental risk and guided the ined the moderating role of culture on the re- culture-specific hypotheses. According to this 6 R. Feldman and S. Masalha perspective, the parenting experiences offered emergence of symbolic competence and be- by the culture differentially support child com- havior adaptation. The parents’ visual and vo- petence and adaptation and parents use the cal coordination with the infant’s signals were resources available in their ecology to achieve expected to have a stronger impact on the emer- developmental milestones. Marked differ- gence of symbolic play in the Israeli group. ences in parental attitudes, believes, and child As to behavior problems, several culture- rearing practices exist between the Israeli and moderated effects were hypothesized. Social Palestinian societies, which are primarily re- support was expected to have a more positive lated to the individualistic and collectivistic impact in the context of extended-family set- orientations of the two cultures, as we have tings and to have a stronger buffering effect found for this sample at 5 months ~Feldman, on the development of behavior problems Masalha, & Nadam, 2001!. The Israeli society among Arab children; maternal depression was is primarily urban, and its orientation is clearly expected to have a stronger effect in the con- individualistic. Young couples live in nuclear text of nuclear-family living and thus to have family arrangements, gender-role philoso- a more negative effect on Israeli children; and phies are more egalitarian, and parenting goals infant negative emotionality was expected to are framed in terms of autonomy and self- have a more negative impact on the develop- actualization ~Feldman, 2000; Israel Central ment of behavior problems in Arab children. Bureau of Statistics, 2002!. Patterns of parent– infant interactions revolve around mutual gaz- Method ing, covocalization, and active involvement with toys, behaviors that support the develop- Participants ment of competence in individualistic societies ~Keller et al., 2004!. The initial cohort of participants included 100 The Arab–Palestinian society is collectiv- Israeli–Jewish couples ~parents of 52 boys and istic in its orientation, and several aspects char- 48 girls! and 62 Arab–Palestinian couples ~par- acterize its specific version of collectivism. ents of 33 boys and 29 girls! and their first- According to the Abudabbeh’s ~1998! model, born child. All parents in the study graduated Arab family life is defined by three main fea- high school, with the vast majority holding a tures; adherence to traditional gender-role vocational or academic degree ~83%!. Only philosophies even among modern couples ~Ab- dual-earner families were recruited in the two dalla, 1995; Mar’i & Mar’i, 1985!, patriarchal societies, and all participating mothers and hierarchies of power that stress deference to fathers were employed in skilled or semi- authority, and the superiority of the family skilled professions. Israeli mothers were on goals to those of the self. Extended family average 27.7 years old ~SD ϭ 3.93!, and Arab living, which stresses the inseparateness of mothers were on average 25.65 years ~SD ϭ the young family from the family of origin, 3.19!; the age difference was significant, the deemphasis on motivations of self- F ~1, 161! ϭ 11.81, p , .01. Israeli mothers fulfillment, particularly among women, and completed an average of 15.25 years of edu- the emphasis on child compliance as a central cation ~SD ϭ 1.69! and Arab mothers 14.68 parenting goal are expressions of these as- years ~SD ϭ 1.35!, but the difference was not pects ~Al-Haj, 1989; Dwairy & van Sickle, significant. Israeli fathers’ age averaged 30.37 1996; El-Islam, 1983!. Similar to other tradi- years ~SD ϭ 4.99!, and their education aver- tional societies, early interactions are based aged 14.54 years ~SD ϭ 1.64!. Arab fathers’ on physical contact between parent and child age averaged 30.35 years with an average ed- rather than on patterns of visual and affective ucation of 14.19 ~SD ϭ 1.85! years, and no coordination ~LeVine, 2002; Tronick, Morelli, significant age or education differences were & Ivey, 1992!. found for fathers. The aforementioned differences in parent- Mothers in the two societies had all worked ing practices and living conditions were ex- prior to childbirth, were on maternity leave, pected to chart different pathways to the and returned to work when the infant was be- Risk, adaptation, and culture 7 tween 12 and 16 weeks, with no cultural dif- groups for any variable, and data were col- ferences in the length of leave. None of the lapsed into a single group. The study was fathers reported taking a paid parental leave. conducted between 1996 and 2000 during a Thirty-eight percent of Israeli mothers and 42% relatively peaceful period in the region. of the Arab mothers were employed in semi- Home visits were made by members of each skilled professions ~secretaries, teachers’ as- cultural community who were trained to con- sistants!, 15% of Israeli mothers and 13% of duct the study in Israel and Ramallah, and Arab mothers were self-employed, and the rest periodic meetings between groups were held. worked as skilled professionals ~teachers, com- In the first visit Israeli infants were an average puter technicians, physicians!. Among fathers, of 20.51 weeks ~SD ϭ 3.14!, and Arab infants 18% of Israeli fathers and 21% of Arab fathers were an average of 21.32 weeks ~SD ϭ 3.98! were employed in semiskilled professions, 22% with no age difference. All infants were born of Israeli fathers and 26% of Arab fathers were at full term, had birthweights of 2700 g and self-employed, and the rest were employed in above, and were in good health since birth. skilled professions. No cultural differences were Of the original 100 Israeli families, 86 fam- found in the professional status of men and ilies were seen again when the child was 34 women or in the family’s economic status, and months ~44 boys, 42 girls!. Among Arabs, 55 all couples considered themselves middle class families were seen again ~29 boys, 26 girls!. according to their society’s standards. Israeli toddlers were on average 33.74 months Israeli families were living in the greater ~SD ϭ 4.43! and Arab toddlers were 34.32 Jerusalem and Tel Aviv areas and were re- months ~SD ϭ 4.18!, and there was no group cruited by nurses in well-baby stations in their difference. No differences were found be- immediate neighborhoods. Nurses introduced tween families who were revisited and those the study to those who fit the study criteria who were not on any of the measures col- after screening the clinic’s records for paren- lected at 5 months. Attrition was mainly re- tal and infant physical and mental health com- lated to inability to locate the families or to plications. Of a pool of 110 considered as the family moving to a far away location. All potential candidates and approached, 100 couples in the study were still married by the agreed to participate. Ten mothers declined, time their child reached 34 months. In 60% of citing father refusal or scheduling difficulties the Israeli families and 62% of the Arab fam- as reasons. The well-baby clinic records of ilies a second child was born by the time the these women did not show differences in de- first child was 34 months old, with no cultural mographic or health variables compared to differences in family composition. Change in the participating families. maternal employment status between 5 and 34 Arab families were recruited in the greater months were as follows: among the Israeli area of Nazareth in the northern part of Israel mothers 27% mothers reported a decrease ~n ϭ 28! and of Ramallah in the West Bank and 19% reported an increase in the amount ~n ϭ 34!. Recruitment and data collection were of out of home hours. In the Arab group 25% conducted by professionals who were mem- reported an increase and 20% reported a de- bers of each cultural community. Seven fami- crease in work hours, and no cultural differ- lies approached declined participation, and ences were found in the proportion of change. these were of similar background as the par- Overall, Israeli mothers were employed 33.4 ticipating families. Of the Arabs, 90% were hr0week ~SD ϭ 6.55! and Arab mothers were Muslims and 10% were Christians, with no employed 32.17 hr0week ~SD ϭ 8.26! with no differences between religions for any demo- cultural differences in maternal employment. graphic or study variables. However, it is possible that the small number of Christians Procedure precluded the detection of significant differences between Christians and Muslims, and 5 months. Parents were contacted by a cultur- this should be noted as a study limitation. No ally matched interviewer, and a time was set differences were found between the two Arab when both parents were home and the infant 8 R. Feldman and S. Masalha was expected to be fed and awake. Home vis- small cars, a telephone, two pet animals and its lasted about 2 hr and began with the two wild animals, and a small tool set. interviewer’s introducing the study’s goals and procedures. Parents then moved to different Measures rooms to be interviewed and to complete a set of self-report measures. Questionnaires were Infancy–5 months. Of the eight infancy risk completed in Hebrew or Arabic. Question- factors studied here, six factors were ex- naires were adapted to Arabic using a back tracted from the parents’ self-report measures and forth translation by a team of profession- and the other two were based on coded obser- als and validated for use in the Palestinian vations. Depressive symptoms, marital satis- culture in two ways. First, a group of mental faction, and social support were reported at health and child development professionals ex- both 5 and 34 months. However, because this amined the suitability of each instrument for study assessed the effects of early ecological parents of young children in their society by risk, only the 5-month measures were used going over each question, assuring its rele- and the 34-month measures were used to val- vance for parents, and making minor changes idate the early measures. Apart from measures in wording for clarity. Second, questionnaires related to mother risk ~maternal depression, were tested on a pilot sample of 20 parents work–family interference, and experience of prior to their use in the current study. Parents childbirth! all self-reported and observed eco- were asked to indicate whether there are ques- logical determinants and outcome measures tions not applicable to children at that age or were based on an average of the mothers’ and to their experience as parents and to change fathers’ self-reports and behaviors. unclear words. Parents indicated that all questions were on target, and slight changes in Risk factor 1: Mother depressive symptoms. wording were made following the parents’sug- The Beck Depression Inventory ~BDI; Beck, gestions. After completion of the question- 1978! includes 21 items that measure the level naires, parents and infants were videotaped in of depressive symptoms on a 3-point scale. three interactions: mother–infant, father–infant The BDI is a widely used self-report instru- ~counterbalanced!, and a triadic family ses- ment for the assessment of depressive symp- sion. In this report, we use data from the two toms, with well-established reliability and parent–infant interactions. validity ~Bumberry, Oliver, & McClure, 1978!. Internal consistency ~Cronbach a! for this sam- 34 months. At 34 months, families and chil- ple was .85. Parents also completed the BDI dren were observed again by a culturally at 34 months, and BDI scores showed medium- matched team of observers. Home visits in- level stability over time ~r ϭ .54, p , .001!. cluded several dyadic and triadic interactions, interviews, and self-report measures. In this Risk factor 2: Parent perception of infant report, we use data from the two parent–infant temperament. The Infant Characteristics Ques- interactions. tionnaire ~Bates, Freeland, & Lounsbury, 1979! consists of 24 items measured on a 9-point Parent–child symbolic play. Fifteen minutes scale and yields four factors. The Fussy- of free play between the child and each parent Difficult factor was used in this study, as it is ~counterbalanced! were videotaped. Parents the most stable dimension in self-report instru- and children were given a set of toys that have ments of temperament ~Goldsmith & Alansky, shown to elicit symbolic play in children of 1987!. Internal consistency ~Cronbach a! was that age ~Keren et al., 2005; Tamis-LeMonda .77. The mothers’ and fathers’ fussy-difficult & Bornstein, 1994!, including two dolls, a scores were related ~r ϭ .73, p , .001! and bottle, a blanket, a tea set including two cups, were averaged into a single score. two plates, sugar and milk pots, a boiler pan, a wallet, a colored necklace, a pair of plastic Risk factor 3: Marital satisfaction. The Mar- sunglasses, a sponge, three work tools, two ital Adjustment Test ~MAT; Locke & Wallace, Risk, adaptation, and culture 9

1959! consists of 15 items, summed to create performance, the degree to which thoughts of a global marital satisfaction score. The MAT the infant interfere with work; a ϭ .74!. is a widely used instrument in the study of marital relationships, with well-demonstrated Risk factor 6: Experience of childbirth. This reliability and validity ~Gottman, Markman, was the average of three PLI items reported & Notarius, 1977!. Internal consistency ~Cron- by the mother on a scale from 1 to 5: effects of bach a! was .76. Mothers’ and fathers’ scores childbirth on self-esteem, effects of childbirth were correlated ~r ϭ .61, p , .001!, and were on marriage, and the degree of recovery from averaged into a single score. Marital satisfac- childbirth ~a ϭ .70!. tion showed individual stability from 5 to 34 Several additional items were extracted from months ~r ϭ .51, p , .001!, validating the use the PLI that address the parents’ attitudes, be- of the MAT in assessing a stable component in liefs, child-rearing goals, child-care practices, the couple’s relationship. and living arrangements. These were used to validate the individualistic versus collectivis- Risk factor 4: Social support. Two mea- tic orientations of the two societies. sures were used to assess social support networks. The Social Support Scale ~Cutrona, Traditional gender-role attitudes. These at- 1984! is a 12-item scale that provides infor- titudes entail the parent’s endorsement of the mation on the parent’s subjective perception following: mothers should stay home when of support availability in different domains infants are young, a mother’s early return to ~e.g., attachment, guidance, reassurance of work is harmful for infants, raising children is worth!, with good reliability and validity. In- the purpose of a woman’s life, men and women ternal consistency ~Cronbach’s a! was .80. should receive equal job opportunities ~nega- Mothers’ and fathers’ scores ~r ϭ .65, p , tive!, and parents should share child-care re- .001! were averaged into a single score. In the sponsibilities ~negative; a ϭ .71!. second measure, adapted from Crockenberg and Litman ~1990!, parents list names of peo- Child-care arrangements. Parents rated ple with whom they have contact within nested whether their infant was in the care of a fam- circles according to the frequency of contact. ily member, in-home babysitting, out of home The final score is the sum of the number of care ~dual employment was an inclusion cri- contacts weighed by the frequency of contact, teria for the present study!, and the number of averaged across mother and father ~a ϭ .87!. infants being cared for by the caregiver. The two social support scales were correlated ~r ϭ .59, p , .001! and their standardized Father involvement. Parents rated on a scale scores were averaged into a Social Support from 1 ~low! to 5 ~high! the degree of father composite ~a ϭ .70!. Social support was sim- involvement in household chores and in ilarly measured at 33 months and was highly child-care responsibilities and the mothers’and stable ~r ϭ .72, p , .001!. fathers’ scores ~r ϭ .57, p , .01! were averaged into a Father Involvement composite ϭ Risk factor 5: Work–family interference. ~a .84!. Parents were interviewed with the Parental Leave Inventory ~PLI!, a validated instrument Reasons for returning to work. Mothers including 108 items ~Feldman et al., 2001, rated yes0no whether the following influenced Feldman, Sussman, et al., 2004!. Mother work– their decision to return to work: financial need, family interference was the average of six ques- career advancement, and self-fulfillment. tions rated from 1 ~low! to 5 ~high! that considered the balance between the work and Experience with infant care. Two items con- family roles following childbirth ~e.g., how sidered the level of involvement the mother well mother is performing at work, how post- had with infant siblings and her general expe- birth job performance compares with prebirth rience with infants. 10 R. Feldman and S. Masalha

Role model. Mothers rated whether their man, Weller, Eidelman, & Sirota, 2003!. The mothers worked outside the home when they parent sensitivity factor was used ~mother a ϭ were growing up. For those whose mothers .93, father a ϭ .91!, which includes parent did work, mothers rated the degree to which acknowledgement of infant signals, visual con- ~1to5! their own mother’s career was impor- tact, warm and positive affect, appropriate vo- tant for her self-concept. cal quality, resourcefulness in handling infant distress or expanding the interaction, consis- Child-rearing goals. Parents rated ~1–5! the tency of style, and adaptation to infant states. degree to which they considered each of the Two coders, one from each culture who following a child-rearing goal: self-expression, spoke Hebrew and Arabic, coded the inter- respect for elders, creativity, compliance to actions and each coded sessions from the two rules, and positive relations with others. Par- cultures. Coders were trained to 90% reliabil- ents also rated the degree to which they con- ity. Interrater reliability, conducted for 25 ses- sider each of the following as an important sions, had an averaged intraclass r value of attribute in a child: kind, assertive, polite, quiet, .90 ~range ϭ .80–.95!. Mothers’ and fathers’ and smart. scores were interrelated ~r ϭ .54, p , .001! and were averaged into a single score. 34 months. Risk factor 8: Infant negative emotionality. Child behavior problems. Mother and Infant negative emotionality was coded from father each completed the Child Behavior the mother–infant and father–infant inter- Checklist02–3 Years ~CBCL02–3; Achen- action using the CIB codes ~mother a ϭ .73, bach, 1992!. Items in the CBCL are rated on a father a ϭ .75!. Codes included were: infant 3-point scale from never applies to almost al- shows fatigue and tiredness, emits fuss-cry ways applies. Three summary scores are ex- vocalization, withdraws, shows discontent- tracted, a total behavior score, an externalizing ment. Infant negative emotionality during symptoms scores, and an externalizing symp- infant–mother and infant–father interactions toms score. Mothers’ and fathers’ scores were were correlated ~r ϭ .55, p , .001! and were highly correlated ~r ϭ .79–.83! and were av- averaged into an infant negative emotionality eraged into a single score. The CBCL02–3 factor. The observed and self-reported mea- has been previously validated in samples of sures showed significant but low correlations Israeli and Palestinian toddlers ~Auerbach et al., ~r ϭ .20, p , .05!, which precluded their in- 1996!. Internal consistency for this sample was tegration into a single score. ~Cronbach’s a! .81. Coding: 34 months. Coding: Infancy–5 months. Child symbolic play. Symbolic play was Risk factor 7: Parent sensitivity. Mother– coded separately for child and parent along infant and father–infant interactions were eight hierarchical levels of symbolization and coded using the Coding Interactive Behavior a default. For each 10-s epoch one out of the Manual ~CIB; Feldman, 1998!. The CIB is a following nine mutually exclusive codes was global rating system that includes 42 codes, applied in line with previous research ~Feld- each rated on a 5-point scale ~1 ϭ low,5ϭ man, Eidelman, et al., 2004; Feldman & high!. The CIB has been validated on sam- Greenbaum, 1997; Melstein-Damast, Tamis- ples from several cultures and has shown sen- LeMonda, & Bornstein, 1996!. Child play lev- sitivity to infant age, cultural setting, and els included three presymbolic levels: ~a! no biological and emotional risk conditions ~Feld- play; ~b! object manipulation ~e.g., touching, man, 2000; Feldman, Eidelman, & Roten- throwing!; and ~c! functional play: use of a berg, 2004; Feldman, Eidelman, Sirota, & toy in its intended way, for example, moving a Weller, 2002; Feldman, Keren, Gross-Rozval, car on the floor. Two simple symbolic levels & Tyano, 2004; Feldman & Klein, 2003; Feld- were coded ~d! self-pretend: unitary symbolic Risk, adaptation, and culture 11 acts around the self, for example, sleeping or Cultural differences in parental attitudes and combing hair; and ~e! other pretend: unitary child-rearing practices: Individualistic symbolic acts that include others in the pre- versus collectivistic orientation tend play, for example, feeding a doll. Three complex symbolic levels were coded: ~f! com- Living arrangements. All young Arab fami- binatorial pretend: combining several play lies, but only 6% of the Israeli families lived schemes into a single act, in one of the fol- within a walking distance to at least one set of lowing three types: a single scheme is played grandparents, confirming to the nuclear ver- with several objects ~e.g., feeding doll and sus extended living arrangements of the two then feeding dog!, several schemes are societies. played with the same object ~e.g., feeding doll Data pertaining to traditional gender-role then putting it to bed!, or different schemes attitudes, child-care arrangements, the moth- are organized in order ~e.g., dressing doll, ers’ role model, reasons for returning to work, putting it inside a car, driving car!; ~g! hier- and experience with infants for this sample at archical pretend: a single act that expresses 5 months are reported elsewhere ~Feldman a hierarchical scheme ~e.g., a child plans et al., 2001!. In short, the data indicate that ahead and fits objects to predetermined roles!; Arab mothers and fathers adhered to more tra- ϭ and ~h! substitutional pretend: a child sub- ditional gender-role philosophies, F ~1, 281! stitutes one object for another in a deliberate 78.46, p , .001. Arab mothers returned to fashion, for example, a stick is used instead work birth for financial reasons and not for of a car. reasons of self-fulfillment or career advance- 2 ϭ Coding of children’s symbolic play was ment, x ~1, 141! 24.22, p , .001. Most of conducted by Israeli and Arab coders who the grandmothers in the Arab but not the Is- 2 did not participate in the 5-month coding, and raeli group did not work outside the home, x ϭ each coded interactions from the two cul- ~1, 141! 10.77, p , .001, and among those tures. Reliability was computed on 25 inter- whose mothers did work, Israeli mothers per- actions and reliability k values averaged .88 ceived their mothers’ career as more central to ϭ ~range ϭ .79–.94!. The proportion of time the their self concept, F ~1, 80! 11.68, p , .001. infant spent in each play level was computed Arab mothers had more experience in caring ϭ and three composites were created: func- for infants in their youth, F ~1, 140! 9.21, tional play ~manipulation ϩ functional play!, p , .001, and 65% of Arab infants, compared simple symbolic play ~self-pretend ϩ other to only 12% of Israeli infants were cared for 2 ϭ pretend!, and complex symbolic play ~com- by a kin, x ~1, 141! 16.97, p , .001. binatorial, hierarchical, and substitutional pretend!. The complex symbolic play com- Father involvement. Arab parents reported posites in the mother–child and father–child lower father involvement in household and interactions were related ~r ϭ .57, p , .001! child-care responsibilities, F ~140! ϭ 54.13, and were averaged into a single score. This p , .001, keeping with the traditional gender score, which indexes the age-appropriate level roles. of symbolic play, was used as a criterion variable. Child-rearing goals. As expected, Israeli parents placed higher emphasis on the child’s “self-expression” and “creativity,” F ~1, 281! ϭ Results 43.54 and 21.24, respectively, p , .001, whereas Arab parents had higher endorsement As a first step, we examined parental atti- for “respect for elders” and “compliance to tudes, beliefs, and practices to validate the rules,” F ~1, 281! ϭ 54.11 and 17.24, respec- individualistic versus collectivistic orienta- tively, p , .001. Parents in the two societies tion of the two societies. Only data from cou- also endorsed different qualities in a child. ples participating in the two assessment points Israeli parents placed a higher value on an are reported here. “assertive” and “smart” child, F ~1, 323! ϭ 12 R. Feldman and S. Masalha

Table 1. Descriptive statistics for study variables

Israeli ~n ϭ 86! Arab ~n ϭ 55! Univariate MSDMSDF ~1, 140!

5 months Child Difficult temperament ~reported! 21.11 6.19 24.96 6.22 10.87*** Negative emotionality ~observed!a 1.23 0.70 1.17 0.78 0.18 Mother Depressive symptoms 5.25 3.81 9.02 7.03 14.66*** Work–family interferencea 3.67 1.54 2.96 0.65 7.49** Childbirth experiencea 3.27 1.28 3.66 1.55 4.12* Context Parents’ marital satisfaction 114.70 12.46 108.66 15.93 7.23** Social supportb Ϫ0.05 0.10 0.47 0.22 15.32** Parent–child interaction Parent sensitivity0responsivenessa 4.04 0.59 3.12 0.66 59.33*** 34 months Symbolic playc 0.11 0.09 0.08 0.10 1.19 Internalizing symptoms 46.97 8.59 51.94 11.13 13.12** Externalizing symptoms 45.35 8.53 47.05 10.36 3.98* Total CBCL score 46.21 8.39 49.98 11.10 8.24** aRated on a scale of 1 ~low! to 5 ~high!. bThe z scores of the two social support questionnaires. cThe proportion of time in complex symbolic play. *p , .05. **p , .01. ***p , .001.

43.11 and 10.33, respectively, p , .001, der. The data at 5 months are presented only whereas Arab parents had higher endorsement for the 141 families who completed the two for a “quiet” and “polite” child, F ~1, 323! ϭ assessment points. 35.97 and 19.25, respectively, p , .001. As seen, cultural differences were found In sum, the cultural comparisons reveal sig- for most factors. Israeli parents scored higher nificant differences between the two societies on parental sensitivity and reported more on dimensions related to collectivism versus marital satisfaction, less depressive symp- individualism and to their specific expres- toms, and lower perceived infant difficulty. sions in Arab cultures ~Abudabbeh, 1998!, in- Arab mothers experienced less work–family cluding traditional gender-role philosophies, interference, recovered better from child- extended living arrangement, child care by a birth, and Arab parents reported higher levels kin, less focus on career as means for women’s of social support. Symbolic play levels were self-expression, and child-rearing goals phrased comparable across societies and higher in terms of honor, respect for elders, and CBCL scores were reported for Arab tod- compliance. dlers. In considering mean-level differences, the differences in sample size between the two cultures should be noted as a study Descriptive statistics for study variables limitation. Descriptive statistics for all study variables Prior to assessing the cumulative and inter- are presented according to culture ~Table 1!. active effects of the ecological determinants No gender differences or Gender ϫ Culture on children’s symbolic skills and social– interactions were found for any of the ecolog- emotional adaptation, bivariate correlations ical determinants or the outcome measures and were computed between the predictor vari- the data were therefore collapsed across gen- ables. To examine cultural differences in the Risk, adaptation, and culture 13 magnitude of the correlations, correlations were .25 .35* .22 computed separately for each group ~Table 2!. .10 Ϫ Ϫ Ϫ The data seen in Table 2 indicate that, for Ϫ the most part, a similar pattern of correlations emerged in the two cultures. In both societies, maternal depression was related to infant ob- .40** .27* served and reported difficult temperament, in- .07 Ϫ Ϫ fant negative emotionality correlated with Ϫ lower parental sensitivity, marital satisfaction was associated with higher social support, social support was related to higher sensitivity, and maternal depression correlated with lower .27* .02 .11 .60*** .28* Ϫ Ϫ Ϫ marital satisfaction. These data suggest that Ϫ the co-occurrence of risk conditions may be universal. Several associations were found only in the Israeli group, such as between maternal .14 .05 .11 depression with higher work–family interfer- .15 Ϫ Ϫ Ϫ ence, more negative experience of childbirth, Ϫ and lower sensitivity, or between perceived infant difficulty and negative birth experience. In general, maternal depression corre- .17 — .13 .14 .27* .14.12 .21* .23* — .42* .36* — .17 .29* .14 .22* .25* .20* — Ϫ Ϫ Ϫ lated with most other risk conditions in the Ϫ Israeli group. Other correlations were significant only in the Arab group, such as the relations between perceived infant difficulty and lower marital satisfaction and less social sup- .22* .26* .33** .22* Ϫ Ϫ Ϫ port. Overall, however, the magnitudes of the Ϫ correlations were not significantly different between the two groups. .11 .09 Predicting child symbolic competence and .27** Ϫ Ϫ social–emotional adaptation Ϫ Tables 3 and 4 present two hierarchical regression models predicting child symbolic play and behavior problems from culture, ecologi- 24678 123456 .10 cal determinants, and their interaction. .09.20* .03 Ϫ Ϫ Ϫ Prior to computing these regressions we examined whether the parental attitudes, beliefs, and child-rearing practices related to the individualistic0collectivistic distinction con- tribute to the prediction of symbolic competence and behavior problems above and beyond .001. , cultural membership. None of these factors p explained unique variance in symbolic play or behavior problems above and beyond culture, .01. ***

and these variables were not entered in the , p next set of regressions. Similarly, maternal Bivariate correlations between predictor variables employment and child gender were unrelated The correlations above the diagonal are for the Arab group; correlations below the diagonal are for the Israeli group. .05. **

to the outcome variables. In the regressions , p Table 2. Note: * 1. Difficult temperament — .13 .29* .07 2. Negative emotionality .20* — .27* .21 3. Mother depressive symptoms .34** .20* — .07 4. Mother work–family interference .15 .16 .21* — 5. Childbirth experience .26** 6. Parents’ marital satisfaction reported here, variables were entered one by 7. Social support 8. Parent sensitivity 14 R. Feldman and S. Masalha

Table 3. Predicting child symbolic competence

Predictors b R DR2 DFdf

Culture .54 .12 .02 3.01 1, 139 Infant difficult temperament Ϫ.21 .16 .01 0.87 2, 138 Infant negative emotionality Ϫ.23 .20 .01 1.62 3, 137 Mother depressive symptoms Ϫ.41* .27 .05 4.19* 4, 136 Work–family interference Ϫ.15 .29 .00 0.36 5, 135 Childbirth experience Ϫ.22 .30 .01 0.74 6, 134 Marital satisfaction .38* .34 .03 2.78 7, 133 Social support .31 .36 .02 1.35 8, 132 Parent sensitivity .56* .42 .06 5.43** 9, 131 Culture ϫ Negative Emotionality Ϫ.13 .43 .00 0.17 10, 130 Culture ϫ Depressive Symptoms Ϫ.59* .50 .05 4.98** 11, 129 Culture ϫ Social Support .22 .52 .01 0.56 12, 128 Culture ϫ Parent Sensitivity .52* .56 .05 5.13** 13, 127

Note: R2 total ϭ .32; F ~13, 127! ϭ 4.92, p , .001. *p , .05. **p , .01.

Table 4. Predicting child behavior problems

Predictors b R DR2 DFdf

Culture .65** .28 .08 10.70*** 1, 139 Infant difficult temperament .42* .33 .04 3.97* 2, 138 Infant negative emotionality .37* .35 .01 1.18 3, 137 Mother depressive symptoms .62* .41 .05 4.24* 4, 136 Work–family interference .39 .42 .00 0.43 5, 135 Childbirth experience Ϫ.36 .44 .01 0.86 6, 134 Marital satisfaction Ϫ.12 .44 .00 0.13 7, 133 Social support Ϫ.46* .49 .05 4.96** 8, 132 Parent sensitivity Ϫ.58* .54 .06 5.43** 9, 131 Culture ϫ Depressive Symptoms Ϫ.59* .58 .05 5.03** 10, 130 Culture ϫ Parent Sensitivity .18 .58 .00 0.28 11, 129 Culture ϫ Negative Emotionality Ϫ.13 .58 .00 0.33 12, 128 Culture ϫ Social Support .49* .62 .04 4.14* 13, 127

Note: R2 total ϭ .39; F ~13, 127! ϭ 5.98, p , .001. *p , .05. **p , .01. ***p , .001.

one in a theoretically determined order. Cul- ence as the maternal factors; marital ture was entered first as a binary variable to satisfaction and social support as the contex- partial out variance related to cultural differ- tual factors; and parental sensitivity as an ob- ences. In the following, the ecological deter- served index of the environmental attunement minants were entered according to Belsky’s to the child. In the next four blocks, the four ~1984! determinants of parenting model, which interaction terms were entered in the same suggests that infant dispositions, maternal per- order as the determinants: culture and infant sonality, and contextual conditions affect chil- negative emotionality, culture and maternal dren and decreasing order. Infant perceived depressive symptoms, culture and social sup- and observed difficulty were thus entered as port, and culture and parent sensitivity. The the child factors; maternal depression, work– criterion variable was the child’s complex sym- family interference, and childbirth experi- bolic play score: the age-appropriate level of Risk, adaptation, and culture 15 symbolic complexity. Results for symbolic significant, F ~1, 54! ϭ .47, p . .10. The competence are presented in Table 3. opposite pattern was found for social support. As seen in Table 3, culture did not make a The difference in the CBCL scores of Arab unique contribution to the prediction of sym- children whose parents reported high ~M ϭ bolic play and children of both cultures dis- 47.44, SD ϭ 11.21, N ϭ 31! or low ~M ϭ played similar levels of symbolic complexity. 51.71, SD ϭ 10.88, N ϭ 24! social sup- Two determinants had a unique contribution port ~using the median split! was significant, to the prediction of symbolic competence: ma- F ~1, 54! ϭ 4.02, p , .05, and no difference ternal depression and parent sensitivity. Two was found between the high ~M ϭ 46.03, SD ϭ interactions were predictive of symbolic play: 8.28, N ϭ 40! and low ~M ϭ 46.47, SD ϭ 8.76, culture and maternal depression and culture N ϭ 46! social support group among Israeli and parent sensitivity. Examination of the first toddlers, F ~1, 85! ϭ .54, p . .10. Contrary to interaction indicated that for the Israeli group, our expectation, the interaction of culture and parent sensitivity was significantly related to infant negative emotionality was not signifi- toddler’s symbolic competence ~r ϭ .33, p , cant. A full moderator model is implied here, .01!, but the associations between sensitivity with culture moderating the relations of early and symbolic skills for the Arab group were maternal depression and family social support not significant ~r ϭ .03, p . .10!. The differ- and later behavior problems ~Baron & Kenny, ence between the magnitudes of these two cor- 1986!. We also examined regression models relations was significant ~Fisher’s Z ϭ 2.07, in which parental sensitivity was entered be- p , .05!. The opposite pattern emerged for fore the mother’s depressive symptoms, as a the interaction of maternal depression and sym- more proximal variable, and the results re- bolic play. Correlations were significant for mained essentially the same. the Arab group ~r ϭϪ.39, p , .01!, but not for the Israeli group ~r ϭϪ.08, p . .10!, and Discussion the difference between the magnitudes of the correlations was significant ~Fisher’s Z ϭ 2.01, The results of this study suggest that culture p , .05!. A note of caution is warranted here, moderates the effects of risk conditions on as a full moderator model was not implied developmental outcomes at the transition from because culture was not independently predic- infancy to early childhood. Risk and protective of symbolic play. tive factors measured at 5 months differen- A similar hierarchical regression model tially affected the development of symbolic was used to predict children’s behavior prob- skills and behavior adaptation at 34 months in lems and is presented in Table 4. As seen in the two cultures. Specifically, the parents’ sen- this table, four ecological determinants were sitive responsiveness, in terms of visual, vo- uniquely predictive of infants’ behavior prob- cal, and affective coordination with the infant’s lems; infant difficult temperament, maternal signals, had a greater impact on the symbolic depressive symptoms, low social support, and complexity of Israeli toddlers. The family’s low parent sensitivity. Two significant inter- social support networks buffered against the actions were found: culture and maternal de- emergence of behavior problems in the Arab pressive symptoms and culture and social group, whereas maternal postpartum depres- support. As to the first interaction, the behav- sive symptoms had a more negative effect on ior problems of Israeli toddlers whose moth- the expression of externalizing and internaliz- ers scored high ~M ϭ 48.23, SD ϭ 7.91, N ϭ ing symptoms in the Israeli group. According 39! or low ~M ϭ 44.98, SD ϭ 8.63, N ϭ 47! on to culture-specific models of human develop- maternal depression ~using the median split!, ment ~Greenfield et al., 2003; Rogoff, 2003; was significant, F ~1, 85! ϭ 4.62, p , .05. Shweder, 2003!, the saliency of risk and pro- Among Arabs, on the other hand, the differ- tective factors may vary according to the ence between children of mothers scoring low cultural context. The present findings are con- ~M ϭ 49.36, SD ϭ 10.87, N ϭ 39! or high sistent with these models and suggest that the ~M ϭ 50.51, SD ϭ 12.32, N ϭ 25! was not resources, living conditions, child-rearing at- 16 R. Feldman and S. Masalha titudes, and parenting practices available within works in collectivistic societies reflect the the infant’s ecology are likely to shape the family’s status in the community, and this effects of risk conditions on the development factor may serve as a proxy for a range of of competence and adaptation. growth-promoting conditions in the child’s Israeli and Palestinian parents differed on a environment. range of child-rearing attitudes and family con- The negative effects of maternal postpar- ditions, and these cultural differences may have tum depression on the emergence of behavior determined the centrality of specific risk fac- problems have been demonstrated in numer- tors to child outcomes. Findings from the ous studies ~Carter et al., 2001; Feldman & Palestinian sample were consistent with Eidelman, 2004; Murray et al., 1999!. The Abudabbeh’s ~1998! model on the special type present findings highlight these effects in re- of collectivism observed in Arab families. Par- lation to the rearing context of individualistic ents in this group adhered to more traditional societies. Several possible explanations may gender-role philosophies, fathers were less in- account for the culture-specific effects of ma- volved in child rearing, mothers perceived their ternal depression on behavior problems. For work as motivated by the family’s financial children growing in nuclear family settings, needs rather than the need for self-fulfillment, the mother is often the infant’s primary care- parents considered deference to authority and giver, the nature of the mother–infant relation- compliance as central parenting goals, moth- ship is critical for optimal growth, and the ers reported caring for infants siblings as girls, infant’s exposure to caregiving females of a approximately two thirds of Palestinian in- kin relationship is usually limited. Within an fants were cared for by a kin, and all young extended family ecology, where several young couples lived within close quarters to their mothers raise their children together and in- family of origin. Within this multigenera- fants are exposed to multiple caregivers on a tional, extended-family context, specific ef- daily basis ~LeVine, 2002; Sharma & Fischer, fects of social support and maternal depression 1998!, the impact of the mother’s depressed on child adaptation were described. Accord- mood may be somewhat attenuated. Another ing to the “cultural pathways” perspective, the possibility is that in societies that stress self- resources available within the infant’s ecol- expression, creativity, and initiative, as seen ogy serve as vehicles for growth and adapta- here in the child-rearing goals of Israeli par- tion, particularly at the transition from infancy ents, child adaptation may be more suscepti- to early childhood ~Keller et al., 2004!. Con- ble to the mother’s depressed mood. Depressed sistent with this perspective, social support, mothers are less competent in supporting the an important feature of the collectivistic con- infant’s intentionality, initiation, autonomy, and text, had a more beneficial impact on child self-sufficiency ~Feldman & Reznick, 1996; adaptation in the Palestinian society, where Kochanska & Kuczynski, 1991! or in provid- young working mothers rely on help from ing sufficient external regulation for the de- the larger family to balance work and parent- velopment of self-regulatory capacities ~Field, ing ~Mar’i & Mar’i, 1985!. A recent report 1992!. As a result, the mother’s depressive from the United Arab Emirates ~Eapen et al., symptoms may be more crucial to child adap- 2004! similarly shows that a central predictor tation in such societies than in cultures that of children’s behavior problems at 3 years emphasize compliance and respect for elders. was the mother’s lack of social support, par- This may be especially salient during the early ticularly among immigrant mothers. These preschool years, a period when autonomy and findings are also consistent with those of initiative are viewed as central developmental Cutrona and colleagues ~2000!, who showed milestones for Western children. This hypoth- that social support had a more positive im- esis, however, is preliminary, and requires rep- pact on mother and child’s functioning in lication in larger samples and various cultures cultural settings guided by a more communal to further specify the links between culture, approach. A somewhat related explanation maternal depressive symptomatology, and child may be that the parents’ social support net- adaptation. Risk, adaptation, and culture 17

Longitudinal associations between the graphic variables, some unknown or unmea- parent’s sensitive responsiveness to infant sig- sured factor may have contributed to the nals and the level of symbolic complexity in reported outcomes. This study presents the early childhood were stronger for the Israeli first effort to follow the development of in- group. In general, parents in individualistic so- fants and families in the Palestinian society, cieties foster the child’s symbolization and tool and much further research is required to chart use through special moments of play that are the growth trajectories of children in that so- organized around face-to-face interactions. Par- ciety across childhood and beyond. ents in more traditional societies are less in- Finally, the findings have implications for clined to set a special time for play and periods the study of resilience, defined as positive ad- of play are less distinguished from the stream aptation in the face of adverse conditions ~Luthar of daily life ~Klein, 1996; Rogoff et al., 1993!. & Cicchetti, 2000!. Resilience has generally In cultures that engage in face-to-face play, the been studied in relation to interpersonal and so- parent’s moment by moment coordination with cial factors, and recently its biological under- the child’s affective expression supports the de- pinnings have been addressed ~Curtis & velopment of symbolic skills ~Melstein-Damast Cicchetti, 2003!. The present findings under- et al., 1996; Slade, 1987!. Still, the fact that no score the need to include culture in the study of cultural differences were found in levels of sym- resilience, in addition to its biological roots and bolic complexity indicates that parents in col- social correlates. Specific cultural provisions, lectivistic societies find effective ways to relational patterns, living arrangements, and facilitate child symbolization and future re- child-rearing philosophies may serve as pro- search is required to chart these pathways more tective factors in the face of adversity in cer- fully. The greater effect of maternal depressive tain cultural or subcultural contexts but not in symptoms on child symbolic play in the Israeli others. Resilience, therefore, need to be stud- group may relate to the use of affect synchrony ied in relation to the core features of the culture as a typical mode of early interactions. Possi- and its availability for a specific child or fam- bly, in cultures that rely on affect matching, ma- ily at important nodes. Further research is re- ternal depression, which compromises the quire to elucidate the theoretical, empirical, and development of mother–infant synchrony ~Feld- clinical aspects of cumulative risk, protective man, 2003; Field, 1992!, has a more negative factors, and resilience in healthy and at risk pop- impact than in societies that rely on other modes ulations of different ages, etiologies, and cul- of parent–infant relatedness. tures. Better understanding of early ecological Limitations of the findings should be noted. risk and its accumulation may enable the con- The two groups were not of equal size, and struction of more efficient interventions that are the smaller number of Arab families should be suited to the child’s social context and can help remembered in the interpretation of the inter- promote competence and adaptation in the early action effects. It is also possible that although years in ways that are consistent with the cul- groups were matched on a range of demo- tural goals and meaning systems.

References

Abdalla, I. A. ~1995!. Sex, sex-role self-concepts, and Auerbach, J. G., Yirmiya, N., & Kamel, F. N. ~1996!. career decision-making self-efficacy among Arab stu- Behavior problem in Israeli Jewish and Palestinian dents. Social Behavior and Personality, 23, 389–402. preschool children. Journal of Clinical Child Psychol- Abudabbeh, N. ~1998!. Counseling Arab–American fam- ogy, 25, 398–405. ilies. In U. P. Gielen & A. L. Comunian ~Eds.!, The Baron, R. M., & Kenny, D. A. ~1986!. The moderator– family and family therapy in international perspective mediator variable distinction in social psychological ~pp. 115–126!. Trieste, Italy: Edizioni Lint Trieste. research: Conceptual, strategic, and statistical consid- Achenbach, T. M. ~1992!. Manual for the Child Behavior erations. Journal of Personality and Social Psychol- Checklist02–3 years. Burlington, VT: University of ogy, 51, 1173–1182. Vermont, Department of Psychiatry. Bates, E., O’Connell, B., & Shore, C. ~1987!. Language Al-Haj, M. ~1989!. Social research on family lifestyles and communication in infancy. In J. D. Osofsky ~Ed.!, among Arabs in Israel. Journal of Comparative Fam- Handbook of infant development ~2nd ed., pp. 149– ily Studies, 20, 175–195. 204!. New York: Wiley. 18 R. Feldman and S. Masalha

Bates, J. E., Freeland, C. A., & Lounsbury, M. L. ~1979!. transition: Infancy to childhood. Chicago: Chicago Measurement of infant difficultness. Child Develop- University Press. ment, 50, 794–803. Cicchetti, D., & Lynch, M. ~1993!. Toward and ecological0 Beck, A. T. ~1978!. Beck Depression Inventory. San transactional model of community violence and child Antonio, TX: The Psychological Corporation0Harcourt maltreatment: Consequences for children’s develop- Brace Jovanovich. ment. Psychiatry, 56, 96–118. Belsky, J. ~1984!. The determinants of parenting: A pro- Cicchetti, D., & Toth, S. L. ~1998!. The development of cess model. Child Development, 55, 83–96. depression in children and adolescents. American Psy- Belsky, J. ~1998!. Theory testing, effect-size evaluation, chologist, 53, 221–241. and differential susceptibility to rearing influence: The Cowan, P., & Cowan, C. ~1992!. When partners become case of mothering and attachment. Child Develop- parents: The big life change for couples. New York: ment, 68, 598–600. Basic Books. Belsky, J., Hsieh, K. H., & Crnic, K. ~1998!. Mothering, Crockenberg, S., & Litman, C. ~1990!. Autonomy as com- fathering, and infant negativity as antecedents of boys’ petence in two-year-olds: Maternal correlates of child externalizing problems and inhibition at age 3 years: compliance, noncompliance, and self-assertion. De- Differential susceptibility to rearing experience? De- velopmental Psychology, 26, 961–971. velopment and Psychopathology, 10, 301–319. Curtis, W. J., & Cicchetti, D. ~2003!. Moving research on Belsky, J., & Pensky, E. ~1988!. Marital change across resilience into the 21st century: Theoretical and meth- the transition to parenthood. Marriage and Family odological considerations in examining the biological Review, 12, 133–156. contributors to resilience. Development and Psycho- Bengi-Arslan, L., Verhulst, F. C., van der Ende, J., & Erol, pathology, 15, 773–810. N. ~1997!. Understanding childhood problem behav- Cutrona, C. E. ~1984!. Social support and stress in the ior from a cultural perspective: Comparison of prob- transition to parenthood. Journal of Abnormal Psy- lem behaviors and competencies in Turkish immigrant, chology, 93, 378–390. Turkish, and Dutch children. Social Psychiatry and Cutrona, C. E., Russell, D. W., Hessling, R. M., Brown, Psychiatric Epidemiology, 32, 477–484. P. A., & Murry, V. ~2000!. Direct and moderating Bornstein, M. H., Haynes, O. M., Pascual, L., Painter, effects of community context on the psychological K. M., & Galperin, C. ~1999!. Play in two societies: well-being of African-American women. Journal of Pervasiveness of process, specificity of structure. Child Personality and Social Psychology, 79, 1088–1101. Development, 70, 317–331. Cutrona, C. E., & Troutman, B. R. ~1986!. Social support, Bronfenbrenner, U. ~1977!. Toward and experimental ecol- infant temperament and parenting self-efficacy: A me- ogy of human development. American Psychologist, diational model of post-partum depression. Child De- 32, 513–531. velopment, 57, 1507–1518. Briggs, J. L. ~1972!. The issue of autonomy and aggres- Dawson, G., Ashman, S. B., Panagiotides, H., Hessl, D., sion in the three-year-old: The Utku Eskimo case. Self, J., Yamada, E., et al. ~2003!. Preschool outcomes Seminars in Psychiatry, 4, 317–329. of children of depressed mothers: Role of maternal Bumberry, W., Oliver, J. M., & McClure, J. M. ~1978!. behavior, contextual risk, and children’s brain activ- Validation of the Beck Depression Inventory in a uni- ity. Child Development, 74, 1158–1175. versity population. Journal of Consulting and Clini- Dwairy, M., & van Sickle, T. D. ~1996!. Western psycho- cal Psychology, 46, 150–155. therapy in traditional Arab societies. Clinical Psychol- Burt, K. B., van Dulmen, M. H., Carlivati, J., Egeland, B., ogy Review, 16, 231–249. Sroufe, A. L., Forman, D. R., et al. ~2005!. Mediating Eapen, V., Yunis, F., Zoubeidi, T., & Sabri, S. ~2004!. links between maternal depression and offspring psy- Problem behaviors in 3-year-old children in the United chopathology: The importance of independent data. Arab Emirates. Journal of Pediatric Health Care, 18, Journal of Child Psychology and Psychiatry, 46, 186–191. 490–499. Eapen, V., Zoubeidi, T., & Yunis, F. ~2004!. Screening for Burt, V. K., & Stein, K. ~2002!. Epidemiology of depres- language delay in the United Arab Emirates. Child: sion throughout the female life-cycle. Journal of Clin- Care, Health, and Development, 30, 541–549. ical Psychiatry, 63~Suppl. 7!, 9–15. Easterbrooks, M. A., Biesecker, G., & Lyons-Ruth, K. Carter, A. S., Garrity-Rokous, F. E., Chazan-Cohen, R., ~2000!. Infancy predictors of emotional availability in Little, C., & Briggs-Gowan, M. J. ~2001!. Maternal middle childhood: The role of attachment security depression and comorbidity: Predicting early parent- and maternal depressive symptomatology. Attach- ing, attachment security, and toddler social-emotional ment and Human Development, 2, 170–187. problems and competencies. Journal of the American El-Islam, M. F. ~1983!. Cultural change and intergenera- Academy of Child & Adolescent Psychiatry, 40, 18–26. tional relationships in Arabian families. International Caspi, A. ~2000!. The child is father of the man: Per- Journal of Family Psychiatry, 4, 321–329. sonality continuities from childhood to adulthood. Feldman, R. ~1998!. Coding interactive behavior manual. Journal of Personality and Social Psychology, 78, Unpublished manuscript, Bar-Ilan University. 158–172. Feldman, R. ~2000!. Parents’ convergence on sharing and Celia, S. ~2004!. Interventions with infants and families marital satisfaction, father involvement, and parent– at risk: Context and culture. Infant Mental Health child relationship at the transition to parenthood. In- Journal, 25, 502–507. fant Mental Health Journal, 21, 176–191. Chang, L., Morrissey, R. F., & Koplewicz, H. S. ~1995!. Feldman, R. ~2003!. Infant–mother and infant–father syn- Prevalence of psychiatric symptoms and their relation chrony: The coregulation of positive arousal. Infant to adjustment among Chinese–American youth. Jour- Mental Health Journal, 24, 1–23. nal of the American Academy of Child & Adolescent Feldman, R., & Eidelman, A. I. ~2004!. Parent–infant Psychiatry, 34, 91–99. synchrony and the social–emotional development of Cicchetti, D., & Beeghley, M. ~Eds.!. ~1990!. The self in triplets. Developmental Psychology, 40, 1133–1147. Risk, adaptation, and culture 19

Feldman, R., Eidelman, A. I., & Rotenberg, N. ~2004!. of verbal and nonverbal behavior. Journal of Mar- Parenting stress, infant emotion regulation, maternal riage and the Family, 9, 461–477. sensitivity, and the cognitive development of triplets: Greenfield, P. M., Keller, H., Fuligni, A., & Maynard, A. A model for parent and child influences in a unique ~2003!. Cultural pathways through universal develop- ecology. Child Development, 75, 1774–1791. ment. Annual Review of Psychology, 54, 461–490. Feldman, R., Eidelman, A. I., Sirota, L., & Weller, A. Han, W. J., & Waldfogel, J. ~2003!. Parental leave: The ~2002!. Comparison of skin-to-skin ~Kangaroo! and impact of recent legislation on parents’ leave taking. traditional care: Parenting outcomes and preterm in- Demography, 40, 191–200. fant development. Pediatrics, 110, 16–26. Heinicke, C. M. ~1984!. Impact of prebirth parent person- Feldman, R., & Greenbaum, C. W. ~1997!. Affect regula- ality and marital functioning on family development: tion and synchrony in mother–infant play as precur- A framework and suggestions for further study. De- sors to the development of symbolic competence. Infant velopmental Psychology, 20, 1044–1053. Mental Health Journal, 18, 4–23. Heubeck, B. G. ~2000!. Cross-cultural generalizability of Feldman, R., Keren, M., Gross-Rozval, O., & Tyano, S. CBCL syndromes across three continents: From the ~2004!. Mother–child’s touch patterns in infant feed- USA to Holland to Australia. Journal of Abnormal ing disorders: Relation to maternal, child, and envi- Child Psychology, 28, 439–450. ronmental factors. Journal of the American Academy Hofstede, G. ~2001!. Culture’s consequences: Comparing of Child & Adolescent Psychiatry, 43, 1089–1097. values, behaviors, institutions, and organizations Feldman, R., & Klein, P. S. ~2003!. Toddlers’ self- across nations ~2nd ed.!. Thousand Oaks, CA: Sage. regulated compliance with mother, caregiver, and Holden, G. W., & Ritchie, K. L. ~1991!. Linking extreme father: Implications for theories of socialization. De- marital discord, child rearing, and child behavior prob- velopmental Psychology, 39, 680–692. lems: Evidence from battered women. Child Develop- Feldman, R., Masalha, S., & Nadam, R. ~2001!. Cultural ment, 62, 311–327. perspective on work and family: Dual-earner Jewish Hyde, J. S., Klein, M. H., Essex, M. J., & Clark, R. ~1995!. and Arab families at the transition to parenthood. Jour- Maternity leave and women’s mental health. Psychol- nal of Family Psychology, 15, 492–509. ogy of Women Quarterly, 19, 257–285. Feldman, R., & Reznick, J. S. ~1996!. Maternal percep- Israel Central Bureau of Statistics. ~2002!. Statistical ab- tion of infant intentionality at 4 and 8 months. Infant stracts of Israel ~No. 53!. Jerusalem: Keter Press. Behavior and Development, 19, 485–498. Javo, C., Heyerdahl, S., & Ronning, J. A. ~2000!. Parent Feldman, R., Sussman, A. L., & Zigler, E. ~2004!. Pater- reports of child behavior problems in young Sami nal leave and work adaptation at the transition to children: A cross-cultural comparison. European Child parenthood: Individual, marital, and social correlates. and Adolescent Psychiatry, 9, 202–211. Journal of Applied Developmental Psychology, 25, Kagan, J. ~1981!. The second year: The emergence of self 459–549. awareness. Cambridge, MA: Harvard University Press. Feldman, R., Weller, A., Eidelman, A. I, & Sirota, L. Kagan, J. ~2001!. Biological constraint, cultural variety, ~2003!. Testing a family intervention hypothesis: The and psychological structures. Annals of the New York contribution of mother–infant skin-to-skin contact Academy of Sciences, 935, 177–190. ~Kangaroo Care! to family interaction and touch. Jour- Kagˇitçibasi,, C. ~1996!. Family and human development nal of Family Psychology, 17, 94–107. across countries: A view from the other side. Hills- Fenson, L., & Ramsay, D. S. ~1980!. Decentration and dale, NJ: Erlbaum. integration of the child’s play in the second year. Child Keller, H. ~2003!. Socialization for competence: Cultural Development, 51, 171–178. models of infancy. Human Development, 46, 288–311. Field, T. M. ~1992!. Infants of depressed mothers. Devel- Keller, H., Yovsi, R., Borke, J., Kartner, J., Jensen, H., & opment and Psychopathology, 4, 49–66. Papaligoura, Z. ~2004!. Developmental consequences Fogel, A. ~1993!. Developing through relationships: Or- of early parenting experiences: Self-recognition and igins of communication, self, and culture. Chicago: self-regulation in three cultural communities. Child University of Chicago Press. Development, 75, 1745–1760. Fogel, A., & Thelen, E. ~1987!. Development of early Keren, M., Feldman, R., Namdari-Weinbaum, I., Spitzer, expressive and communicative action: Reinterpreting S., & Tyano, S. ~2005!. Symbolic play in toddlers: the evidence from a dynamic system perspective. De- Relations with dyadic and triadic parent–child play velopmental Psychology, 23, 747–761. interaction styles, child characteristics, and marital Franz, M., Lensche, H., & Schmitz, N. ~2003!. Psycho- satisfaction. American Journal of Orthopsychiatry, 75, logical distress and socioeconomic status in single 599–607. mothers and their children in a German city. Social Klein, P. S. ~1996!. Early intervention: Cross-cultural Psychiatry and Psychiatric Epidemiology, 38, 59–68. experiences with a mediational approach. New York: Goldsmith, H. H., & Alansky, J. A. ~1987!. Maternal and Garland. infant temperamental predictors of attachment: A meta- Kochanska, G., & Kuczynski, L. ~1991!. Maternal auton- analytic review. Journal of Consulting and Clinical omy granting: Predictors of normal and depressed Psychology, 55, 805–816. mothers’ compliance and noncopmpliance with the Goldstein-Ferber, S., & Feldman, R. ~2005!. Delivery requests of five-year olds. Child Development, 62, pain and the development of mother–infant inter- 1449–1459. actions. Infancy, 8, 43–62. Kroes, M., Kalff, A. C., Steyaert, J., Kessels, A. G., Feron, Goodman, S. H., & Gotlieb, I. H. ~1999!. Risk for psy- F. J., Hendriksen, J. G., et al. ~2002!. A longitudinal chopathology in the children of depressed mothers: A community study: Do psychosocial risk factors and developmental model for understanding mechanisms child behavior checklist scores at 5 years of age pre- of transmission. Psychological Review, 106, 458–490. dict psychiatric diagnoses at a later age? Journal of Gottman, J., Markman, H., & Notarius, C. ~1977!. The the American Academy of Child & Adolescent Psychi- topography of marital conflict: A sequential analysis atry, 41, 955–963. 20 R. Feldman and S. Masalha

LeVine, R. A. ~2002!. Contexts and culture in psycholog- chopathology: Concepts and challenges. Develop- ical research. New Directions for Child Development, ment and Psychopathology, 12, 265–296. 96, 101–106. Sameroff, A. J. ~1997!. Understanding the social context Levy-Shiff, R. ~1994!. Individual and contextual corre- of early psychopathology. In J. Noshpitz ~Ed.!, Hand- lates of marital change across the transition to parent- book of child and adolescent psychiatry ~pp. 224– hood. Developmental Psychology, 30, 591–601. 235!. New York: Wiley. Locke, H., & Wallace, K. ~1959!. Short marital-adjustment Sameroff, A. J., & Fiese, B. ~2000!. Models of develop- and prediction tests: Their reliability and validity. Mar- ment and developmental risk. In C. H. Zeanah ~Ed.!, riage and Family Living, 21, 251–255. Handbook of infant mental health ~pp. 3–19!. New Luthar, S. S., & Cicchetti, D. ~2000!. The construct of re- York: Guilford Press. silience: Implications for interventions and social pol- Sameroff, A. J., Seifer, R., Barocas, R., Zax, M., & icies.DevelopmentandPsychopathology,12,857–885. Greenspan, S. ~1987!. Intelligence quotient scores of Lynch, M., & Cicchetti, D. ~1998!. An ecological- 4-year-old children: Social-environmental risk fac- transactional analysis of children and contexts: tors. Pediatrics, 79, 343–350. Longitudinal interplay among child maltreatment, com- Schwartzberg, N. S., & Dytell, R. S. ~1996!. Dual-earner munity violence, and children’s symptomatology. De- families: The importance of work stress and family velopment and Psychopathology, 10, 235–257. stress for psychological well-being. Journal of Occu- Mahoney, A., Jouriles, E. N., & Scavone, J. ~1997!. Mar- pational Health Psychology, 1, 211–223. ital adjustment, marital discord over childrearing, Seifer, R., Sameroff, A. J., Barrett, L. C., & Krafchuk, E. and child behavior problems: Moderating effects of ~1994!. Infant temperament measured by multiple ob- child age. Journal of Clinical Child Psychology, 26, servations and mother report. Child Development, 65, 415–423. 1478–1490. Mar’i, S. K., & Mar’i, M. M. ~1985!. The role of women Seiffge-Krenke, I., & Kollmar, F. ~1998!. Discrepancies as change agents in Arab society in Israel. In M. Shafir, between mothers’ and fathers’ perceptions of sons’ M. S. Mednick, D. Izraeli, & J. Bernard ~Eds.!, Women’s and daughters’problem behavior: A longitudinal analy- worlds: From the new scholarship ~pp. 122–153!.New sis of parent–adolescent agreement on internalizing York: Praeger. and externalizing problem behavior. Journal of Child Markus, H. R., & Kitayama, S. ~1991!. Culture and the Psychology and Psychiatry, 39, 687–697. self: Implications for cognition, emotion, and motiva- Sharma, D., & Fischer, K. W. ~1998!. Socioemotional tion. Psychological Review, 98, 224–253. development across cultures: Context, complexity, and McCune, L. ~1995!. A normative study of representa- pathways. New Directions for Child Development, 81, tional play at the transition to language. Developmen- 3–20. tal Psychology, 31, 198–206. Shweder, R. A. ~2003!. Why do men barbecue? Recipes Melstein-Damast, A., Tamis-LeMonda, C. S., & Born- for cultural psychology. Cambridge, MA: Harvard Uni- stein, M. H. ~1996!. Mother–child play: Sequential versity Press. interactions and the relations between maternal Shweder, R. A., Goodnow, J., Hatano, G., LeVine, R. A., beliefs and behaviors. Child Development, 67, Markus, H., & Miller, P. ~1998!. The cultural psychol- 1752–1766. ogy of development: one mind, many mentalities. In Murray, L., Sinclair, D., Cooper, P., Ducournau, P., Turner, R. M. Lerner ~Ed.!, Handbook of child psychology: P., & Stein, A. ~1999!. The socioemotional develop- Vol. 1. Theoretical models of human development ment of 5-year-old children of postnatally depressed ~pp. 865–937!. New York: Wiley. mothers. Journal of Child Psychology and Psychia- Slade, A. ~1987!. A longitudinal study of maternal involve- try, 40, 1259–1271. ment and symbolic play during the toddler period. Noll, L. M., & Harding, C. G. ~2003!. The relationship of Child Development, 21, 558–567. mother–child interaction and the child’s development Slade, A., & Wolf, D. P. ~1994!. Children at play. New of symbolic play. Infant Mental Health Journal, 24, York: Oxford University Press. 557–570. Stern, D. N. ~1995!. The motherhood constellation. New Pauli-Pott, U., Mertesacker, B., & Beckmann, D. ~2004!. York: Basic Books. Predicting the development of infant emotionality from Stuchbery, M., Matthey, S., & Barnett, B. ~1998!. Post- maternal characteristics. Development and Psychopa- natal depression and social supports in Vietnamese, thology, 16, 19–42. Arab, and Anglo-Celtic mothers. Social Psychiatry Piaget, J. ~1962!. Play, dreams, and imitation in child- and Psychiatric Epidemiology, 33, 483–490. hood. New York: Norton. Super, C. M., & Harkness, S. ~1986!. The developmental Rogoff, B. ~2003!. The cultural nature of human develop- niche: A conceptualization at the interface of child ment. New York: Oxford University Press. and culture. International Journal of Behavioral De- Rogoff, B., Mistry, J., Goncu, A., & Mosier, C. ~1993!. velopment, 9, 545–569. Guided participation in cultural activity by toddlers Tamis-LeMonda, C. S., & Bornstein, M. H. ~1994!. Spec- and caregivers. Monographs of the Society for Re- ificity in mother–toddler language-play relations across search in Child Development, 58~8!. the second year. Developmental Psychology, 30, Rubin, K., Burgess, K. B., Dwyer, K. M., & Hastings, 283–292. P. D. ~2003!. Predicting preschoolers’ externalizing Tomlinson, M., & Swartz, L. ~2003!. Imbalances in the behaviors from toddler temperament, conflict, and ma- knowledge about infancy: The divide between rich ternal negativity. Developmental Psychology, 39, and poor countries. Infant Mental Health Journal, 24, 164–176. 547–556. Rutter, M. ~1987!. Psychosocial resilience and protective Tramontina, S., Schmitz, M., Polanczyk, G., & Rohde, mechanisms. American Journal of Orthopsychiatry, L. A. ~2003!. Juvenile bipolar disorder in Brazil: Clin- 57, 316–331. ical and treatment findings. Biological Psychiatry, 53, Rutter, M., & Sroufe, L. A. ~2000!. Developmental psy- 1043–1049. Risk, adaptation, and culture 21

Triandis, H. C. ~1989!. Cross-cultural studies of individ- Voydanoff, P. ~1989!. Work and family: A review and ualism and collectivism. In R. A. Dienstbier ~Ser. Ed.! extended conceptualization. Journal of Social Behav- & J. J. Berman ~Vol. Ed.!, Nebraska Symposium on ior and Personality, 3, 1–22. Motivation ~Vol. 37, pp. 41–134!. Lincoln, NE: Uni- Vygotsky, L. S. ~1978!. Mind in society. Cambridge, MA: versity of Nebraska Press. Harvard University Press. Tronick, E. Z. ~1995!. Touch in mother–infant inter- Wachs, T. D., Bishry, Z., Sobhy, A., McCabe, G., Galal, action. In T. M. Field ~Ed.!, Touch in early develop- O., & Shaheen F. ~1993!. Relation of rearing environment ~pp. 53–65!. Mahwah, NJ: Erlbaum. ment to adaptive behavior of Egyptian toddlers. Child Tronick, E. Z., Morelli, G. A., & Ivey, P. K. ~1992!. The Development, 64, 586–604. Efe forager infant and toddler’s pattern of social rela- Weinberg, M. K., & Tronick, E. Z. ~1998!. The impact of tionships: Multiple and simultaneous. Developmental maternal psychiatric illness on infant development. Psychology, 28, 568–577. Journal of Clinical Psychiatry, 59, 53–61. Tucker, D. M. ~1992!. Developing emotions and cortical Werner, E. E., & Smith, R. S. ~1979!. An epidemiologic networks. In M. R. Gunnar & C. A. Nelson ~Eds.!, perspective on some antecedents and consequences of Developmental behavioral neuroscience: The Minne- childhood mental health problems and learning dis- sota Symposia on Child Psychology ~Vol. 24!. Hills- abilities: A report from the Kauai Longitudinal Study. dale, NJ: Erlbaum Journal of the American Academy of Child & Adoles- Van Bakel, H. J., & Riksen-Walraven, J. M. ~2002!. Par- cent Psychiatry, 18, 292–306. enting and development of one-year-olds: Links with Whiting, B. B., & Whiting, J. W. M. ~1975!. Children of parent, contextual, and child characteristics. Child De- six cultures: A psycho-cultural analysis. Cambridge, velopment, 73, 256–273. MA: Harvard University Press.