Cross-Cultural Patterns of Attachment

790

It was in Uganda, a former British protectorate in East Africa, that Mary Ainsworth (1967) began to create the famous tripartite classification sys-tem of infant–mother attachment relationships. In her short-term longitudinal field study, carried out in 1954–1955, she found three patterns of at-tachment behavior in a small sample of 28 infants. The “securely attached group” of 16 infants cried infrequently, seemed especially content when they were with their mothers, and used their mothers as a secure base from which to explore the environ-ment. The “insecurely attached group” of seven infants cried frequently, not only when left alone by their mothers but also in the mothers’ pres-ence; they cried to be picked up, cried when they were put down, and wanted continuous physical contact with their mothers, mingled with ambiva-lence about their mothers’ presence. A “nonat-tached” group of five infants responded similarly to their mothers and to other adults, were not upset about being left alone by their mothers, and avoided interaction with the mothers upon their return. From Ainsworth’s detailed case studies of these infants, it can be inferred that in the Strange Situation procedure (Ainsworth, Blehar, Waters, & Wall, 1978) they would have been classified as “avoidant.”

The Uganda study laid the foundation for not only the notion of attachment classification but also the development of the sensitivity con-struct that would later be captured in the still widely used Sensitivity–Insensitivity to Infant Signals and Communications Observational Scale (Ainsworth, Bell, & Stayton, 1971, 1974). Ainsworth described how it was not warmth that seemed to distinguish between the different at-tachment classifications because almost all of the observed mothers in the Uganda sample showed warmth in interaction with their babies, but rath-er the amount of caregiving for the baby, and the mother’s excellence as an informant about the baby. From Ainsworth’s descriptions it becomes clear that these two factors amount to a pattern of proximity and availability, interest in the baby, perceptiveness about the baby’s needs, and prompt responsiveness to the baby’s signals. Ainsworth’s work in a context with multiple caregivers in the non-Western setting of Uganda was the starting point from which she initiated a replication study in a Western setting, as described in her Balti-more study (Ainsworth & Wittig, 1969). Thus, the Uganda study provided all the main ingredi-ents for decades of research and theorizing about attachment and the role of parental sensitivity

Chapter 37

Cross-Cultural Patterns of AttachmentUniversal and Contextual Dimensions

Judi MesmanMarinus H. van IJzendoorn

Abraham Sagi-Schwartz

Cassidy3E_Book_2.indb 790 1/13/2016 2:20:21 PM

37. Cross-Cultural Patterns of Attachment 791

in predicting differences in the quality of attach-ment.

Bowlby (1969/1982) suggested that the for-mation of an attachment relationship between infants and their protective caregivers is the out-come of evolution; “inclusive fitness” (Trivers, 1974) was deemed to be facilitated by an etho-logically based innate bias to become attached to a conspecific (see Simpson & Belsky, Chapter 5, this volume). Therefore, a core element of attachment theory is the universality of this bias in infants to become attached, regardless of their specific cul-tural niche. Furthermore, secure attachment is expected to be normative within contexts that are not inherently threatening to human health, survival, and successful reproduction. In circum-stances not too strongly deviating from the “en-vironment of evolutionary adaptedness” (Bowlby, 1969/1982), it is assumed to be most adaptive to develop an attachment pattern that allows for exploration of the environment from the security of a safe haven in case of distress. However, if a cultural niche is harsh and socioeconomic circum-stances are extremely stressful for parents, infants may be prepared by their parents to develop, for example, an avoidant attachment pattern to cope with the socioeconomic stresses. In such a niche, the avoidant attachment pattern may well be normative in that it promotes inclusive fitness by stimulating a quantitative strategy of procre-ation with lower investment in more offspring at an earlier age (Simpson & Belsky, Chapter 5, this volume). Similarly, a resistant attachment pattern may be more common in environments that induce stress associated with intractable wars (Belsky, 2008). Evolution may not have equipped human beings with rigid behavioral strategies that would have made it difficult to adapt to changing (natural and social) environments (Hinde, 1982; Simpson & Belsky, Chapter 5, this volume). Ac-cording to Hinde’s (1982) position on conditional strategies, both mothers and babies are equipped to elicit not one set of interactions but a wide range of potential relationships compatible with specific environmental requirements. Neverthe-less, one may wonder whether the secure attach-ment pattern is the primary strategy for adapting to a social environment that is basically supportive of the infant, and whether the insecure strategies should be considered as secondary, in that they constitute deviating but adaptive patterns pro-voked by less supportive contexts (Main, 1990). Although insecure attachments may emerge to the benefit of inclusive fitness, the long-term costs for

the individual’s psychological and physical well-being might be considerable.

Regarding the antecedents of attachment, consistent responsive care is expected to foster a secure attachment bond, converging with observa-tional studies of both human and nonhuman pri-mates (e.g., Harlow, 1958; Nelson, Fox, & Zeanah, 2014; Van IJzendoorn, Bard, Bakermans-Kranen-burg, & Ivan, 2009). Even in rodents it is found that early-life stress is provoked by fragmentation and unpredictability of parental care and signal-ing, and such stress leads to persistent emotional dysfunction in offspring (Baram, Obenaus, Sand-man, & Small, 2012; Meaney, 2010). Indeed, the role of parental sensitivity is to provide a secure base from which the child can explore in the se-cure knowledge that the caregiver will be physical-ly and emotionally available in case of distress, and will alleviate this distress (Ainsworth et al., 1974, 1978). Genetics seems to play only a minor role in forming variations in attachment patterns (Baker-mans-Kranenburg & Van IJzendoorn, Chapter 8, this volume).

It can be argued that when the broader chil-drearing context allows for consistent responsive care, the formation of a secure attachment to this responsive caregiver is most likely to result in adaptive development and inclusive fitness. Similarly, in environments not deviating strongly from the environment of evolutionary adapted-ness (Bowlby, 1969/1982), secure attachment is expected to lead to positive child outcomes across domains given that attachment security has been found to predict more optimal basic human pre-requisites for adaptive functioning such as stress regulation (Lupien, McEwen, Gunnar, & Heim, 2009), nutritional status (Black & Aboud, 2011), and immune system functioning (Ehrlich, Miller, Jones, & Cassidy, Chapter 9, this volume; Miller, Chen, & Parker, 2011; see also Thompson, Chap-ter 16, this volume).

These universality assumptions have been widely tested and confirmed in North American and European samples, but the question is: What has nearly half a century of cross-cultural attach-ment research yielded regarding these issues? In this chapter, we describe and evaluate the cross-cultural attachment studies that have followed Ainsworth’s Uganda example. We limit our dis-cussion to cultures other than the Anglo-Saxon and European cultures because these are amply represented in other chapters in this volume (e.g., Fearon & Belsky, Chapter 14; Thompson, Chapter 16; Lyons-Ruth & Jacobvitz, Chapter


792 VI. SYSTeMS, CuLTure, AND CONTexT

29; DeKlyen & Greenberg, Chapter 28; Chase-Stovall & Dozier, Chapter 31). We discuss the following hypotheses (originally suggested by Van IJzendoorn, 1990):

1. The universality hypothesis: When given an opportunity, all infants without severe neu-rophysiological impairments will become at-tached to one or more specific caregivers.

2. The normativity hypothesis: The majority of in-fants are securely attached in contexts that are not inherently threatening to human health and survival.

3. The sensitivity hypothesis: Attachment security is dependent on childrearing antecedents, par-ticularly sensitive and prompt responses to in-fant attachment signals.

4. The competence hypothesis: Secure attachment leads to positive child outcomes in a variety of developmental domains.

attachment and Sensitivity in africa: a Network of Caregivers

In her Uganda study, Ainsworth (1967, 1977) described the development of attachment in a multiple-caregiver context. Since then, it has also been established that in a Western context, infants’ attachment to their mothers does not impede attachment relationships with other care-givers, such as fathers and professional caregivers in day care (Ahnert, Pinquart & Lamb, 2006; Goossens & Van IJzendoorn, 1990; Lucassen et al., 2011; Sagi et al., 1995). Thus, even in a chil-drearing environment in which mothers share their caregiving responsibilities with several other adults (and in Uganda also with older children), infants nevertheless become attached to their mothers and use them as a secure base to explore the world. The Uganda study, however, was rather small and exploratory, and certainly not represen-tative of the various African cultures, each with its own patterns of multiple caregiving. In this section we discuss studies of attachment in Af-rica conducted in the years following Ainsworth’s (1967) research. We place special emphasis on child development in a network of (child and adult) caregivers, in order to examine whether a multiple-caregiver environment is compatible with a unique attachment relationship between child and parent. Attachment in a network of multiple caregivers (or alloparents) is of crucial im-

portance because cross-cultural evidence indicates that in most societies, nonparental caretaking is either the norm or a common form (Hewlett & Winn, 2014; Hrdy, 2009), although this view has been contested (Bogin, Bragg, & Kuzawa, 2014), and paternal grandparents in a patrilineal society are likely to compete for the same resources as the child, challenging the often hypothesized benefits of alloparenting (Strassman, 2011). In this section we discuss ethnographic studies and studies using standardized instruments addressing attachment in the African context, which often involves mul-tiple caregivers.

Ethnographic Studies of Attachment and Sensitivity in Africa

We start by describing five studies of multiple caregiving in African societies that provide ethno-graphic descriptions of secure-base behavior and discriminative attachment to one or more caregiv-ers, and patterns of sensitive responsive care. Four of these studies concern hunter–gatherer societies, characterized by small seminomadic groups with a fluid group structure, absence of strict social rules, and flexible subsistence strategies (Lee & Daly, 1999). Bowlby (1969/1982) developed his evolu-tionary theory of attachment on the basis of specu-lations about child development and childrearing in the original environment in which the human species spent about 99% of its historical time as hunters and gatherers. In this “environment of evolutionary adaptedness” an infant would be pro-tected against predators and other dangers by stay-ing in close proximity to a protective adult. The four societies resembling this original way of living are described here.

Konner (1977, 2005) studied the !Kung San or Bushmen of northwestern Botswana, and described the general rules of childrearing in the !Kung society as reflecting indulgence, stimula-tion, and nonrestriction (Konner, 1977). The !Kung infants were fed whenever they cried and reached for the breast. At night they slept in close proximity to their mothers and were also fed on demand—even without the mothers’ awakening. An infant was carried around in a sling, giving the infant constant access to the mother’s breast and to decorative objects hanging around her neck. The infant could look around freely and experi-ence extensive physical and cognitive stimulation. The 2- and 3-year-old children studied by Konner were involved in multiage peer groups, spending



more time there than with their mothers, and readily establishing new bonds. Nevertheless, !Kung mothers were most comforting in response to infant crying, even when excluding breast feed-ing, always responding to crying bouts lasting lon-ger than 30 seconds (Konner, 2005). The !Kung study showed that there is room for both multiple attachment relationships and a preferred primary bond with the mother. It further highlights that sensitive responsiveness is an integral part of in-fant care within a multiple-caregiver context.

Morelli and Tronick (1991) studied the Efé of the Ituri forest in northeastern Zambia, a group that employs a system of multiple care-givers throughout the first few years of life, with newborns allowed to suckle other adult females even when their mothers are present (Tronick, Morelli, & Winn, 1987). Even the physical care is shared with other caregivers. Morelli and Tron-ick reported that the number of caregivers in the first 18 weeks amounted to 14.2 on average. This extremely dense social network led to prompt re-sponses to any sign of infant distress. During the second half of the first year, the infants began to show preference for the care of their own mothers, were more likely to protest against their mothers leaving, and wanted to be carried by their moth-ers on trips outside the camp. Morelli and Tron-ick pointed to the 1-year-olds’ interference with adults’ work activities, which prevented nonma-ternal caregivers from taking on caregiving re-sponsibility during work. They also noted that at night, only the mothers cared for their infants, and sleep was regularly interrupted by episodes of play-ful interaction exclusively between infants and their mothers (Morelli & Tronick, 1991), which may have facilitated the emergence of a special infant–mother bond. From the perspective of at-tachment theory, the night may be an especially stressful time, during which infants need a protec-tive caregiver the most (see description of Israeli communal kibbutzim, below). The Efé study sup-ports the universality hypothesis, also suggesting that sensitive responsiveness at night fosters at-tachment relationships.

The child care pattern among the Hadza of Tanzania also consists of multiple caregivers, with close, indulgent, and affectionate physical care-giver–infant proximity and responsiveness in the first years of life (Marlowe, 2005). Hadza women usually forage in groups containing only women and children, taking young infants with them and leaving toddlers at the camp. Marlowe (2005) found that mothers held infants more often than

all other caregivers combined, also showing how despite mothers’ willingness to hand their chil-dren over to other available caregivers, the child was not always equally willing, signifying a pref-erence for mothers as primary attachment figures. When a child started crying when handed over to another caregiver, the mother usually took the child back, which is consistent with the notions of secure-base behavior and maternal sensitive re-sponsiveness. In addition, Marlowe noted how all caregivers seemed to be equally sensitive to fussing and crying, but that mothers, compared with other caregivers, were generally much more effective at soothing and calming their child. Thus, descrip-tions of Hadza child care are consistent with the universality and sensitivity hypotheses.

For part of the year, the Bofi live in settle-ments near farming villages in the southwest re-gion of the Central African Republic, and live in camps in the forest for the other part of the year. In their study of 22 Bofi children who also experi-enced multiple careving, Fouts and Lamb (2005) reported that the average length of bouts of cry-ing and/or fussing was substantially longer when the mother was absent than when she was present. The difference was not statistically significant, but given the small sample size, this is not surprising. Calculating the effect size reveals a large one (d = 0.70), suggesting that in maternal absence, chil-dren try to elicit the mother’s presence and care, which is consistent with other descriptions of at-tachment behavior. Furthermore, crying children were more likely to be responded to when the mother was present than when she was absent, and context- and caregiver-specific patterns of sensitive responsiveness appeared to explain in-dividual differences in crying frequency. These findings confirm the universality hypothesis con-cerning a preferential attachment bond with the mother within a multiple-caregiving context, and they also support the sensitivity hypothesis, in that patterns of responsive caregiving, albeit specific to the ecology of each child’s family circumstances, were predictive of less crying.

In contrast to these four societies, the Hausa in Nigeria represent a polymatric culture that practices mostly agriculture. An average of four caregivers share the tasks of social, verbal, and playful interactions with children, but the biologi-cal mothers take almost complete responsibility for physical care activities, such as feeding and bathing (Marvin, VanDevender, Iwanaga, LeVine, & LeVine, 1977). In their ethnographic study of 18 infants, Marvin and colleagues found that



Hausa infants were almost always in close physi-cal contact with or close proximity to available adult caregivers, while not allowed to explore the wider environment alone because of the dangers involved. The high social density of the Hausa compound led to prompt adult or older sibling re-sponses to any infant attachment signals, such as crying. The Hausa caregivers therefore appeared to be indulgent and sensitive, and at the same time restrictive toward their infants. The restric-tion of locomotion also led to a different use of adult caregivers as a secure base: Hausa infants ex-plored their immediate environment in visual and manipulative ways, but only in close proximity to an attachment figure, and they ceased to explore as soon as the caregiver left. Nevertheless, Hausa infants clearly used adult caregivers as safe bases from which to explore, and they differentiated be-tween attachment figures and strangers. Further-more, all infants displayed attachment behavior toward an average of three to four different figures. Although being raised in an attachment relation-ships network, most Hausa infants were prin-cipally attached to one adult (generally the one holding the baby and interacting with him or her most often, whether or not this was the mother) to whom they addressed their attachment behavior most frequently (Marvin et al., 1977). This study provides evidence for the universality hypothesis and suggests that sensitive responsiveness is part of routine care for infants.

Standardized Observations of Attachment and Sensitivity in Africa

Only a few studies in multiple-caregiver African societies have used standardized procedures such as the Strange Situation procedure (Ainsworth et al., 1978) or the Attachment Q-Sort (AQS; Vaughn & Waters, 1990) to classify infants for-mally as securely or insecurely attached. This is, of course, due to the substantial practical barriers to conducting standardized procedures in contexts lacking laboratory facilities. The four studies that successfully did so are therefore very important and are discussed further here.

Childrearing among the Gusii of Kenya is characterized by alloparenting, but the division of tasks between mothers and other caregivers is rather strict. Mothers provide most of the physi-cal care and are responsible for their children’s health, whereas the activities of child caregiv-ers are limited to social and playful interactions

(Kermoian & Leiderman, 1986). In their study, Kermoian and Leiderman (1986) included 26 infant–caretaker dyads. Outside each mother’s hut, a modified Strange Situation procedure was implemented, with two separation–reunion epi-sodes for mother, caregiver, and stranger each. The extra separations were meant to compensate for the lack of a Strange Situation laboratory en-vironment. Gusii infants are used to being greeted with a handshake by their mothers and caretak-ers, and during the reunions, the Gusii infants an-ticipated the handshake in the same way as North American or European infants anticipate a hug. The secure Gusii infants would reach out to an adult with one arm, to receive the handshake en-thusiastically, whereas the insecure infants would avoid the adult or reach and then pull away after the adult approached. The distribution of patterns of attachment was comparable with Western find-ings: Secure attachment was found in 61 and 54% of infants to mothers and to nonmaternal care-givers, respectively. The authors concluded that the development of differential or person-specific attachment behaviors for “polymatric” infants is similar to that observed in “monomatric” West-ern societies; that is, infants do become uniquely attached to a protective adult caregiver, regard-less of the presence of one or more mother figures (Reed & Leiderman, 1981). This study also pro-vides evidence for the universality and normativ-ity hypotheses. Interestingly, positive nutritional or health status of the infants was related to the security of the infant–mother bond, whereas the infants’ cognitive development was related to the security of attachment to the nonmaternal care-giver. These findings might be expected given the clear task division between mothers (physical care, focus on health) and other caregivers (social and playful interactions), and provide evidence also for the competence hypothesis.

Attachment classifications based on the tra-ditional Strange Situation procedure were avail-able in a study of 26 mothers and their 1-year-old infants among the Dogon subsistence farmers in Mali, among whom maternal care is flexibly sup-plemented with care from siblings and other fam-ily members (True, 1994; True, Pisani, & Oumar, 2001). The Dogon mothers breast-fed their infants on demand very frequently and kept them in close proximity almost all the time. The percentage of secure infant–mother dyads was high (69%), whereas the avoidant classification appeared to be absent, and few resistant infant–mother dyads were found (8%). The percentage of disorganized



infants was high (23%) compared to percentages in normal Western samples (15–20%). The study supports the universality hypothesis in showing how the Strange Situation procedure is classifi-able with the ABCD (anxious-avoidant, secure, anxious-resistant, disorganized) coding system in an African culture, and supports the normativity hypothesis, because the majority of infants were securely attached (True et al., 2001). To explain the lack of avoidant attachments, True (1994) hy-pothesized that the Strange Situation procedure in the Dogon society may have been experienced as highly stressful instead of mildly stressful, forcing avoidant infants to seek proximity, and may also have increased the number of disorganized infants. Finally, True also found that infant attachment se-curity was related to caregiving patterns character-ized by sensitivity, a lack of frightening/frightened behaviors, and fewer violations of communica-tion coherence and cooperativeness. This seminal study among the Dogon therefore provides evi-dence for the universality hypothesis, the norma-tivity hypothesis, and the sensitivity hypothesis. Importantly, it also shows that contrary to some claims (e.g., Otto [2014], who calls the procedure “cruel”), it is possible to find creative and culture-sensitive ways of administering the Strange Situa-tion procedure that are not overly stressful for the infants.

There is some evidence that the extensive shared infant care in forager communities might influence infant behavior in separation–reunion situations. In a study among the Aka foragers of the Central African Republic, sensitive respon-siveness by nonmaternal caregivers negatively predicted infant’s distress during mother’s absence (Meehan & Hawks, 2013), and positively predict-ed infant engagement of mother upon her return (Meehan & Hawks, 2015), suggesting that infant behavior toward mothers in separation–reunion situations needs to be interpreted with consider-ation for the caregiving context.

In South Africa, Tomlinson, Cooper, and Murray (2005) studied attachment in a black sample of 98 mother–infant dyads, and assessed them at 2, 6, and 18 months postpartum. Families were living in Khayelitsha, an impoverished black settlement close to Cape Town. Only 5% lived in brick houses, and 49% of the houses were with-out modern plumbing, 58% of the families had no regular income, and 51% of the pregnancies were unplanned. The researchers undertook the chal-lenging job of conducting the Strange Situation, observing the quality of the home environment

and maternal sensitive responsiveness. Despite the poor living circumstances, the majority of infants were securely attached to their mothers (62%), although a rather large number of infants did de-velop a disorganized attachment (26%). Only 4% were avoidantly attached, and the remaining 8% were resistantly attached (Tomlinson et al., 2005). Furthermore, a remarkably high incidence of post-partum depression was found in the mothers (35% when infants were 2 months old), compared to similar samples in Western countries (with about 10% of mothers experiencing postpartum depres-sion). The presence of postpartum depression was strongly associated with attachment insecurity and disorganization, and sensitivity at 2 months, as well as at 18 months, predicted attachment se-curity significantly and independently of depres-sion. The study therefore provides support for the universality hypothesis, the normativity hypoth-esis, and the sensitivity hypothesis.

In another South African study by Minde, Minde, and Vogel (2006), 46 mother–child dyads (children ages 18-40 months) living in an im-poverished black township of Johannesburg were observed using the AQS. The results showed a roughly equal distribution of secure attachment (47%) and insecure attachment (53%), although we should mention that the AQS was not de-signed to yield attachment classifications. The relatively low rate of secure attachment is likely to be due to the extremely disadvantaged nature of the sample. Not only did all mothers report financial problems but also almost half of them reported a history of abuse, and about one-third had been placed away from home for prolonged periods before they were 5 years old (Minde et al., 2006). Furthermore, the sample’s depression and anxiety levels were just under the 90th percentile of the norm distribution, and similar to findings in Western countries, both the severity of financial problems and the experience of abuse were related to lower rates of attachment security (Minde et al., 2006). The AQS was successfully applied in this South African sample, providing evidence in sup-port of the universality hypothesis. The normativ-ity hypothesis was not properly tested because the AQS was not designed to yield classifications. The sensitivity hypothesis was indirectly confirmed, as higher rates of problems known to adversely affect maternal sensitivity (financial problems and a his-tory of abuse) were related to insecure attachment.

Family life in Africa has changed drastically in the past decades, due to the HIV/AIDS pandem-ic; many infants of infected mothers are born with



HIV. Peterson, Drotar, Olness, Guay, and Kiziri-Mayengo (2001) investigated a Ugandan sample of 35 HIV-positive mothers with or without AIDS, and 25 HIV-negative mothers, all with infants in their first year of life, 10 of whom were HIV-in-fected themselves. The researchers used the AQS to rate the attachment security of the infants dur-ing a 4-hour home visit. The average AQS score reflecting infants’ secure-base behavior in the sub-sample of 50 infants who were not infected with HIV (.35) was very similar to those found in other countries in different parts of the world (average of 34 samples = .31; Van IJzendoorn, Vereijken, Bakermans-Kranenburg, & Riksen-Walraven, 2004). Peterson and colleagues also found that 32% of the variance of AQS attachment security in the Ugandan sample was predicted by maternal affect, which consisted of ratings of expressivity and affective involvement, stimulation and activ-ity, and responsivity and sensitivity. Thus, almost half a century after the original Uganda study, the new Ugandan findings support Ainsworth’s (1967) observations of attachment bonds within a multiple-caregiver context (universality hypoth-esis) and her suggestion that sensitive parenting fosters a secure attachment relationship (the sen-sitivity hypothesis). Furthermore, the normativity hypothesis was supported by showing that security scores were similar to those found elsewhere.

attachment and Sensitivity in East asia

Countries in East Asia have been described as favoring the cultural model of interdependence (Kagitcibasi, 2007) and have been labeled as hav-ing a collectivistic culture (Oyserman, Coon, & Kemmelmeier, 2002). Both terms refer to a cultur-al context in which the importance of the group and social harmony are emphasized more than the needs of the individual, and in which adherence to group norms, filial piety, and self-sacrifice are expected (Hofstede, 1984). The label autonomous-relatedness has been coined to describe cultural groups that allow room for a focus on autonomy within an interdependent context, which general-ly characterizes urbanized or migrated groups that originally came from collectivistic cultural back-grounds but now reside in a more individualistic context (Kagitcibasi, 2007). Indeed, the rapid ur-banization, globalization, and increased wealth in

several regions of East Asia may have moved the social and psychological identities of its (urban) inhabitants closer to those of Western countries (e.g., Naftali, 2010). Consistent with this idea, a recent meta-analysis showed that urban Chinese parents display lower levels of authoritative par-enting (i.e., warmth and support for autonomy) than parents from North American countries, but not necessarily higher authoritarian parenting in terms of a focus on unquestioning obedience (Wang & Mesman, 2015).

Multiple caregivers are also common in most East Asian cultures, although nonparental care is mostly restricted to care by grandparents, who often coreside with the nuclear family and tend to provide extensive care for their grandchildren when both parents are working (e.g., Nauck & Suckow, 2006), and may in those situations be the primary attachment figures for the children. Grandparental coresidence is also fostered by housing shortages in urban areas and the absence of professional child care facilities for families in which both parents work outside the home (Goh & Kuczynski, 2010). Sibling care is very uncom-mon given the one-child policy in China and the very low fertility rates in Japan and South Korea. However, we note that demographic and cultural differences within the East Asian region are sub-stantial. For instance, in a country such as Indo-nesia, fertility rates are higher than elsewhere in the region (although declining fast), and the rela-tively recent colonial past of this country and its current predominant Muslim identity also sets it apart from countries such as China and Japan. A total of 18 studies using standardized observational instruments focusing on attachment are available from Papua New Guinea, China, Taiwan, Japan, South Korea, and Indonesia. We discuss these in some detail below.

Attachment among the Trobriander

In a unique study among 20 mother–toddler dyads living in the horticulturalist village of Tauwema on the Trobriand Islands (part of Papua New Guinea), written ethnographic records were col-lected, as well as standardized observations in the Strange Situation procedure (Grossmann, Grossmann, & Keppler, 2005). All observed tod-dlers showed attachment behavior and used their mothers as safe havens in times of distress in both field observations and the standardized Strange Situation, supporting the universality hypothesis.



In addition, all toddlers showed separation distress when involuntarily separated from their mothers (Grossmann et al., 2005). Secure attachment was normative, with 16 of the 20 infants falling into this category, supporting the normativity hypothe-sis. Maternal sensitivity was not systematically ob-served, but the ethnographic records suggest that mothers of insecure infants were less accessible to their toddlers, and slower and harsher in their re-sponses (Grossmann et al., 2005), which indicates tentative support for the sensitivity hypothesis.

Attachment in China and Taiwan

Nine studies of attachment have been conducted in normative urban samples in China, but only three were published in English, and only one obtained attachment classifications by coders of-ficially trained by experts (Archer et al., 2009). In that study, 62 infants and toddlers were exam-ined in the Strange Situation procedure, yielding a distribution of 57% secure, 13% avoidant, 16% resistant, and 13% disorganized (with forced clas-sification: B = 62%, A = 15%, C = 23%). The other two studies found a rate of 68% of securely attached infants (Ding, Xu, Wang, Li, & Wang, 2012; Hu & Meng, 1996), which is very similar to results from studies conducted in other parts of the world (Van IJzendoorn & Kroonenberg, 1988). The pattern within the insecure category was somewhat different, with Hu and Meng (1996) reporting 16% avoidant and 16% resistant (dis-organized was not coded), whereas Ding and col-leagues (2012) reported 7.5% avoidant and 21.8% resistant (and 2.5% disorganized). The six studies published in Chinese yielded the following distri-butions: B = 65%, A = 17%, C = 13% , D = 4% (Li et al., 2004); B = 73%, A = 11%, C = 7%, D = 9% (Liang, Chen, & Chen, 2000); B = 53%, A = 13%, C = 27%, D = 7% (Hu & Meng, 2003); B = 72%, A = 10%, C = 10%, D = 8% (Yue, Zhang, Chen, Liang, & Zhang, 2010); B = 68%, A = 6%, C = 26%, D = 0% (Ding et al., 2008); and B = 67%, A = 6%, C = 23%, D = 4% (Gu et al., 1997). As can be readily noticed, all studies conducted in China confirm both the universality hypothesis (by successfully classifying all children into one of the attachment categories) and the normativity hypothesis (by showing that the majority of chil-dren are securely attached).

What is also notable is the relatively low prevalence of avoidant attachment in the major-ity of these studies compared to the rates of avoid-

ance reported in Western studies (Van IJzendoorn & Kroonenberg, 1988; Van IJzendoorn, Sagi, & Lambermon, 1992). Hu and Meng (1996) ex-pressed doubts about the validity of the avoidant category, noting that the avoidant infants did not show stranger anxiety, and expressed indifference toward their mothers at reunion. The Chinese mothers’ stress on early independence in their in-fants, as well as their reliance on nonparental care-givers, may have been responsible for this pattern. In some cases, grandparents may have served as the primary attachment figures. An alternative inter-pretation, however, might be based on the subtle avoidant behaviors that are difficult for even well-trained coders to observe in infants. However, as we see later, the lower prevalence of the avoidant category is in line with findings of other studies in non-Western cultures and may reflect the cultural context of parenting.

In view of the (until recently) strict birth control policy in China and the traditional prefer-ence for a male child, it is important to note that the distributions of attachment classifications for male and female infants were virtually the same in the two studies that reported gender (Ding et al., 2012; Hu & Meng, 1996). It has been sug-gested that firstborn daughters who “survive” pa-rental choices regarding gender-driven abortion or postnatal abandonment are apparently welcome (Short, Fengying, Siyuan, & Mingliang, 2001) and have indeed been shown to have been treat-ed better than those with a male sibling (Fong, 2002). Regarding the role of grandparents, all but one family in the Hu and Meng study (1996) co-resided with one or more grandparents. The rate of grandparental coresidence was not reported in the Ding and colleagues study (2012), but they did report that infants with an insecure attachment to their mothers were also more likely to sleep with nonmaternal caregivers at night. In both studies, the possibility that a grandparent was the primary caregiver of the infants in those samples was sug-gested as a potential explanation for the results, namely, the high rate of avoidance and what ap-peared to be infant indifference toward the moth-ers (Hu & Meng, 1996), and the lower rate of se-curity in infants with more caregivers during the day and at night (Ding et al., 2012). The issue of sleeping arrangements is also discussed in the sec-tion on the Israeli kibbutz below.

Of the nine studies conducted in China, none examined the sensitivity hypothesis using observational studies. However, three studies ex-



amined the competence hypothesis and reported evidence to support this assumption. In one study, securely attached 2-year-old children exhibited fewer internalizing and externalizing problems than nonsecurely attached children both concur-rently and at ages 4 and 7 years (Yue et al., 2010). In the second study, securely attached infants pos-sessed more advanced cognitive skills than inse-curely attached infants, but there were no differ-ences in physical development (Ding, Wang, Li, Chi, & Xu, 2008), and the final study to examine outcomes of attachment found that secure toddlers had fewer behavior problems and displayed better self-regulation (Gu et al., 1997).

Posada and colleagues (2013) conducted a cross-cultural study involving nine countries, in-cluding a sample of 68 mothers and their 4-year-old children in Taiwan, which we discuss under the same heading as the studies from mainland China because of geographical and cultural prox-imity. Importantly, Taiwan does not have a one-child policy, but its fertility rate is also very low. In the Posada and colleagues study, 3-hour home observations were conducted, and the AQS was used to rate the child’s secure-base behavior. The results showed that the average security scores for the Taiwanese sample (.32) were similar to scores found in other studies (.31; see Van IJzendoorn et al., 2004).

The Japanese Case: Amae, Dependence, and Attachment

One of the most severe critiques of attachment theory has come in the form of an accusation of “cultural blindness” among attachment researchers to alternative conceptions of relatedness (Roth-baum, Weisz, Pott, Miyake, & Morelli, 2000). This “cultural blindness” accusation of a Western bias in attachment theory was based specifically on the Japanese case (rebutted in Van IJzendoorn & Sagi, 2001). The Japanese case can indeed be considered a real challenge to attachment theory’s universality, normativity, and sensitivity hypoth-eses, and we address these challenges below.

It has been argued that the concept of attach-ment may not be relevant to the Japanese culture, in which the idea of amae (Doi, 1973, 1992) seems to play a more prominent and effective role in describing family relationships and their societal implications (Emde, 1992). (Amae refers to rela-tional dependence, including aspects such as the desire for interpersonal closeness, the presence of

indulgence, and relying on the other person [Beh-rens, 2004].) Furthermore, it has been argued that “Amae has an advantage over attachment because it implies a psychological dependence” (Doi, 1989, p. 350). Vereijken (1996) asked eight native Japa-nese behavioral scientists to describe the concepts of amae, attachment, and dependence with the help of the AQS. The descriptions of amae and dependence were very similar, whereas the de-scriptions of attachment security and amae were not associated (Vereijken, 1996). Furthermore, when the descriptions of the ideal child accord-ing to Japanese mothers were compared with the Japanese experts’ definitions, only attachment se-curity (and not amae) appeared to be desirable (cf. Posada et al., 1995). Furthermore, an observation-al study of Japanese sojourners in the united states by Mizuta, Zahn-Waxler, Cole, and Hiruma (1996; see also Nakagawa, Teti, & Lamb, 1992) confirmed Vereijken’s (1996) conclusion that amae and at-tachment are orthogonal dimensions that can be reliably distinguished. Furthermore, Behrens (2004), a psychologist of Japanese origin, analyzed the use of the term and related terminology in natural Japanese discourse. she concluded that for many native Japanese speakers, amae has a nega-tive connotation involving social enforcement of obligations, and that amae lacks the biological roots of the attachment concept, is not associated with the regulation of stress, but can occur any time there is a desire or a motive on the part of the amae provider (Behrens, 2004; Behrens, Main, & Hesse, 2007). Because of the fundamental concep-tual, linguistic, and biological differences between the two constructs, attachment research in japan cannot be considered a challenge to the concept of amae, and amae cannot be considered a refutation of the concept of attachment.

Studies Using the Strange Situation Procedure in Japan

We now turn to observational studies of attach-ment in Japan. Three studies using the Strange Situation procedure to assess attachment secu-rity in Japan have been reported in the interna-tional literature. The first was conducted by Dur-rett, Otaki, and Richards (1984), who studied a middle-class sample of 39 intact families with their 12-month-old firstborns in Tokyo. This study showed a pattern of avoidant, secure, and resis-tant infant–mother attachments, consistent with the global distribution, with 13% A, 61% B, 18%



C, and 8% unclassifiable cases. The Tokyo study confirms the universality and normativity hypoth-eses because the authors did not report difficulties in applying the attachment coding system to this population, and because the normative “modal” category was secure. The unclassifiable cases could be a sign of culture-specific problems in applying the tripartite model but could also be due to the fact that the disorganized (D) classification was not yet available (notably, the D category was particu-larly instrumental in solving unclassifiable cases; Main & Solomon, 1990; see also Lyons-Ruth & Jacobvitz, Chapter 29, and Solomon & George, Chapter 18, in this volume). Furthermore, the mothers of securely attached infants indicated that they felt more supported by their husbands than did the mothers of avoidantly attached infants, but they did not differ in this regard from the moth-ers of resistantly attached infants (Durrett et al., 1984). This finding provides indirect support for the sensitivity hypothesis because partner support has been found to foster maternal sensitivity (e.g., Van Bakel & Riksen-Walraven, 2002).

The second Japanese study, conducted in Sapporo (Miyake, Chen, & Campos, 1985; Na-kagawa, Lamb, & Miyake, 1992; Takahashi, 1986), was unfortunately weakened by some methodological flaws and unclear reporting (see Van IJzendoorn & Sagi-Schwartz, 2008, for a full discussion of this study) and is therefore not discussed further in this section. The third Japa-nese study was also conducted in Sapporo with 43 mother–child dyads (Behrens, Hesse, & Main, 2007). Attachment distributions were based on Main and Cassidy’s sixth-year reunion assessment procedure (see Solomon & George, Chapter 18, this volume). In contrast to the previous Sapporo study, children’s three-way or forced “organized” attachment classification distribution did not dif-fer from the global distribution: 68% secure, 22% avoidant, 10% resistant (three children remained unclassified). The ABCD distribution was 49% B, 2% A, 0%C, and 49% D. The high proportion of D could be due to the fact that the non-Japanese coders had to judge maternal talk from translated transcripts and may have missed important cues from the tone of voice (Behrens et al., 2007). Fur-thermore, maternal attachment representations as measured with the Adult Attachment Interview (AAI; see Hesse, Chapter 26, this volume) pre-dicted child reunion classification (Behrens et al., 2007), with proportions very similar to matches reported worldwide. This is indirect support for the sensitivity hypothesis because adult attach-

ment representations have been found to predict maternal sensitivity (also in Asian samples, see, e.g., Liang et al., 2015), which in turn is known to predict infant attachment security (De Wolff & Van IJzendoorn, 1997). Thus, maternal sensitivity is likely to have (partially) mediated the relation between maternal attachment representations and infant attachment security. Notably, mothers’ unresolved attachment status strongly predicted child D status (r = .65).

Studies Using the Attachment Q-Sort in Japan

Three studies using the AQS have been con-ducted in Japan. Vereijken (1996) studied a Tokyo sample of 48 families with 14-month-old infants. More sensitive mothers had more secure children, and the association between sensitivity and at-tachment was impressively strong (all correla-tions—based on reports from independent cod-ers—were .59 or higher). In a follow-up study 10 months later, the association between sensitivity and attachment was replicated (Vereijken, 1996). In another study, 50 Japanese mothers and their preschool children were observed (Kazui, Endo, Tanaka, Sakagami, & Suganuma, 2000). The results showed that the children of secure moth-ers (as assessed with the AAI) had the highest security scores on the AQS, whereas children of unresolved mothers had the lowest AQS scores. The children of the dismissing and preoccupied mothers scored in between. The majority of the mothers were classified as secure (66%). Knowing that secure mothers generally have more sensitive interactions with their children than do insecure mothers, this outcome represents another indirect confirmation of attachment theory’s sensitivity hypothesis. The last Japanese study was conducted in the context of a larger cross-cultural context (Posada et al., 2013). In a sample of 45 infants, the average security score was the lowest (.19) of those reported for nine countries in total, but the majority of scores were positive, indicating a predominance of secure-base behavior. The low average score appeared to be predominantly due to lower scores on the Smooth Interactions with Mother and the Interactions with Other Adults scales (and not the scales reflecting physical prox-imity and contact with the mother). These find-ings may reflect the potential discomfort that Japa-nese mothers convey when observed in the home (Nakagawa, Lamb, et al., 1992).



Attachment and Sensitivity in South Korea

In the first and only study to use the Strange Situ-ation procedure in South Korea, Jin, Jacobvitz, Hazen, and Jung (2012) observed 87 mother–in-fant dyads. The results showed that 70% of the infants were classified as securely attached, 1% as avoidant, 17% as resistant, and 9% as disorganized. Jin and colleagues noted that in the reunion epi-sodes of the Strange Situation, the mothers were more likely to approach their infants immediately and sit beside them than was the case in Ain-sworth’s Baltimore sample, even when the infants were no longer distressed. The authors also dis-cussed the low prevalence of avoidant attachment in their sample by pointing to the generally lower percentage of infants in the avoidant category in other studies of East Asian samples, such as those from Indonesia (Zevalkink, Riksen-Walraven, & Van Lieshout, 1999) and China (e.g., Archer et al., 2015; Ding et al., 2012). Jin and colleagues (2012) speculated that a caregiving context in which mother–infant relationships are generally characterized by physical closeness, indulgence, and the discouragement of mother–infant separa-tion is very unlikely to produce avoidant children. As they put it, “Instead, when insecure attach-ment occurs, it is more likely to be resistant, since mothers are more likely to err on the side of being overly enmeshed, overprotective, or overinvolved, rather than rejecting” (p. 41). As we see later in this chapter, this may be relevant even for other non-Western cultures outside of East Asia. Finally, observations of maternal sensitivity during free play in this study revealed that infants of more sensitive mothers were more likely to be securely attached than infants of less sensitive mothers. Thus, this unique study in South Korea confirmed the universality hypothesis, the normativity hy-pothesis, and the sensitivity hypothesis.

Muslim Families in Indonesia

Indonesia is the fourth most populous country in the world and has the largest Islamic population of any country in the world. Zevalkink (1997) conducted the first attachment study on Islamic families of Sundanese Indonesian origin in West Java. Sundanese Indonesian children generally experience relatively extensive periods of close physical proximity to their mothers because they are carried in a carrying cloth, or slendang, during the first year. They are breast-fed on demand until 2 or 3 years of age and sleep in the same beds as

their mothers during the first 4 years of life. When an infant is fussy or cries, the mother promptly re-sponds with soothing or feeding. Sundanese Indo-nesian women, however, marry at a very young age, and their divorce rate is high. A stable income and permanent job are rare, which adds to the instabil-ity of family life. Poverty and health problems lead to rather high infant mortality (Zevalkink et al., 1999). Zevalkink and colleagues (1999) reported Strange Situation assessments of 46 children, ages 12–30 months. They also conducted extensive home observations on maternal sensitivity and observed maternal support in structured play ses-sions. The distribution of attachment classifica-tions was as follows: 52% secure, 7% avoidant, 20% resistant, and 22% disorganized (Zevalkink et al., 1999). More maternal support in structured play sessions was associated with attachment secu-rity; disorganized children received low maternal support. In addition, a higher-quality home care-giving environment was related to higher rates of secure attachment (Zevalkink, Riksen-Walraven, & Bradley, 2008). Thus, in these Muslim fami-lies, the universality hypothesis was confirmed because all children could be classified based on the Strange Situation procedure. Furthermore, a majority of children were securely attached (sup-porting the normativity hypothesis), and secure attachment was associated with the quality of caregiving in the predicted way (supporting the sensitivity hypothesis).

attachment and Sensitivity in Latin america

In this section, we examine the four hypotheses of attachment theory within the Latin American cultural context, mostly reflecting urban parent-ing, but also in one case of rural parenting. We found seven studies using standardized methods to assess attachment conducted in Latin American countries, including studies from Chile, Colombia, Peru, and Mexico. Each is discussed below.

Chile: A Study on Malnourished Infants

The first study using the strange situation proce-dure in a Latin American context was conducted in Chile by Valenzuela (1997) among 84 mother–infant dyads living in poverty, with half of the in-fants classified as chronically underweight. For the



total sample, only 27% of infants were categorized as securely attached, 26% as avoidant, 25% as re-sistant, and 21% as A/C (probably classifiable now as D). However, when distinguishing two groups based on infant nutritional status, the results re-vealed that the attachment distribution was much closer to the global distribution in the adequately nourished infants: 50% secure, 23% avoidant, 22% resistant, and 2% A/C (Valenzuela, 1997). Indeed, consistent with findings in the Gusii study (Kermoian & Leiderman, 1986), attachment secu-rity was significantly related to infant nutritional status. Finally, in the Chilean study, observations of maternal sensitivity were also significantly re-lated to attachment security in the expected di-rection (Valenzuela, 1997). This set of findings is consistent with studies showing that responsive feeding (which basically refers to sensitivity dur-ing feeding) is related to children’s more optimal weight status (Black & Aboud, 2011). In fact, both the World Health Organization and United Nations International Children’s Emergency Fund (Who/UNICEF; 2003) have incorporated respon-sive feeding into their guidelines for promoting healthy child development. In summary, the study provides evidence for all four key hypotheses for-mulated to represent the key tenets of attachment theory.

Studies from Colombia and Peru

The AQS was used to assess child secure-base be-havior in four studies from urban Colombia (the first three with partially overlapping samples), each of which revealed high average security scores (comparing to the average of .31 across 34 samples in Van IJzendoorn et al., 2004). First, an average security score of .43 was found by Posada and colleagues (2002) in a sample of 61 infants, and observed maternal sensitivity was positively related to attachment security in this study. Sec-ond, an average AQS score of .46 was found in a sample of 30 infants (Posada, Carbonell, Al-zate, & Plata, 2004). Third, Vaughn and col-leagues (2007) reported an average score of .48 for a sample of 25 Colombian infants; they found that maternal secure-base narratives were related to attachment security in the expected direction. Finally, Posada and colleagues (2013) report an average score of .32 in a group of 83 toddlers. In one of these four studies, ethnographic descrip-tions of maternal care were analyzed to examine the appropriateness of attachment theory’s con-

ceptualization of early care (Posada et al., 2004). The authors concluded that “most of the domains of maternal behavior found in this study matched well those identified by Ainsworth; our character-ization of maternal early care displays a direct rela-tion to that of attachment theory (e.g., Ainsworth et al., 1978)” (p. 516). Within the cross-cultural study by Posada and colleagues (2013), an urban sample of 30 mothers and preschoolers from Peru was also included, showing an average AQS se-curity scores of .30, similar to that found for the Colombian sample reported in the same article. Given the positive and moderate-to-high security scores found in the five studies described here, sup-port is provided for the universality and norma-tivity hypotheses. Furthermore, the results of two studies directly support the sensitivity hypothesis, and another provides indirect support by showing that maternal secure-base narratives (conceptually relevant to sensitive parenting) relate positively to infant secure attachment.

Attachment and Sensitivity in Urban and Rural Mexico

In a study by Gojman and colleagues (2012), in-tergenerational relations of attachment and care-giving quality were investigated in a sample of 35 urban, upper-middle-class Mestizo families, and a sample of 31 rural poor Indian families in Mexico. For the urban middle-class families, the Strange Situation procedure revealed a rate of 77% secure attachment, 3% avoidant, 3% resistant, and 14% disorganized (Gojman et al., 2012). In the poor rural sample, rates of security (32%) were clearly lower, and disorganization (35%) was more preva-lent than in the upper-middle-class urban sample. The finding that attachment security is less com-mon in poor rural samples than in more affluent urban samples is consistent with the family stress model, which describes how financial strains lead to parental stress, which in turn compromises fam-ily functioning in general and parenting quality in particular, leading to unfavorable child outcomes (Conger & Donnellan, 2007). This indirect evidence for the sensitivity hypothesis was cor-roborated by results showing a partial mediation model, with maternal autonomous attachment representations, as measured with the AAI, pre-dicting observed maternal sensitivity, which in turn predicted infant attachment security (Goj-man et al., 2012). As noted by the authors, these results clearly support the cross-cultural robustness of core features of attachment theory.



attachment in Israel

For the purpose of this chapter, we include Israel within the Anglo-Saxon and European cultures, and therefore do not review urban-normative sam-ples that have been studied with similar outcomes as those found in many Western countries (e.g., Sagi-Schwartz, Koren-Karie, Gini, Ziv, & Joels, 2002). We focus on two unique samples in Israel: infants raised in kibbutzim and the Arab minority.

Attachment and Collective Sleeping Practices

Following a visit to Israeli kibbutzim in the early 1950s, Bowlby (1951) noted the rich opportunities for research provided by kibbutz upbringing and predicted that this childrearing context, though clearly different from institutional care, might produce higher rates of attachment insecurity. The communal bedrooms in the children’s house were shared by three or four children, who each had private corners in which they kept their per-sonal things, and which were decorated according to each child’s preference. When collective sleep-ing was still in effect (starting a few months after birth), family time was in the afternoon and eve-ning, and children were returned to the children’s house for the night by their parents, who put them to bed. A caregiver or a parent then remained with them until the night watchwomen took over (Aviezer, Van IJzendoorn, Sagi, & Schuengel, 1994). Whereas most institutionalized childrear-ing in Western cultures involves multiproblem populations, the collective sleeping arrangements in the kibbutzim were designed for middle-class, well-functioning families. Although the system of multiple caregiving in the kibbutz was in many ways similar to multiple-caregiver contexts else-where (Rabin & Beit-Hallahmi, 1982), a world-wide sample of 183 societies showed that in none of them did infants sleep away from their parents (Barry & Paxton, 1971).

In support of the universality hypothesis, several studies of the attachments of communal-ly sleeping kibbutz children to their parents and caregivers revealed that they do form attachment bonds with their caregivers, and most often with their mothers (Aviezer & Sagi-Schwartz, 2008). The first communal kibbutz study conducted with the ABC system found 57% of securely attached infants (Sagi, Lamb, Lewkowicz, Shoham, Dvir, & Estes, 1985). Using the ABCD system in a more

recent sample, a rate of 26% secure infant–moth-er attachment was found (Sagi, Van IJzendoorn, Aviezer, Donnell, & Mayseless, 1994). In the ABC classification system D/B cases could have been classified as secure, hence inflating the rate of security. The rate of security in the communal sleep kibbutz was lower than that found in Israeli nonkibbutz samples (e.g., 66% in Koren-Karie, Oppenheim, Dolev, Sher, & Etzion-Carasso, 2002; 70% in Sagi, et al., 2002), and lower than that in a kibbutz sample with family-based sleeping (60% in Sagi et al., 1994). To rule out alternative explana-tions for the effect of communal sleeping arrange-ments, assessments were also made of the ecology of the children’s house during the day, maternal separation anxiety, infants’ temperaments, and mother–infant play interactions. The two groups (i.e., family-based and communal sleepers) were found to be comparable on all of these variables. Thus, it was concluded that collective sleeping, experienced by infants as a time when mothers were largely unavailable and inaccessible, was re-sponsible for the greater insecurity found in this group.

Interestingly, resistant attachment was found to be overrepresented in the communal-sleeping kibbutz samples (Sagi et al., 1985, 2002). Incon-sistent responsiveness was inherent in the reality of these infants, given that sensitive responding by a mother or caregiver during the day contrasted sharply with the presence of an unfamiliar person at night. This supports the sensitivity hypothesis because inconsistent responsiveness has been de-scribed as an important antecedent of resistant attachment (Ainsworth et al., 1978; Cassidy & Berlin, 1994). The rate of disorganized infants was also rather high for the second communal-sleep-ing kibbutz sample (44%), which could be attrib-uted to the unpredictable circumstances that these children experienced, especially during the nights. It should also be noted that in several Israeli stud-ies (Jewish sample: Koren-Karie et al., 2002; Sagi et al., 1985, 1994, 1997, 2002; Sagi-Schwartz, Van IJzendoorn, & Bakermans-Kranenburg, 2008; Arab sample: Zreik, 2014; see details below), the resistant classification appeared to be overrepre-sented and the avoidant classification to be under-represented compared to the global distribution (Van IJzendoorn & Kroonenberg, 1988). We spec-ulate that the resistant attachment strategy may be elicited in the context of continual threats to na-tional and personal security more readily than the avoidant strategy—the threat is external to the family. Parental preoccupation with these daily



stresses may lead to exaggerated overprotective-ness and impaired sensitivity to children’s attach-ment signals (see Belsky, 2008, for an evolutionary perspective).

Networks of Attachment Relationships

The kibbutz context has also made a unique con-tribution to the evaluation of the competence hypothesis. In a follow-up of most of the subjects in the Sagi and colleagues’ (1985) sample when they were 5 years old, Oppenheim, Sagi, and Lamb (1990) found that secure attachment to a nonpa-rental caregiver (the metapelet) during infancy was the strongest predictor of a child’s being empathic, dominant, independent, achievement-oriented, and behaviorally purposive in kindergarten. No significant associations were found between these socioemotional developments and the quality of children’s attachment to their parents, suggest-ing that the influence of attachment relationships may be domain-specific (see also the Gusii study described earlier). Because the infants’ relations with caregivers had been formed in the context of the infant house, those relationships were the best predictor of children’s socioemotional behav-ior in similar contexts. Furthermore, the extended network of infants’ attachments to the three types of caregivers (i.e., mothers, fathers, and metapelet) was a better predictor of later functioning than attachments in the family network only (attach-ments to mothers and fathers) and the attachment to mothers only (Sagi & Van IJzendoorn, 1996; Van IJzendoorn et al., 1992), which is consistent with findings in other samples (Howes, Rodning, Galluzzo, & Myers, 1988; Tavecchio & Van IJzen-doorn, 1987). This outcome may be interpreted as support for the integration model, which as-sumes that in a network of multiple-attachment relationships, secure attachments may compen-sate for insecure attachments in a linear way (Van IJzendoorn et al., 1992). Beyond kindergarten, however, networks of infant attachment relations did not contribute as much to the explanation of later behavior (Sagi-Schwartz & Aviezer, 2005). Instead, the data were more supportive of the hi-erarchy model (Van IJzendoorn et al., 1992), sug-gesting that early relations with mother as the pri-mary caregiver contributed most to later adaptive functioning at all ages (except in kindergarten), even in the kibbutz environment (Aviezer, Resn-ick, Sagi, & Gini, 2002).

Ecological Constraints on Intergenerational Transmission of Attachment

Intergenerational transmission of attachment refers to the process through which parents’ mental rep-resentations of their past attachment experiences influence their parenting behavior and the quality of their children’s attachment to them (Bowlby, 1969/1982; see Hesse, Chapter 26, this volume). In several studies of Western cultures, a concor-dance rate of about 75% has been found between the security of the parents’ mental representation of attachment and the security of the child–par-ent attachment (for a review, see Van IJzendoorn, 1995). In an Israeli study, Sagi and colleagues (1997) administered the AAI to 20 mothers from kibbutzim maintaining collective sleeping arrange-ments and to 25 mothers from home-sleeping kib-butzim. Parent–child concordance in attachment classifications was low for the communally sleep-ing group (40%), whereas it was rather high for the home-sleeping group (76%). Thus, contextual factors such as communal sleeping may override the influence of parents’ attachment representa-tion and their sensitive responsiveness. This find-ing indicates the limits of a context-free, universal model of transmission.

Attachment in the Arab Minority Residing in Israel

In a unique study on attachment in Arab families, infant–mother attachment relationships among Arabs living in Israel were assessed (Zreik, 2014; Zreik, Oppenheim, & Sagi-Schwartz, 2015). The sample included 85 Arab mother-infant dyads, and attachment was assessed using the Strange Situa-tion Procedure. Maternal sensitivity was assessed using the Emotional Availability Scales (Biringen, Robinson, & Emde, 2000) and the Maternal Be-havior Q-Sort (Pederson & Moran, 1995). Secure attachment was found to be the modal classifica-tion, with a rate of 67%, which supports the uni-versality and normativity hypotheses. Moreover, similar to the studies involving Jewish infants in Israel, the ambivalent classification (13%) also ap-peared to be overrepresented in this Arab sample, with an underrepresentation of the avoidant clas-sification (4%). Perhaps the fact that both groups live in a similar stressful geopolitical ecology con-tributed to this finding. Sixteen percent of the sample was classified as disorganized. The expect-ed association between sensitivity and attachment



security is not clear-cut when religion, maternal education, and measurement of sensitivity are taken into consideration, thus only partially sup-porting the sensitivity hypothesis (see Zreik, 2014, for details).

Universal and Contextual Dimensions of attachment

Figures 37.1 and 37.2 provide overviews of the re-sults presented in each of the earlier sections. In this section, we review the findings of cross-cultur-al attachment research in relation to the four core hypotheses of attachment theory, discussing both their universal and their contextual aspects.

The Universality Hypothesis

The universality hypothesis appears to be support-ed most strongly. In every study on attachment(-related) caregiving patterns in non-Anglo-Saxon and European countries, children were observed to show attachment behavior in stressful circum-stances and to have a preferential bond with one or more caregivers, based on both ethnographic descriptions and standardized assessments. The cross-cultural studies included here support Bowl-by’s (1969/1982) idea that attachment is indeed a universal phenomenon, and an evolutionary explanation seems to be warranted. Although in many cultures children grow up with a network of attachment figures, the parent or caregiver who takes responsibility for the care of a child during part of the day or the night becomes the favorite target of infant attachment behaviors. Not only the attachment phenomenon itself, but also the different types of attachment, appear to be pres-ent in various Western and non-Western cultures. Avoidant, secure, and resistant attachments have been observed in the African, East Asian, and Latin American studies, in samples ranging from hunter–gatherer societies characterized by high levels of alloparenting to affluent and deprived urban contexts. Even in the extremely diverging childrearing context of the Israeli kibbutzim with communal sleep at night, the differentiation be-tween secure and insecure attachment could be made.

These results do not, however, preclude cul-ture-specific patterns. Several studies have noted the attachment to multiple caregivers, and the identity of those caregivers depends on the specific

caregiving arrangements that are common in a given cultural context. For instance, infant secure attachment was found in relation to not only the infant’s mother but also nonmaternal caregivers (Goossens & Van IJzendoorn, 1990; Kermoian & Leiderman, 1986; Sagi et al., 1995). In addition, infants’ ways of expressing attachment and explo-ration behaviors have been found to vary depend-ing on cultural norms and customs. Hausa infants are generally physically restricted in their locomo-tion by their caregivers and are thus less free to explore the environment by themselves (Marvin et al., 1977). Instead, these infants explored their immediate environment in visual and manipula-tive ways, but only in close proximity to an attach-ment figure, and they ceased to explore as soon as the caregiver left. Nevertheless, the Hausa infants clearly used adult caregivers as secure bases from which to explore, and they differentiated between attachment figures and strangers (Marvin et al., 1977). Among the Gusii, culture-specific modes of attachment behavior were described, in that infants are accustomed to greeting their caregiv-ers with a handshake, which was differentially observed depending on attachment classifications (Kermoian & Leiderman, 1986), showing how local cultural customs can be taken into account in attachment research using standardized assess-ments.

The Normativity Hypothesis

The cross-cultural evidence for the normativity hypothesis is rather strong as well. In almost all cross-cultural studies included here, the majority of infants were classified as securely attached. Fig-ure 37.2 summarizes the attachment classification distributions found in the studies discussed earlier. When these are combined with AQS findings about the cross-cultural preference among experts, as well as mothers, for securely attached children, we may be confident that secure attachment is not just a North American invention or a West-ern ideal, but instead is a rather widespread and preferred phenomenon. As we described early on in this chapter, the category of secure attachments emerged from Ainsworth’s Uganda study, not from her Baltimore study (as is often suggested). In fur-ther support of the normativity hypothesis, Posada and colleagues (1995, 2013) showed that mater-nal beliefs about ideal child behavior overlapped considerably with attachment theory’s notion of the secure-base phenomenon, and that the secure-base phenomenon is clearly present in maternal


805

FIG

Ur

E 3

7.1.

Res

ults

of s

tudi

es te

stin

g at

tach

men

t the

ory

hypo

thes

es.

Afr

ica:

Bot

swan

a–1

(!K

ung

San;

Kon

ner,

1977

); C

entr

al A

fric

an R

epub

lic–1

(B

ofi; F

outs

& L

amb,

200

5); K

enya

–1 (

Gus

ii; K

erm

oian

& L

eide

rman

, 198

6); M

ali–

1 (D

ogon

; Tru

e et

al.,

200

1); N

iger

ia–1

(H

ausa

; Mar

vin

et a

l., 1

977)

; Sou

th A

fric

a, C

ape

Tow

n–1

(Tom

linso

n et

al.,

200

5); S

outh

Afr

ica,

Joha

nnes

burg

–1 (

Min

de e

t al.,

200

6); T

anza

nia–

1 (H

adza

; Mar

low

e, 2

005)

; Uga

nda–

1 (G

anda

; Ain

swor

th

& W

itti

g, 1

969)

; Uga

nda–

2 (P

eter

son

et a

l., 2

001)

; Zam

bia–

1 (E

fé; M

orel

li &

Tro

nick

, 199

1).

Eas

t A

sia:

Chi

na, B

eijin

g–1

(Hu

& M

eng,

199

6); C

hina

–Bei

jing–

2 (L

iang

et a

l., 2

000)

; Chi

na–B

eijin

g–3

(Hu

& M

eng,

200

3); C

hina

, Bei

jing–

4 (Y

ue e

t al.,

201

0); C

hina

, Gua

ngzh

ou–1

(Li

et a

l., 2

004)

; C

hina

, Sha

ngha

i–1

(Gu

et a

l., 1

997)

; Chi

na, S

hang

hai–

2 (D

ing

et a

l., 2

008)

; Chi

na, S

hang

hai–

3 (D

ing

et a

l., 2

012)

; Chi

na, X

i’an

(Arc

her e

t al.,

200

9); I

ndon

esia

–1 (

Zeva

lkin

k et

al.,

199

9); J

apan

, To

kyo–

1 (D

urre

tt e

t al.,

198

4); J

apan

, Tok

yo–2

(K

azui

et a

l., 2

000)

; Pap

ua N

ew G

uine

a–1

(Tro

bria

nd Is

land

s, G

ross

man

n et

al.,

200

5); J

apan

, Sap

poro

–1 (

Taka

hash

i, 19

86; N

akag

awa

et a

l., 1

992)

; Jap

an,

Sapp

oro–

2 (B

ehre

ns e

t al.,

200

7); J

apan

, Sap

poro

–3 (

Posa

da e

t al.,

201

3); S

outh

Kor

ea–1

(Ji

n et

al.,

201

2); T

aiw

an–1

(Po

sada

et a

l., 2

013)

.L

atin

Am

eric

a: C

hile

–1 (

Val

enzu

ela,

199

7); C

olom

bia–

1 (P

osad

a et

al.,

200

2, 2

004;

Vau

ghn

et a

l., 2

007)

; Col

ombi

a–2

(Pos

ada

et a

l., 2

013)

; Mex

ico–

1 (G

ojm

an e

t al.,

201

2); P

eru–

1 (P

osad

a et

al.,

201

3).

Isra

el: I

srae

l–1

(firs

t kib

butz

coh

ort c

omm

unal

slee

ping

; Sag

i et a

l., 1

985)

; Isr

ael–

2 (s

econ

d ki

bbut

z coh

ort c

omm

unal

slee

ping

; Sag

i et a

l., 1

994)

; Isr

ael–

3 (s

econ

d ki

bbut

z coh

ort f

amily

slee

ping

; Sag

i et a

l.,

1994

); Is

rael

–4 (

first

urb

an c

ohor

t; Sa

gi e

t al.,

200

2); I

srae

l–5

(sec

ond

urba

n co

hort

; Kor

en–K

arie

et a

l., 2

002)

; Isr

ael–

6 (A

rabs

; Zre

ik, 2

014)

.


Laura

Sticky Note

AU: Can you provide better art for this figure and the next? Very hard to read.

806

FIG

Ur

E 3

7.2.

Res

ults

of s

tudi

es e

xam

inin

g th

e di

stri

buti

on o

f att

achm

ent c

lass

ifica

tion

s. Se

e pa

ragr

aphs

that

follo

w th

e ca

ptio

n in

Fig

ure

37.1

.



descriptions of the behaviors of their own children across different cultural contexts (Posada et al., 1995).

Even though secure attachment appears to be the norm across cultures, there are variations in the rates of secure attachment reported in the studies in non-Anglo-Saxon and European con-texts. For instance, rates of security were particu-larly low in the rural poor Mexican sample (32%; Gojman et al., 2012) and in the undernourished Chilean sample (7%; Valenzuela, 1997). These findings emphasize the importance of socioeco-nomic circumstances in shaping family life and parenting patterns. This is consistent with the family stress model, which describes how adverse economic circumstances interfere with optimal parenting through their influence on parental stress, ultimately leading to negative effects on child development (Conger & Donnellan, 2007; see also McLoyd, 1998). Not having the economic resources to reliably provide basic care for children in terms of nutrition, health care, safe housing, and clothing is likely to be universally stressful for parents. We suggest investing in studies on the relation between economic stress and attachment-related processes across cultures to evaluate the family stress model. In many studies comparing parent–child relations in different cultural groups, cultural and socioeconomic variables are con-founded (e.g., Bakermans-Kranenburg, Van IJzen-doorn, & Kroonenberg, 2004). Therefore, as also shown by the Mexican and Chilean studies, it is pivotal for future cross-cultural studies of attach-ment to examine different socioeconomic groups within cultures.

The normativity hypothesis does not address the distribution of classifications within the in-secure category. However, interesting patterns of the prevalence rates of avoidant versus resistant attachment emerge from the studies discussed in this chapter (see Figure 37.2). Whereas results from Anglo-Saxon and European samples gener-ally show a higher rate of A than C classifications, with A mostly (well) above 20% (Van IJzendoorn et al., 1992; Van IJzendoorn & Kroonenberg, 1988), the range of avoidant attachment classifi-cations in the studies discussed in this chapter lies between 0 and 23%, with only two studies report-ing a rate higher than 20% (Behrens et al., 2007; Valenzuela, 1997), and half of the studies reporting a rate lower than 10% in a variety of non-Western countries. Because avoidant attachment is gener-ally associated with consistently unresponsive par-enting, we hypothesize that in countries empha-

sizing relatedness, customs of highly proximal and indulgent parenting may be responsible for the low rates of avoidance. As Jin and colleagues (2012) noted, insensitive care in these cultures may be more likely to be intrusive rather than unavailable (and therefore potentially inconsistently sensitive as well), which is generally found to foster resis-tant attachment rather than avoidance.

The Sensitivity Hypothesis

The sensitivity hypothesis was not tested in most of the studies, but when examined, the data almost invariably supported it (except for the procedur-ally flawed first Sapporo study). Indirect support for the sensitivity hypothesis comes from studies reporting that secure attachment was related to factors known to be associated with maternal sen-sitivity. Furthermore, several ethnographic studies provide accounts of sensitive parenting as appar-ently normative in hunter–gatherer societies. It is also interesting to note that in a sample of no less than 1,150 Asian families living in the United States, 87% of whom were born in Asia, obser-vations of maternal sensitivity were significantly related to observations of infant attachment se-curity (Huang, Lewin, Mitchell, & Zhang, 2012). Further support for the cross-cultural relevance of the sensitivity construct was provided by Mesman and colleagues (2015), who reported strong con-vergence between maternal beliefs about the ideal mother and attachment theory’s description of the sensitive mother across 26 cultural groups from 15 countries (see also Emmen et al., 2012).

The potentially universal role of sensitive caregiving in fostering attachment security is con-sistent with the idea that when the socioeconomic context allows for sensitive care, the formation of a secure attachment to a sensitive caregiver is most likely to result in adaptive functioning and inclusive fitness. Notably, the function and outcome of the parental behaviors are more im-portant than their specific manifestation, which is consistent with the form–function distinction drawn by Bornstein, Cote, Haynes, Suwalsky, and Bakeman (2012), which states that different pa-rental behaviors can have the same function in different cultural contexts. As such, the sensitiv-ity construct reflects an organizational approach to the attachment relationship in that neither parent nor child behaviors can be captured by a predefined and exhaustive set of concrete behav-iors (Sroufe & Waters, 1977), but reflect a history of patterns of interaction rather than individual



behaviors (Weinfield, Sroufe, Egeland, & Carl-son, 2008). Indeed, there is evidence that whereas the rates of contingent responding are very simi-lar across very different cultures, the modalities through which responsiveness is channeled are culture-dependent (e.g., Kärtner, Keller, & Yovsi, 2010). In some cultures, appropriate responding to infant vocalization may consist of touching or stroking the infant, whereas in other cultures it may be to imitate the sound that the infant made or to smile at the infant (Kärtner, Keller, & Yovsi, 2010; Keller, Borke, Yovsi, Lohaus, & Jensen, 2005; Keller et al., 2009). Similarly, infant distress may be dealt with appropriately in many different ways. For instance, soothing through nursing/feed-ing is much more common in non-Western than in Western cultures (e.g., Ainsworth, 1967; True et al., 2001). Many more concrete aspects of re-sponsiveness may be examined to fine-tune our understanding of what it means to be sensitive in different cultures both in infancy and beyond, to paint a comprehensive picture of the similarities and differences in the concrete behavioral charac-teristics of sensitive responsiveness across cultures and their particular role in fostering secure attach-ment.

The Competence Hypothesis

The competence hypothesis has been tested only sporadically in cross-cultural research; this con-curs with the relative lack of Western studies on the association between attachment and (later) competence (Groh et al., 2014; Sroufe, Egeland, Carlson, & Collins, 2005; Van IJzendoorn, Dijk-stra, & Bus, 1995). In the Gusii study (Kermoian & Leiderman, 1986), the nutritional status of the secure infants was better than that of the insecure infants. This outcome has been replicated by Va-lenzuela (1997) in her Chilean study of under-nourished infants. Although the relation between attachment and health status is truly remarkable, it is still not clear whether attachment security serves only as the cause and nutritional status only as the effect. It may be that more healthy infants evoke more care in general and more sensitive care in particular, especially in situations of eco-nomic deprivation when parents are forced to be selective in their investment of time and energy (Finerman, 1994; Scheper-Hughes, 1993). A more intricate causal pattern cannot be completely ex-cluded on the basis of the correlational evidence

that Kermoian and Leiderman (1986) and Va-lenzuela (1997) have provided. The Dogon study (True, 1994; True et al., 2001) does not allow for differentiation between cause and effect, either. Only the Israeli studies on children raised in a kib-butz and city children showed some longitudinal relations between secure attachment and future adaptive functioning (Aviezer, Sagi, & Van IJzen-doorn, 2002; Gini, Oppenheim, & Sagi-Schwartz, 2007; Oppenheim, Koren-Karie, & Sagi-Schwartz, 2007; Sagi- Schwartz & Aviezer, 2005). We should therefore conclude that the cross-cultural support for the competence hypothesis is still insufficient. Given the convincing support for the other three key hypotheses postulated by attachment theory, it seems more than feasible to extend attachment re-search in non-Western countries to children’s de-velopmental outcomes. Furthermore, the concept of attachment networks may be fruitfully applied in studies of the competence hypothesis in societ-ies where multiple caregivers are the norm.

Conclusions

Our analysis and integration of cross-cultural at-tachment research suggest a balance between universal trends and contextual determinants. Attachment theory without contextual compo-nents is as difficult to conceive of as attachment theory without a universalistic perspective. If all infants across cultures used the same fixed strate-gies to deal with attachment challenges, it would leave no room for adaptation to dynamic changes in the environment (Hinde & Stevenson-Hinde, 1990) and to the constraints imposed by different developmental niches (DeVries, 1984; Harkness & Super, 1992, 1996; LeVine & Miller, 1990; see Simpson & Belsky, Chapter 5, this volume). With-out variation, selection of optimal behavioral strat-egies would become obsolete (Darwin, 1859/1985; see Van IJzendoorn, Bakermans-Kranenburg, & Sagi-Schwartz, 2006, for an evolutionary attach-ment model integrating universal and contextual dimensions of attachment).

The studies reviewed in this chapter em-phasize the importance of investigating attach-ment within the wider social networks in which children grow and develop. In Western as well as non-Western cultures, most children commu-nicate with several attachment figures, including siblings. Examining the competence hypothesis



only on the basis of infant–mother attachment may decrease predictive power substantially, and more conceptual and empirical work is needed to determine how experiences with different attach-ment figures become organized to form a coherent internal working model (Sagi-Schwartz & Aviezer, 2005). Furthermore, several studies reviewed here emphasize the importance of including variations in the socioeconomic context within countries, so that attachment patterns can be elucidated from both a cultural and a resource perspective.

Last, the current cross-cultural database is almost absurdly small compared to the domain that should be covered. Data on attachment in a populous country such as India and most Islamic countries are still lacking, and large parts of Africa, Asia, and Latin America are uncharted territories with respect to the development of attachment. Although the studies including samples from these parts of the world represent admirable contribu-tions to the attachment literature, they cannot be considered to be representative of their respective continents. This lack of studies outside of North America and Europe, of course, is partly due to the difficulties of conducting standardized research in countries where facilities and resources are lacking. And even if researchers manage to, for instance, conduct the Strange Situation procedure, in many cases the coding is done by researchers who have not been formally trained by experts, which makes the quality of the classifications unclear. It is also the task of Western attachment researchers to pro-vide adequate and affordable training opportuni-ties to those researchers in other parts of the world who want to study attachment.

In our chapter of this third edition, which includes attachment data from five new countries or cultures compared to the previous edition (Van IJzendoorn & Sagi-Schwartz, 2008), we pay spe-cial attention to cross-cultural beliefs about at-tachment and sensitivity, and we provide more details on attachment and sensitivity in Latin America and with Arabs living in Israel. What has not changed since 2008 is that the available cross-cultural studies have not refuted the bold conjectures of attachment theory about the uni-versality of attachment, the normativity of secure attachment, the link between sensitive caregiving and attachment security, and the competent child outcomes of secure attachment. In fact, taken as a whole, the studies are remarkably consistent with the theory. Until further notice, attachment theo-ry may therefore claim cross-cultural validity.

acknowledgments

Support for the preparation of this chapter was provided by the Spinoza Prize of the Netherlands Foundation for Scientific Research (NWO) to Marinus H. van IJzendoorn, and by NORFACE (New Opportunities for Research Funding Agen-cy Cooperation in Europe, Grant No. 292) to Judi Mesman.

references

Ahnert, L., Pinquart, M., & Lamb, M. E. (2006). Secu-rity of children’s relationships with nonparental care providers: A meta-analysis. Child Development, 74, 664–679.

Ainsworth, M. D. S. (1967). Infancy in Uganda: Infant care and the growth of love. Baltimore: Johns Hopkins University Press.

Ainsworth, M. D. S. (1977). Infant development and mother–infant interaction among Ganda and Ameri-can families. In P. H. Leiderman, S. R. Tulkin, & A. H. Rosenfeld (Eds.), Culture and infancy (pp. 119–150). New York: Academic Press.

Ainsworth, M. D. S., Bell, S. M., & Stayton, D. J. (1971). Individual differences in Strange Situation behavior of one year olds. In H. R. Schaffer (Ed.), The origins of human social relations (pp.17–58). New York: Academic Press.

Ainsworth, M. D. S., Bell, S. M., & Stayton, D. J. (1974). Infant–mother attachment and social devel-opment: Socialisation as a product of reciprocal re-sponsiveness to signals. In M. P. M. Richards (Ed.), The introduction of the child into a social world (pp. 9–135). London: Cambridge University Press.

Ainsworth, M. D. S., Blehar, M. C., Waters, E., & Wall, S. (1978). Patterns of attachment. Hillsdale, NJ: Erlbaum.

Ainsworth, M. D. S., & Wittig, B. A. (1969). Attach-ment and exploratory behavior of one year olds in a strange situation. In B. M. Foss (Ed.), Determinants of infant behavior (Vol. 4, pp. 113–136). London: Methuen.

Archer, M., Steele, M., Lan, J., Jin, X., Herreros, F., & Steele, H. (2015). Attachment between infants and mothers in China: Strange Situation procedure find-ings to date and a new sample. International Journal of Behavioral Development, [Epub ahead of print].

Aviezer, O., Resnick, G., Sagi, A., & Gini, M. (2002). School competence in young adolescence: Links to early attachment relationships beyond concurrent self-perceived competence and representations of re-lationships. International Journal of Behavioral Develop-ment, 26, 397–409.

Aviezer, O., Sagi, A. & Van IJzendoorn, M.H. (2002). Collective sleeping for kibbutz children: An experi-ment in nature predestined to fail. Family Process, 41, 435–454.



Aviezer, O., & Sagi-Schwartz, A. (2008). Attachment and non-maternal care: Towards contextualizing the quantity versus quality debate. Attachment and Human Development, 10, 275–285.

Aviezer, O., Van IJzendoorn, M. H., Sagi, A., & Schuen-gel, C. (1994). “Children of the dream” revisited: 70 years of collective early child care in Israeli kibbut-zim. Psychological Bulletin, 116, 99–116.

Bakermans-Kranenburg, M. J., Van IJzendoorn, M. H., & Kroonenberg, P. M. (2004). Differences in attach-ment security between African-American and white children: Ethnicity or socio-economic status? Infant Behavior and Development, 27, 417–433.

Barry, H. I., & Paxton, L. M. (1971). Infancy and early childhood: Cross-cultural codes 2. Ethnology, 10, 466–508.

Behrens, K. (2004). A multifaceted view of the concept of amae: Reconsidering the indigenous Japanese con-cept of relatedness. Human Development, 47, 1–27.

Behrens, K. Y., Hesse, E., & Main, M. (2007). Moth-ers’ attachment status as determined by the Adult Attachment Interview predicts their 6-year-olds’ re-union responses: A study conducted in Japan. Devel-opmental Psychology, 43, 1553–1567. [Verified online]

Belsky, J. (2008). War, trauma and children’s develop-ment: Observations from a modern evolutionary per-spective. International Journal of Behavioral Develop-ment, 32, 260–271.

Biringen, Z., Robinson, J. L., & Emde, R. N. (2000). Ap-pendix B: The Emotional Availability Scales (3rd ed.; an abridged infancy/early childhood version). Attach-ment and Human Development, 2, 256–270.

Black, M. M., & Aboud, F. E. (2011). Responsive feed-ing is embedded in a theoretical framework of respon-sive parenting. Journal of Nutrition, 141, 490–494.

Bogin, B., Bragg, J., & Kuzawa, C. (2014). Humans are not cooperative breeders but practice biocultural re-production. Annals of Human Biology, 41, 368–380.

Bornstein, M. H., Cote, L. R., Haynes, O. M., Suwalsky, J. T. D., & Bakeman, R. (2012). Modalities of infant–mother interaction in Japanese, Japanese American immigrant, and European American dyads. Child De-velopment, 83, 2073–2088.

Bowlby, J. (1951). Maternal care and mental health. Ge-neva: World Health Organization.

Bowlby, J. (1982). Attachment and loss: Vol. 1. Attach-ment. New York: Basic Books. (Original work pub-lished 1969)

Cassidy, J., & Berlin, L. J. (1994). The insecure/am-bivalent pattern of attachment: Theory and research. Child Development, 65, 971–991.

Conger, R. D., & Donnellan, M. B. (2007). An inter-actionist perspective on the socioeconomic context of human development. Annual Review of Psychology, 58, 175–199.

Darwin, C. (1985). On the origin of species by means of natural selection, or the preservation of favoured races in the struggle for life. Harmondsworth, UK: Penguin. (Original work published 1859)

De Wolff, M., & Van IJzendoorn, M. H. (1997). Sensi-tivity and attachment. A meta-analysis on parental antecedents of infant attachment. Child Development, 68, 571–591.

DeVries, M. W. (1984). Temperament and infant mor-tality among the Masai of East Africa. American Jour-nal of Psychiatry, 141, 1189–1194.

Ding, Y. H., Wang, Z. Y., Li, H. R., Chi, Z. Y., & Xu, H. (2008). Study of 62 infants’ attachment in Shanghai and their relationship with infants’ temperament and cognitive development. Chinese Child Healthcare Jour-nal, 16, 163–165.

Ding, Y. H., Xu, X., Wang, Z., Li, H., & Wang, W. (2012). Study of mother–infant attachment patterns and influence factors in Shanghai. Early Human De-velopment, 88, 295–300.

Doi, T. (1973). The anatomy of dependence. New York: Kodansha.

Doi, T. (1989). The concept of amae and its psychoana-lytic implications. International Review of Psychoanaly-sis, 16, 349–354.

Doi, T. (1992). On the concept of amae. Infant Mental Health Journal, 13, 7–11.

Durrett, M. E., Otaki, M., & Richards, P. (1984). At-tachment and the mother’s perception of support from the father. International Journal of Behavioral De-velopment, 7, 167–176.

Emde, R. N. (1992). Amae, intimacy, and the early moral self. Infant Mental Health Journal, 13, 34–42.

Emmen, R. A. G., Malda, M., Mesman, J., Ekmekci, H., & Van IJzendoorn, M. H. (2012). Sensitive parenting as a cross-cultural ideal: Sensitivity beliefs of Dutch, Moroccan, and Turkish mothers in the Netherlands. Attachment and Human Development, 14, 601–619.

Finerman, R. (1994). “Parental incompetence” and “se-lective neglect”: Blaming the victim in child survival. Social Science and Medicine, 40, 5–15.

Fong, V. L. (2002). China’s one-child policy and the em-powerment of urban daughters. American Anthropolo-gist, 104, 1098–1109.

Fouts, H. N., & Lamb, M. E. (2005). Weaning emotion-al patterns among the Bofi foragers of Central Africa: The role of maternal availability and sensitivity. In M. E. Lamb & B. S. Hewlett (Eds.), Hunter–gatherer childhoods: Evolutionary, developmental, and cultural perspectives (pp. 19–64). New Brunswick, NJ: Trans-action.

Gini, M., Oppenheim, D., & Sagi-Schwartz, A. (2007). Negotiation styles in mother–child narrative co-construction in middle childhood: Associations with early attachment. International Journal of Behavioral Development, 31, 149–160.

Goh, E. C. L., & Kuczynski, L. (2010). “Only children” and their coalition of parents: Considering grandpar-ents and parents as joint caregivers in urban Xiamen, China. Asian Journa of Social Psychology, 13, 221–231.

Gojman, S., Millán, S., Carlson, E., Sánchez, G., Rod-arte, A., González, P., et al. (2012). Intergenerational relations of attachment: A research synthesis of



urban/rural Mexican samples. Attachment and Human Development, 14, 553–566.

Goossens, F. A., & Van IJzendoorn, M. H. (1990). Qual-ity of infants’ attachments to professional caregivers: Relation to infant–parent attachment and day-care characteristics. Child Development, 61, 832–837.

Groh, A. M., Fearon, R. P., Bakermans-Kranenburg, M. J., Van IJzendoorn, M. H., Steele, R. D., & Roisman, G. I. (2014). The significance of attachment security for children’s social competence with peers: A meta-analytic study. Attachment and Human Development, 16, 103–136.

Grossmann, K. E., Grossmann, K., & Keppler, A. (2005). Universal and culture-specific aspects of human be-havior: The case of attachment. In W. Friedlmeier, P. Chakkarath, & B. Schwarz (Eds.), Culture and human development: The importance of cross-cultural research for the social sciences (pp. 71–91). New York: Psychol-ogy Press.

Gu, H., Cen, G., Li, D., Gao, X., Li, Z., & Chen, X. (1997). Two-year-old children’s social behaviour de-velopment and related family factors. Psychological Science, 20, 519–524.

Harkness, S., & Super, C. M. (1992). Shared child care in East Africa: Sociocultural origins and developmen-tal consequences. In M. E. Lamb, K. J. Sternberg, C. P. Hwang, & A. G. Broberg (Eds.), Child care in con-text: Cross-cultural perspectives (pp. 441–459). Hills-dale, NJ: Erlbaum.

Harkness, S., & Super, C. M. (Eds.). (1996). Parents’ cultural belief systems: Their origins, expressions, and consequences. New York: Guilford Press.

Harlow, H. (1958). The nature of love. American Psy-chologist, 13, 673–685.

Hewlett, B. S., & Winn, S. (2014). Allomaternal nurs-ing in humans. Current Anthropology, 55, 200–229.

Hinde, R. A. (1982). Attachment: Some conceptual and biological issues. In C. Parkes, & J. Stevenson-Hinde (Eds.), The place of attachment in human behav-ior (pp. 60–76). New York: Basic Books.

Hinde, R. A., & Stevenson-Hinde, J. (1990). Attach-ment: Biological, cultural, and individual desiderata. Human Development, 33, 62–72.

Hofstede, G. (1984). Culture’s consequences: Interna-tional differences in work-related values. Newbury Park, CA: Sage.

Howes, C., Rodning, C., Galluzzo, D. C., & Myers, L. (1988). Attachment and child care: Relationships with mother and caregiver. Early Childhood Research Quarterly, 3, 403–416.

Hrdy, S. B. (2009). Mothers and others: The evolutionary origins of mutual understanding. Cambridge, MA: Har-vard University Press.

Hu, P., & Meng, Z. (1996). An examination of infant–mother attachment in China. Poster presented at the 14th meeting of the International Society for the Study of Behavioral Development, Quebec City, Canada.

Hu, P., & Meng, Z. (2003). Research on discrimination

of mother-infant attachment type. Acta Psychologica Sinica, 35, 201–208.

Huang, Z. J., Lewin, A., Mitchell, S. J., & Zhang, J. (2012). Variations in the relationship between ma-ternal depression, maternal sensitivity, and child attachment by race/ethnicity and nativity: Findings from a nationally representative cohort study. Mater-nal and Child Health Journal, 16, 40–50.

Jin, M. K., Jacobvitz, D., Hazen, N., & Jung, S. H. (2012). Maternal sensitivity and infant attachment security in Korea: Cross-cultural validation of the Strange Situation. Attachment and Human Develop-ment, 14, 33–44.

Kagitcibasi, C. (2007). Family, self, and human develop-ment across cultures: Theory and applications (2nd ed.). Mahwah, NJ: Erlbaum.

Kärtner, J., Keller, H., & Yovsi, R. D. (2010). Mother–infant interaction during the first 3 months: The emergence of culture-specific contingency patterns. Child Development, 81, 540–554.

Kazui, M., Endo, T., Tanaka, A., Sakagami, H., & Sug-anuma, M. (2000). Intergenerational transmission of attachment: Japanese mother–child dyads. Japanese Journal of Educational Psychology, 48, 323–332.

Keller, H., Borke, J., Staufenbiel, T., Yovsi, R. D., Abels, M., Papaligoura, Z., et al. (2009). Distal and proximal parenting as alternative parenting strategies during infants’ early months of life: A cross-cultural study. International Journal of Behavioral Development, 33, 412–420.

Keller, H., Borke, J., Yovsi, R., Lohaus, A., & Jensen, H. (2005). Cultural orientations and historical changes as predictors of parenting behaviour. International Journal of Behavioral Development, 29, 229–237.

Kermoian, R., & Leiderman, P. H. (1986). Infant at-tachment to mother and child caretaker in an East African community. International Journal of Behavioral Development, 9, 455–469.

Konner, M. (1977). Infancy among the Kalahari Desert San. In P. H. Leiderman, S. R. Tulkin, & A. Rosen-feld (Eds.), Culture and infancy: Variations in the human experience (pp. 287–328). New York: Academic Press.

Konner, M. (2005). Hunter–gatherer infancy and child-hood: The !Kung and others. In M. E. Lamb & B. S. Hewlett (Eds.), Hunter–gatherer childhoods: Evo-lutionary, developmental, and cultural perspectives (pp. 19–64). New Brunswick, NJ: Transaction.

Koren-Karie, N., Oppenheim, D., Dolev, S., Sher, E., & Etzion-Carasso, A. (2002). Mothers’ insightfulness regarding their infants’ internal experience: Relations with maternal sensitivity and infant attachment. De-velopmental Psychology, 38, 534–542.

Lee, R. B., & Daly, R. (1999). The Cambridge encyclo-pedia of hunters and gatherers. Cambridge, UK: Cam-bridge University Press.

LeVine, R. A., & Miller, P. M. (1990). Commentary. Human Development, 33, 73–80.

Li, X., Jing, J., Yang, D., Cai, X., Chen, X., & Su, X. (2004). Characters of 75 infants’ attachment towards



their mothers. Chinese Mental Health Journal, 18, 291–293.

Liang, X., Wang, Z.-Y., Liu, H.-Y., Lin, Q., Wang, Z., & Liu, Y. (2015). Adult attachment status predicts the developmental trajectory of maternal sensitivity in new motherhood among Chinese mothers. Mid-wifery, 31(1), 68–73.

Lucassen, N., Tharner, A., Van IJzendoorn, M. H., Bak-ermans-Kranenburg, M. J., Volling, B. L., Verhulst, F. C., et al. (2011). The association between paternal sensitivity and infant–father attachment security: A meta-analysis of three decades of research. Journal of Family Psychology, 25, 986–992.

Lupien, S. J., McEwen, B. S., Gunnar, M. R., & Heim, C. (2009). Effects of stress throughout the lifespan on the brain, behavior and cognition. Nature Reviews Neuroscience, 10, 434–445.

Main, M. (1990). Cross-cultural studies of attachment organization: Recent studies, changing method-ologies, and the concept of conditional strategies. Human Development, 33, 48–61.

Main, M., & Solomon, J. (1990). Procedures for iden-tifying infants as disorganized/disoriented during the Ainsworth Strange Situation. In M. T. Greenberg, D. Cicchetti, & E. M. Cummings (Eds.), Attachment in the preschool years: Theory, research, and intervention (pp. 121–160). Chicago: University of Chicago Press.

Marlowe, F. W. (2005). Who tends Hadza children? In M. E. Lamb & B. S. Hewlett (Eds.), Hunter–gatherer childhoods: Evolutionary, developmental, and cultural perspectives (pp. 19–64). New Brunswick, NJ: Trans-action.

Marvin, R. S., VanDevender, T. L., Iwanaga, M. I., LeVine, S., & LeVine, R. A. (1977). Infant–caregiver attachment among the Hausa of Nigeria. In H. Mc-Gurk (Ed.), Ecological factors in human development (pp. 247–259). Amsterdam: North-Holland.

McLoyd, V. C. (1998). Socioeconomic disadvantage and child development. American Psychologist, 53, 185–204.

Meaney, M. J. (2010). Epigenetics and the biological definition of gene × environment interactions. Child Development, 1, 41–79.

Meehan, C. L., & Hawks, S. (2013). Cooperative breed-ing and attachment among the Aka foragers. In N. Quinn & J. M. Mageo (Eds.), Attachment reconsidered: Cultural perspectives on a Western theory. New York: Palgrave Macmillan.

Meehan, C. L., & Hawks, S. (2015). Multiple attach-ments: Allomothering, stranger anxiety, and inti-macy. In H. Otto & H. Keller (Eds.), Different faces of attachment: Cultural variations on a universal human need (pp. 113–140). Cambridge, UK: Cambridge University Press.

Mesman, J., Van IJzendoorn, M., Behrens, K., Car-bonell, O. A., Cárcamo, R., Cohen-Paraira, I., et al. (2015). Is the ideal mother a sensitive mother? Beliefs about early childhood parenting in mothers across the globe. [Epub ahead of print]

Miller, G. E., Chen, E., & Parker, K. J. (2011). Psycho-logical stress in childhood and susceptibility to the chronic diseases of aging: Moving toward a model of behavioral and biological mechanisms. Psychological Bulletin, 137, 959–997.

Minde, K., Minde, R., & Vogel, W. (2006). Culturally sensitive assessment of attachment in children aged 18–40 months in a South African township. Infant Mental Health Journal, 27, 544–558.

Mizuta, I., Zahn-Waxler, C., Cole, P. M., & Hiruma, N. (1996). A cross cultural study of preschoolers’ attach-ment: Security and sensitivity in Japanese and U.S. dyads. International Journal of Behavioral Development, 19, 141–159.

Morelli, G. A., & Tronick, E. Z. (1991). Efé multiple caretaking and attachment. In J. L. Gewirtz & W. M. Kurtines (Eds.), Intersections with attachment (pp. 41–52). Hillsdale, NJ: Erlbaum.

Naftali, O. (2010). Recovering childhood: Play, peda-gogy, and the rise of psychological knowledge in contemporary urban China. Modern China, 36, 589–616.

Nakagawa, M., Lamb, M. E., & Miyake, K. (1992). An-tecedents and correlates of the Strange Situation be-havior of Japanese infants. Journal of Cross-Cultural Psychology, 23, 300–310.

Nakagawa, M., Teti, D. M., & Lamb, M. E. (1992). An ecological study of child–mother attachments among Japanese sojourners in the United States. Develop-mental Psychology, 28, 584–592.

Nauck, B., & Suckow, J. (2006). Intergenerational re-lationships in cross-cultural comparison: How social networks frame intergenerational relations between mothers and grandmothers in Japan, Korea, China, Indonesia, Israel, Germany, and Turkey. Journal of Family Issues, 27, 1159–1185.

Nelson, C. A., Fox, N. A., & Zeanah, C. H. (2014). Ro-mania’s abandoned children: Deprivation, brain develop-ment, and the struggle for recovery. Cambridge, MA: Harvard University Press.

Oppenheim, D., Koren-Karie, N., & Sagi-Schwartz, A. (2007). Emotional dialogues between mothers and children at 4.5 and 7.5 years: Relations with chil-dren’s attachment at 1 year. Child Development, 78, 38–52.

Oppenheim, D., Sagi, A., & Lamb, M. E. (1990). In-fant–adult attachments on the kibbutz and their rela-tion to socioemotional development four years later. In S. Chess & M. E. Hertzig (Eds.), Annual progress in child psychiatry and child development (pp. 92–106). New York: Brunner/Mazel.

Otto, H. (2014). Don’t show your emotions!: Emotion regulation and attachment in the Cameroonian Nso. In H. Otto & H. Keller (Eds.), Different faces of at-tachment: Cultural variations on a universal human need (pp. 215–229). Cambridge, UK: Cambridge Univer-sity Press.

Oyserman, D., Coon, H. M., & Kemmelmeier, M. (2002). Rethinking individualism and collectivism:


Laura

Sticky Note

AU: Provide page nos. PAGE NOS ARE 85-113


Evaluation of theoretical assumptions and meta-anal-yses. Psychological Bulletin, 128, 3–72.

Pederson, D. R., & Moran, G. (1995). A categorical de-scription of infant–mother relationships in the home and its relation to Q-sort measures of infant–mother interaction. Monographs of the Society for Research in Child Development, 60 (2-3), 111–132.

Peterson, N. J., Drotar, D., Olness, K., Guay, L., & Kiziri-Mayengo, R. (2001). The relationship of maternal and child HIV infection to security of attachment among Ugandan infants. Child Psychiatry and Human Development, 32, 3–17.

Posada, G., Carbonell, O. A., Alzate, G., & Plata, S. J. (2004). Through Colombian lenses: Ethnographic and conventional analyses of maternal care and their associations with secure base behavior. Developmental Psychology, 40, 508–518.

Posada, G., Gao, Y., Wu, F., Posado, R., Tascon, M., Schoelmerich, A., et al. (1995). The secure-base phenomenon across cultures: Children’s behavior, mothers’ preferences, and experts’ concepts. In E. Waters, B. E. Vaughn, G. Posada, & K. Kondo-Ikemura (Eds.), Caregiving, cultural, and cognitive perspectives on secure-base behavior and working models: New growing points of attachment theory and research. Monographs of the Society for Research in Child Development, 60 (2–3, Serial No. 244), 27–48.

Posada, G., Jacobs, A., Richmond, M., Carbonell,O. A., Alzate, G., Bustamante, M. R., et al. (2002). Mater-nal caregiving and infant security in two cultures. De-velopmental Psychology, 38, 67–78.

Posada, G., Lu, T., Trumbell, J., Kaloustian, G., Tru-del, M., Plata, S. J., et al. (2013). Is the secure-base phenomenon evident here, there, and anywhere?: A cross-cultural study of child behavior and experts’ definitions. Child Development, 84, 1896–1905.

Rabin, A. I., & Beit-Hallahmi, B. (1982). Twenty years later. New York: Springer.

Reed, G., & Leiderman, P. H. (1981). Age-related changes in attachment behavior in polymatrically reared infants: The Kenyan Gusii. In T. M. Field, A. M. Sostek, P. Vietze, & P. H. Leiderman (Eds.), Culture and early interactions (pp. 215–236). Hillsdale, NJ: Erlbaum.

Rothbaum, F., Weisz, J., Pott, M., Miyake, K., & Morelli, G. (2000). Attachment and culture: Security in the United States and Japan. American Psychologist, 55, 1093–1104.

Sagi, A., Koren-Karie, N., Gini, M., Ziv, Y., & Joels, T. (2002). Shedding further light on the effects of vari-ous types and quality of early child care on infant–mother attachment relationship: The Haifa study of early child care. Child Development, 73, 1166–1186.

Sagi, A., Lamb, M. E., Lewkowicz, K. S., Shoham, R., Dvir, R., & Estes, D. (1985). Security of infant–mother, –father, and –metapelet attachments among kibbutz-reared Israeli children. In I. Bretherton & E. Waters (Eds.), Growing points of attachment theory

and research. Monographs of the Society for Research in Child Development, 50 (1–2, Serial No. 209), 257–275.

Sagi, A., & Van IJzendoorn, M. H. (1996). Multiple caregiving environments: The kibbutz experience. In S. Harel & J. P. Shonkoff (Eds.), Early childhood inter-vention and family support programs: Accomplishments and challenges (pp. 143–162). Jerusalem: JDC–Broo-kale Institute.

Sagi, A., Van IJzendoorn, M. H., Aviezer, O., Donnell, F., Koren-Karie, N., Joels, T, et al. (1995). Attach-ments in a multiple-caregiver and multiple-infant environment: The case of the Israeli kibbutzim. In E. Waters, B. E. Vaughn, G. Posada, & K. Kondo-Ikemura (Eds.), Caregiving, cultural, and cognitive per-spectives on secure-base behavior and working models: New growing points of attachment theory and research (Special issue). Monographs of the Society for Research on Child Development, 60(Serial No. 244, No. 2–3), 71–91.

Sagi, A., Van IJzendoorn, M. H., Aviezer, O., Donnell, F., & Mayseless, O. (1994). Sleeping out of home in a kibbutz communal arrangement: It makes a difference for infant–mother attachment. Child Development, 65, 992–1004.

Sagi, A., Van IJzendoorn, M. H., Scharf, M., Joels, T., Koren-Karie, N., Mayseless, O., et al. (1997). Eco-logical constraints for intergenerational transmission of attachment. International Journal of Behavioral De-velopment, 20, 287–299.

Sagi-Schwartz, A., & Aviezer, O. (2005). Correlates of attachment to multiple caregivers in kibbutz children from birth to emerging adulthood: The Haifa longi-tudinal study. In K. E. Grossmann, K. Grossmann, & E. Waters (Eds.), Attachment from infancy to adulthood (pp. 165–197). New York: Guilford Press.

Sagi-Schwartz, A., Van IJzendoorn, M. H., Bakermans-Kranenburg, M. J. (2008). Does intergenerational transmission of trauma skip a generation?: No meta-analytic evidence for tertiary traumatization with third generation of Holocaust survivors. Attachment and Human Development, 10, 105–121.

Scheper-Hughes, N. (1993). Death without weeping: The violence of everyday life in Brazil. Berkeley: University of California Press.

Short, S. E., Fengying, Z., Siyuan, X., & Mingliang, Y. (2001). China’s one-child policy and the care of children: An analysis of qualitative and quantitative data. Social Forces, 79, 913–943.

Sroufe, L. A., Egeland, B., Carlson, E., & Collins, W. A. (2005). The development of the person: The Minnesota Study of Risk and Adaptation from Birth to Adulthood. New York: Guilford Press.

Sroufe, L. A., & Waters, E. (1977). Attachment as an organizational construct. Child Development, 48, 1184–1199.

Strassman, B. I. (2011). Cooperation and competition in a cliff-dwelling people. PNAS, 108, 10894-10901.

Takahashi, K. (1986). Examining the Strange Situation



procedure with Japanese mothers and 12-month-old infants. Developmental Psychology, 22, 265–270.

Takahashi, K. (1990). Are the key assumptions of the “Strange Situation” procedure universal?: A view from Japanese research. Human Development, 33, 23–30.

Tavecchio, L. W. C., & Van IJzendoorn, M. H. (1987). Attachment in social networks: Contributions to the Bowlby–Ainsworth attachment theory. Amsterdam: North-Holland.

Tomlinson, M., Cooper, P., & Murray, L. (2005). The mother–infant relationship and infant attachment in a South-African peri-urban settlement. Child Devel-opment, 76, 1044–1054.

Trivers, R. L. (1974). Parent–offspring conflict. Ameri-can Zoologist, 14, 249–264.

Tronick, E. Z., Morelli, G. A., & Winn, S. (1987). Mul-tiple caretaking of Efé (Pygmy) infants. American An-thropologist, 89, 96–106.

True, M. M. (1994). Mother–infant attachment and com-munication among the Dogon of Mali. Unpublished doctoral dissertation, University of California, Berkeley.

True, M. M., Pisani, L., & Oumar, F. (2001). Infant–mother attachment among the Dogon of Mali. Child Development, 72, 1451–1466.

Valenzuela, M. (1997). Maternal sensitivity in a de-veloping society: The context of urban poverty and infant chronic undernutrition. Developmental Psychol-ogy, 845–855.

Van Bakel, H. J. A., & Riksen-Walraven, J. M. A. (2002). Parenting and development of one-year-olds: Links with parental, contextual, and child character-istics. Child Development, 73, 256–273.

Van IJzendoorn, M. H. (1990). Developments in cross-cultural research on attachment: Some methodologi-cal notes. Human Development, 33, 3–9.

Van IJzendoorn, M. H. (1995). Adult attachment rep-resentations, parental responsiveness, and infant at-tachment: A meta-analysis on the predictive validity of the Adult Attachment Interview. Psychological Bul-letin, 117, 387–403.

Van IJzendoorn, M. H., Bakermans-Kranenburg, M. J., & Sagi-Schwartz, A. (2006). Attachment across diverse sociocultural contexts. The limits of univer-sality. In K. Rubin & O. B. Chung (Eds.), Parenting beliefs, behaviors, and parent–child relations: A cross-cultural perspective (pp. 107–142). New York: Psychol-ogy Press.

Van IJzendoorn, M. H., Bard, K. A., Bakermans-Kranenburg, M. J., & Ivan, K. (2009). Cognitive development of young nursery-reared chimpanzees in responsive versus standard care. Developmental Psy-chobiology, 51, 173–185.

Van IJzendoorn, M. H., Dijkstra, J., & Bus, A. G. (1995). Attachment, intelligence, and language. Social Devel-opment, 4, 115–128.

Van IJzendoorn, M. H., & Kroonenberg, P. M. (1988). Cross-cultural patterns of attachment: A meta-anal-

ysis of the Strange Situation. Child Development, 59, 147–156.

Van IJzendoorn, M. H., & Sagi, A. (2001). Cultural blindness or selective inattention? American Psycholo-gist, 56, 824–825.

Van IJzendoorn, M. H., Sagi, A., & Lambermon, M. W. E. (1992). The multiple caretaker paradox: Data from Holland and Israel. New Directions for Child Develop-ment, 57, 5–24.

Van IJzendoorn, M. H., & Sagi-Schwartz, A. (2008). Cross-cultural patterns of attachment: Universal and contextual dimensions. In J. Cassidy & P. R. Shaver (Eds.), Handbook of attachment: Theory, research, and clinical applications (2nd ed., pp. 880–905). New York: Guilford Press.

Van IJzendoorn, M. H., Vereijken, C. M. J. L., Baker-mans-Kranenburg, M. J., & Riksen-Walraven, J. M. (2004). Assessing attachment security with the At-tachment Q-Sort: Meta-analytic evidence for the validity of the observer AQS. Child Development, 75, 1188–1213.

Vaughn, B. E., Coppola, G., Verissimo, M., Monteiro, L., José Santos, A., Posada, G., et al. (2007). The qual-ity of maternal secure-base scripts predicts children’s secure-base behavior at home in three sociocultural groups. International Journal of Behavioral Develop-ment, 31, 65–76.

Vaughn, B. E., & Waters, E. (1990). Attachment behav-ior at home and in the laboratory: Q-sort observations and Strange Situation classifications of one-year-olds. Child Development, 61, 1965–1973.

Vereijken, C. M. J. L. (1996). The mother–infant relation-ship in Japan: Attachment, dependency, and amae. Un-published doctoral dissertation, Catholic University of Nijmegen, The Netherlands.

Vereijken, C. M. J. L., Riksen-Walraven, M., & Kondo-Ikemura, K. (1997). Maternal sensitivity and infant attachment security in Japan: A longitudinal study. International Journal of Behavioral Development, 21, 35–49.

Wang, L., & Mesman, J. (2015). Parenting and child devel-opment in China: A narrative and meta-analytic review of the cross-cultural literature. Manuscript submitted for publication.

Weinfield, N. S., Sroufe, L. A., Egeland, B., & Carlson, E. (2008). Individual differences in infant-caregiver attachment: Conceptual and empirical aspects of se-curity. In J. Cassidy & P. R. Shaver (Eds.), Handbook of attachment: Theory, research, and clinical applications (2nd ed., pp. 78–101). New York: Guilford Press.

WHO/UNICEF (2003). Global strategy for infant and young child feeding. Geneva: WHO.

Yue, Y., Zhang, G., Chen, H., Liang, Z., & Zhang, P. (2010). The relations between children’s attachment and problem behaviour. Psychological Science, 33, 318–320.

Zevalkink, J. (1997). Attachment in Indonesia: The moth-er–child relationship in context. Ridderkerk, The Neth-erlands: Ridderprint.



Zevalkink, J., Riksen-Walraven, J. M., & Van Lieshout, C. F. M. (1999). Attachment in the Indonesian care-giving context. Social Development, 8, 21–40.

Zevalkink, J., Riksen-Walraven, M., & Bradley, R. H. (2008). The quality of children’s home environment and attachment security in Indonesia. Journal of Ge-netic Psychology: Research and Theory on Human De-velopment, 169, 72–91.

Zreik, G (2014). Attachment in a cross-cultural perspec-

tive: The case of Arab infants and mothers in Israel. Unpublished doctoral thesis, University of Haifa, De-partment of Psychology, Faculty of Social Sciences, Ramat-Gan, Israel.

Zreik, G., Oppenheim, D., Sagi-Schwartz, A. (2015, March). Attachment in a cross-cultural perspective: The case of Arab infants and mothers in Israel. Poster presented at the biennial meeting of the Society for Research in Child Development Philadelphia .


Documents

Cross-Cultural Patterns of Attachment