Reading G

Developmental theorists agree that social relationships are critical in human development, but they have disagreed about which relationships are most critical. Many noted theorists have argued that no social relationship is more important than the first: the one between parent and child. Sigmund Freud (1930) left no doubt: A warm and stable mother–child relationship is essential for normal personality development. His follower Erik Erikson tended to agree, emphasizing the importance of responsive parenting to the development of trust in the parent–infant relationship. These theorists, in turn, influenced the architects of today’s most influential theory of close human relationships, attachment theory, to focus on the lasting significance of early parent–child relationships.

Attachment Theory

Attachment theory was formulated by British psychiatrist John Bowlby (1969, 1973, 1980, 1988), and it was elaborated on by his colleague Mary Ainsworth, an American developmental psychologist (1989; Ainsworth et al., 1978; and see Cassidy & Shaver, 2016). The theory was based primarily on ethology, the study of the behavior of various species in their natural environments and the evolution of that behavior (Archer, 1992). Attachment theory therefore asked how attachment might have helped our ancestors adapt to their environment. Bowlby also drew on concepts from psychoanalytic theory (he was a psychoanalytic therapist who studied the contribution of mother–child relationships to later psychopathology) and cognitive theory (he was interested in expectations about self and other, as you will see).

According to Bowlby (1969), an attachment is a strong affectional tie that binds a person to an intimate companion. It is also a behavioral system through which humans regulate their emotional distress when under threat and achieve security by seeking proximity to another person. For most of us, the first attachment we form, around 6 or 7 months of age, is to a parent.

How do we know when a baby becomes attached to his mother? He will try to maintain proximity to her—crying, clinging, approaching, following, doing whatever it takes to maintain the desired closeness to her and expressing his displeasure when he cannot. He will prefer her to other people, reserving his biggest smiles for her and seeking her when he is upset, discomforted, or afraid; she is irreplaceable in his eyes. He will also be confident about exploring his environment as long as he knows that his mother is there to provide the security he needs.

Notice that an infant attached to a parent is rather like an adult “in love” (like Mike or Marcy at the start of the chapter). True, close emotional ties are expressed in different ways, and serve different functions, at different points in the life span. Adults, for example, do not usually feel compelled to follow their mates around the house, and they look to their loved ones for more than comforting hugs and smiles. Nonetheless, there are basic similarities among the infant attached to a caregiver, the child attached to a best friend, and the adolescent or adult attached to a romantic partner. Throughout the life span, the objects of our attachments are special, irreplaceable people to whom we want to be close and from whom we derive a sense of security (Ainsworth, 1989).

Nature, Nurture, and Attachment

Drawing on ethological theory and research, Bowlby argued that both infants and parents are biologically predisposed to form attachments because they contribute to survival. It makes sense to think, for example, that young birds tended to survive if they stayed close to their mothers so that they could be fed and protected from predators—but that they starved or were gobbled up, and therefore failed to pass their genes to future generations, if they strayed. Thus, chicks, ducks, and goslings may have evolved to engage in imprinting, an innate form of learning in which the young will follow and become attached to a moving object (usually the mother) during a critical period early in life.

Noted ethologist Konrad Lorenz (1937) observed imprinting in young goslings and found that it is automatic and unlearned, that it occurs only within a critical period shortly after the bird has hatched, and that it is irreversible—once the gosling begins to follow a particular object, whether its mother or Lorenz, it will remain attached to that object. The imprinting response is considered a prime example of a species-specific and largely innate behavior that has evolved because it has survival value. Subsequent research has shown that it is not quite as automatic and different from other learning as originally claimed, that the “critical” period is more like a “sensitive” period, and that imprinting can be reversed (Spencer et al., 2009).

What about human infants? Babies may not become imprinted to their mothers, but they certainly form attachments in infancy and follow their love objects around. Bowlby argued that they come equipped with several other behaviors besides proximity-seeking that help ensure adults will love them, stay with them, and meet their needs. Among these behaviors are sucking and clinging, smiling and vocalizing (crying, cooing, and babbling), and expressions of negative emotion (fretting and crying).

Moreover, Bowlby argued, just as infants are programmed to respond to their caregivers, adults are biologically programmed to respond to an infant’s signals. Just see if you can ignore a baby’s cry or fail to smile when a baby grins at you. Adults are even hormonally prepared for caregiving. Oxytocin is a hormone produced primarily in the hypothalamus that plays roles in facilitating caregiving and parent–infant attachment in animals and humans (Feldman, 2016). It affects the muscles that facilitate contractions during labor and the release of milk during breastfeeding. Moreover, mothers with high levels of oxytocin before birth engage in higher levels of positive attachment behavior after the birth, think more about their relationships with their infants, and check more often on their infants than mothers with lower levels of the hormone (Feldman et al., 2007). Oxytocin seems to prime parents (fathers too) to form attachments to their babies, then. Administered by nasal spray, it also seems to increase interpersonal trust among adult partners, although only for some people in some contexts (Bartz, 2016; Schneiderman, Zagoory-Sharon, Leckman, & Feldman, 2012).

Just as the imprinting of goslings occurs during a critical period, human attachments form during what Bowlby viewed as a sensitive period for attachment, the first 3 years of life. But attachments do not form automatically. According to Bowlby, a responsive social environment is critical: An infant’s preprogrammed signals to other people may eventually wane if caregivers are unresponsive to them. Ultimately, the security of an attachment relationship depends on both nature and nurture—on the interaction over time between a biologically prepared infant and a biologically prepared caregiver and on the sensitivity of each partner to the other’s signals.

Attachment and Later Development

Bowlby maintained that the quality of the early parent–infant attachment has lasting impacts on development, especially on later relationships. He proposed that, based on their interactions with caregivers, infants construct expectations about relationships called internal working models—cognitive representations of themselves and other people that guide their processing of social information and their behavior in relationships (Bowlby, 1973; see also Bretherton, 1996). Securely attached infants who have received responsive care will form internal working models suggesting that they are lovable and that other people can be trusted to care for them. By contrast, insecurely attached infants subjected to insensitive, neglectful, or abusive care may conclude that they are difficult to love, that other people are unreliable, or both. Because of their internal working models, insecure infants would be expected to avoid or to be anxious in close relationships later in life. They may, for example, be wary of getting too close to anyone or become jealous and overly dependent if they do.

In sum, attachment theory, as developed by John Bowlby and elaborated with Mary Ainsworth, claims the following:

• The capacity to form attachments is part of our evolutionary heritage.
• Attachments unfold through an interaction of biological and environmental forces during a sensitive period early in life.
• The quality of the attachment between infant and caregiver shapes later development and the quality of later relationships.
• Internal working models of self and others are the mechanism through which early experience affects later development.

Ainsworth’s most important contribution to attachment theory was to devise a way to assess differences in the quality of parent–infant attachments. She and her associates created the Strange Situation, a now-famous procedure for measuring the quality of an attachment (Ainsworth et al., 1978). It consists of eight episodes that gradually escalate the amount of stress infants experience as they react to the approach of an adult stranger and two departures and returns of their caregiver. On the basis of an infant’s pattern of behavior across the episodes, the quality of his attachment to a parent can be characterized as one of four types, one secure and three insecure (resistant, avoidant, and disorganized disoriented).

1. Secure Attachment

About 60% of 1-year-olds in the many societies that have been studied are securely attached to their mothers or primary caregivers (Colin, 1996; Howe, 2011). The securely attached infant actively explores the room when alone with his mother because she serves as a secure base. The infant is upset by separation but greets his mother warmly and is quickly comforted by her presence when she returns. The securely attached child is outgoing with a stranger when his mother is present. In the Bowlby–Ainsworth view, the securely attached infant “stays close and continuously monitors [the caregiver’s] whereabouts (proximity maintenance), retreats to her for comfort if needed (safe haven), resists and is distressed by separations from her (separation distress), and explores happily as long as she is present and attentive (secure base)” (Hazan, Campa, & Gur-Yaish, 2006, p. 190).

2. Resistant Attachment

About 10% of 1-year-olds show a resistant attachment, an insecure attachment characterized by anxious, ambivalent reactions (and also called anxious/ ambivalent attachment). The resistant infant does not dare venture off to play even when her mother is present; her mother does not seem to serve as a secure base for exploration. Yet this infant becomes distressed when her mother departs, often showing stronger separation anxiety than the securely attached infant. When her mother returns, the infant is ambivalent: she may try to remain near her but is not comforted and does not calm down; she seems to resent the mother for having left, may resist if she tries to make physical contact, and may even hit and kick her in anger (Ainsworth et al., 1978). Resistant infants are also wary of strangers, even when their mothers are present. It seems, then, that resistant or ambivalent infants do all they can to get affection and comfort but never quite succeed.

3. Avoidant Attachment

Infants with avoidant attachments (up to 15% of 1-year-olds) may play alone but are not very adventuresome, show little apparent distress when separated from their mothers, and avoid contact or seem indifferent when their mothers return. These insecurely attached infants are not particularly wary of strangers but sometimes avoid or ignore them, much as these babies avoid or ignore their mothers. In contrast to resistant infants, avoidant infants seem to have shut down their emotions and distanced themselves from their parents, as if they were denying their need for affection or had learned not to express their emotional needs.

4. Disorganized–Disoriented Attachment

Ainsworth’s work initially focused on secure, resistant, and avoidant attachment styles, but some infants do not develop any of these consistent attachment styles. They seem confused. Up to 15% of infants—more in high-risk families—display what is now recognized as a fourth attachment classification, one that is associated with later emotional problems (Main & Solomon, 1990; van IJzendoorn, Schuengel, & Bakermans-Kranenburg, 1999).

Reunited with their mothers after a separation, these infants may act dazed and freeze or lie on the floor immobilized—or they may seek contact but then abruptly move away as their mothers approach them, only to seek contact again. Infants with a disorganized–disoriented attachment have not been able to devise a consistent strategy for regulating their negative emotions; they seem frightened of their parent and stuck between approaching and avoiding this frightening figure (Hesse & Main, 2006). She is clearly not a source of comfort for them.

Table 14.1 Child Behaviors in the Strange Situation Associated with Attachment Types and Related Parenting Styles

Table 14.1 summarizes the features of these four patterns of attachment. What determines which of these attachment patterns will characterize a parent–infant relationship? The caregiver, the infant, and the cultural and social context all contribute.

The Caregiver’s Contributions

According to Freud, infants in the oral stage of psychosexual development become attached to the individual who provides them with oral pleasure, and the attachment bond will be most secure if a mother is relaxed and generous in her feeding practices.

In a classic study conducted by Harry Harlow and Robert Zimmerman (1959), Freud’s psychoanalytic view was put to the test—and failed. Monkeys were reared with two surrogate mothers: a wire “mother” and a cloth “mother” wrapped in foam rubber and covered with terrycloth. Half the infants were fed by the cloth mother, and the remaining infants were fed by the wire mother. To which mother did these infants become attached? There was no contest: Infants strongly preferred the cuddly cloth mother, regardless of which mother had fed them. They spent most of their time clinging to her, ran to her when they were upset or afraid, and showed every sign of being attached to her.

Harlow’s research demonstrated that contact comfort , the pleasurable tactile sensations provided by a soft and cuddly “parent,” is a more powerful contributor to attachment in monkeys than feeding. Contact comfort also promotes human attachments (Anisfeld et al., 1990). Moreover, many infants become attached to someone other than the adult who feeds them (Schaffer & Emerson, 1964).

In the end, it is the parent’s behavior that most influences the quality of attachment, as shown in Table 14.1. Infants who enjoy secure attachments to their parents have parents who are sensitive and responsive to their needs and emotional signals, just as Bowlby and Ainsworth proposed (Ainsworth et al., 1978; De Wolff & van IJzendoorn, 1997). The importance of sensitive, responsive parenting in creating a secure attachment is one of the most established findings in developmental science.

Babies who show a resistant pattern of attachment often have parents who are inconsistent in their caregiving; they react enthusiastically or indifferently, depending on their moods, and are frequently unresponsive (Isabella, 1993; Isabella & Belsky, 1991). Mothers who are depressed, for example, often have difficulty responding sensitively to their babies’ signals and do not provide the comforting that helps babies regulate their negative emotions (Dawson & Ashman, 2000). The resistant infant may learn to cope with such unreliable caregiving by trying desperately—through clinging, crying, and other attachment behaviors—to obtain emotional support and comfort, and then becomes sad and resentful when these efforts fail.

The parents of infants with an avoidant attachment sometimes deserve to be avoided. Some are rejecting; they are impatient, unresponsive, and resentful when the infant interferes with their plans (Ainsworth, 1979; Isabella, 1993). Others have been called “intrusive”; often meaning well, they provide high levels of stimulation even when their babies become uncomfortably aroused and need a break so that they can regain control of their emotions (Isabella & Belsky, 1991). Infants with an avoidant attachment style may learn to avoid and make few emotional demands on adults who either seem to dislike their company or bombard them with stimulation they cannot handle.

Finally, a disorganized–disoriented style of attachment is evident in as many as 80% of infants who have been physically abused or maltreated (Carlson et al., 1989; and see Baer & Martinez, 2006) and is also common among infants whose mothers are severely depressed or abuse alcohol and drugs (Beckwith, Rozga, & Sigman, 2002). The parents of infants with a disorganized attachment pattern have been described as frightening and frightened—as fearful adults who are not up to the challenge of caring for an infant and create an unpredictable, scary environment for their babies (Hesse & Main, 2006). Infants with a disorganized attachment are understandably confused about whether to approach or avoid a parent who can be loving one minute but angry and abusive or indifferent the next. You can see each of the four types of attachment, then, as a quite reasonable way of coping with a particular brand of parenting.

The Infant’s Temperament

An infant’s temperament can also influence the quality of an attachment. Attachments tend to be insecure when infants are by temperament fearful, irritable, or unresponsive (Beckwith et al., 2002). The caregiver’s style of parenting and the infant’s temperament can also interact to determine the outcome. To illustrate, Figure 14.3 shows the percentages of 12-month-olds who tested as securely attached as a function of whether they were difficult-to-read infants born prematurely and whether their mothers were depressed (Poehlmann & Fiese, 2001). When a difficult-to-read, premature infant was paired with a depressed mother, the odds of a secure attachment were especially low (see Leerkes, Blankson, & O’Brien, 2009).

Figure 14.3 The combination of a depressed mother and a pre-mature infant means low odds that a secure attachment will form.

^{Source: Data from Poehlmann, J. & Fiese, B. H. The interaction of maternal and infant vulnerabilities on developing attachment relationships, Development and Psychopathology, 13, 1-11. (c) 2001 Guildfod. Reprinted by permission from Cambridge University Press.}

So who affects the quality of parent–infant attachment more, the parent or the infant? The parent is clearly more influential. We know the infant’s temperament is not the main influence on security of attachment because:

• Relationships between infant temperament and quality of attachment are often quite weak (Roisman & Groh, 2011).
• Many infants are securely attached to one parent but insecurely attached to the other (Kochanska & Kim, 2013; van IJzendoorn & De Wolff, 1997).
• An infant’s genetic makeup, although it influences the infant’s temperament, has little influence on quality of attachment (Fearon et al., 2006; Raby et al., 2012; Roisman & Fraley, 2008).
• Caregivers who are responsive and adjust to their baby’s temperamental quirks are able to establish secure relationships even with temperamentally difficult babies (Leerkes et al., 2009; Mangelsdorf et al., 1990).

Cultural and Social Context

Finally, the broader social context surrounding caregiver and infant can affect how they react to each other. For example, the stresses associated with living in poverty or experiencing marital conflict may make it difficult for parents to provide sensitive care and may contribute to insecure attachments (Roisman & Booth- LaForce, 2014).

The cultural context in which caregiver and baby interact also colors their relationship. For instance, in Germany, an individualistic culture, parents strongly encourage independence and discourage clingy behavior, fearing that if they are too responsive to cries they will spoil their infants. This may explain why many German infants make few emotional demands on their parents and are often classified as avoidantly attached (Grossmann, Grossmann, & Keppler, 2005). By contrast, babies in the collectivist culture of Japan are rarely separated from their mothers early in life and are encouraged to be dependent on their mothers. They become highly distressed by separations such as those they must endure in the Strange Situation. As a result, they are more likely than American babies to be classified as resistantly attached (Takahashi, 1990; van IJzendoorn & Sagi, 1999).

Could findings like this mean that research on infant attachment is culturally biased? Might those Japanese babies who freak out in the Strange Situation actually be securely attached? Very possibly. In addition, we must appreciate that in many cultures infants do not have a single primary caregiver but instead are passed from caregiver to caregiver during the day. Among the Efe of the Democratic Republic of Congo, for example, infants were observed to be in contact with an average of 9 to 14 different people during a two-hour observation period (Morelli, 2015).

A number of researchers conclude that secure attachment may mean somewhat different things in different societies (Miller, Goyal, & Wice, 2015; Rothbaum & Morelli, 2005). At the same time, many of the key signs of a secure attachment are consistent across cultures (Posada et al., 2013). And the main predictions of attachment theory—especially the critical importance of parent sensitivity and responsiveness in fostering a secure attachment— are well supported across cultures (see Morelli, 2015; van IJzendoorn & Sagi-Schwartz, 2008). Overall, then, sensitivity and responsiveness of the caregiver (or in many cultures, the caregivers) is the primary influence on the quality of caregiver–infant attachment, but characteristics of both the baby’s temperament and the surrounding cultural and social environment also affect it.

From Freud on, psychologists have assumed that a secure parent– child relationship is critical in human development. Just how important is it? Several lines of research offer answers: studies of socially deprived infants; studies of infants who experience separations from their caregivers; and studies comparing the later development of securely and insecurely attached infants.

Social Deprivation

What happens to infants who never have an opportunity to form an attachment? It is better to have loved and lost than never to have loved at all, say studies of infants who grow up in deprived institutional settings where they are unable to form attachments (Leiden Conference, 2012; MacLean, 2003). In the 1990s, children from deprived institutions in Romania were adopted into homes in the United States, the United Kingdom, and Canada after the fall of the Romanian government (Gunnar, Bruce, & Grotevant, 2000). These adoptees reportedly spent their infancies in orphanages with 20–30 children in a room and only one caregiver for every 10–20 children; they spent most of their time rocking in their cribs with little human contact, much less hugs and synchronous routines (Fisher et al., 1997). How did they turn out?

Infants who spent their first several months or more in deprived orphanages displayed a host of problems—poor growth, medical problems, brain abnormalities, and delays in physical, cognitive, and social-emotional development (Leiden Conference, 2012; MacLean, 2003; McCall & Groark, 2015). What of their attachments after they were adopted? The good news is that about 90% had clearly become attached to their adoptive parents 9 months after adoption (Carlson et al., 2014). However, high proportions showed disorganized attachment behavior, as though they saw their parents as both comforting and threatening (see also Bakermans- Kranenburg et al., 2011; Zeanah, Berlin, & Boris, 2011). Many institutionalized children continue to show insecure and atypical attachment behavior as adolescents (Palacios et al., 2014).

A meta-analysis of many studies of institutionalized and otherwise maltreated and neglected children concluded that those who are adopted before 1 year of age have a good chance of becoming as securely attached to their caregivers as other children (van den Dries et al., 2009). However, high rates of insecure and disturbed attachment are observed among children adopted after their first birthday, and the problem is not lack of a single “mother figure.” In adequately staffed institutions and communes, infants cared for by a few responsive caregivers turn out fine (Groark et al., 2005; St. Petersburg-USA Orphanage Research Team, 2008).

Apparently, then, normal development requires sustained interactions with sensitive and responsive caregivers—whether one or a few. Apparently too, children are resilient if they are given a reasonable chance to find someone to love. However, as Bowlby claimed, early social deprivation can leave lasting marks on development.

Separations

Now consider babies who form an attachment but are separated from their caregivers as a result of illness, war, death, divorce, or other circumstances. Bowlby (1960, 1980) discovered that these infants go through a grieving process in which they first protest and search frantically for their loved one, then become sad and listless once they give up, and sometimes then ignore or avoid their caregiver if she returns, only gradually warming up to her again.

Long-term separation of young children from a parent is an issue for military families when a family member is deployed. This can be a Strange Situation-like stress test not only for infants and children but for their parents, whose attachments to their partners are disrupted. Although most families cope well, some infants and children develop emotional and behavioral problems (Maholmes, 2012; Riggs & Riggs, 2011). The most important influence on whether children adapt well is whether the at-home parent or substitute caregiver provides sensitive and responsive care (Posada, Longoria, Cocker, & Lu, 2011). Programs specially designed for military families can help reduce the negative effects of deployment on infants and children (Lester et al., 2012).

Similarly, infants who are permanently separated from an attachment figure caregiver due to death, divorce, or another reason often have adjustment difficulties but normally recover if they have the security of an existing attachment or can form a new one (Bowlby, 1960, 1980; van IJzendoorn & Juffer, 2006). We should worry more about children who experience a series of permanent separations from caregivers, as happens to some infants and children who are repeatedly moved from one foster care placement to another. They may have long-lasting attachment problems and psychological disorders as a result of their repeated experiences of loving and losing (Tarren-Sweeney, 2013; Ward, Munro, & Dearden, 2006).

Day Care

What of the daily separations from their parents that infants experience when they are placed in day care centers or in family care homes (in which caregivers take children into their homes)? A major national longitudinal study of child care supported by the National Institute of Child Health and Human Development tracked infants through adolescence (NICHD, 2006; NICHD Early Child Care Research Network [ECCRN], 1997, 2006; Vandell et al., 2010). Overall, infants who experienced routine care by someone other than their mothers were not much different than infants cared for almost exclusively by their mothers in the many aspects of development studied. Most importantly, infants who received alternative care were no less securely attached to their mothers overall than infants who were tended by their mothers (Friedman & Boyle, 2008; Phillips, 2015; NICHD ECCRN, 1997).

Quality of parenting was a much stronger influence on these infants’ attachments and later development than whether they had day care experience. Quality of day care also had impacts, though; infants fared poorest if their mothers were insensitive and unresponsive and they were subjected to poor-quality day care on top of it (NICHD ECCRN, 1997, 2006; Vandell et al., 2010). In sum, infants and young children who spend time in day care are not less securely attached to their parents than infants and young children cared for at home, on average. However, they are most likely to thrive if they receive both good quality parenting and good quality child care. From a policy perspective, it makes sense to make affordable and high-quality child care available to more families (Phillips, 2015).

Later Development of Securely and Insecurely Attached Infants

Finally, we can assess the importance of early attachment experiences by asking the following question: How much difference does having a secure rather than insecure attachment in infancy make later in life? Three main qualities distinguish children who were securely attached infants from those who were insecurely attached:

• Intellectual competence. Children who were securely attached as infants are described by teachers as more curious, selfdirected, and eager to learn than insecurely attached children (Waters, Wippman, & Sroufe, 1979) and they are more engaged in classroom activities (Drake, Belsky, & Pasco Fearon, 2014). They have benefitted from having a secure base for exploration.
• Social competence. Children who had been securely attached as infants are more able to initiate play activities, are more sensitive to the needs and feelings of other children, and are more popular and socially competent (Booth-LaForce & Kerns, 2009; Waters et al., 1979). Their positive interactions with caregivers have allowed them to develop good social skills and positive internal working models of self and others.
• Emotion regulation. Secure attachment in infancy is also linked to good emotion regulation and coping skills (Gunnar, 1998; Kochanska, 2001). The comforting and help with emotion regulation securely attached babies receive from their sensitive parents make them less reactive to stress than children who had insecure attachments (Diamond & Fagundes, 2010). Good emotion regulation, as you saw earlier, is in turn associated with fewer emotional and behavioral problems in childhood.

In late childhood, adolescence, and even adulthood, children who have enjoyed secure relationships with their parents early in life often continue to be well adjusted and to enjoy good relationships later in life (Elicker, Englund, & Sroufe, 1992; Simpson et al., 2007). In a revealing longitudinal study spanning the years from infancy to the early 20s, Jeffrey Simpson and his colleagues (2007) found that secure attachment in the Strange Situation at 12 months of age was linked to the quality of a child’s peer relations in elementary school, which in turn predicted quality of friendships in adolescence, which in turn predicted the emotional quality of romantic relationships in early adulthood. Although quality of attachment during infancy and quality of romantic relationship in adulthood were not significantly correlated, they were indirectly associated through a chain of influence in which the quality of relationships in each developmental period affects the quality of relationships in the next period (see Figure 14.4); see also Cassidy & Shaver, 2016; Fraley et al., 2013).

Figure 14.4 Simpson et al. (2007) found that relationship quality at each step in development affects relationship quality at the next step.

^{Source: From Simpson, J. A., Collins, W. A., Tran, S., & Haydon, K. C. (2007). Attachment and the experience and expression of emotions in romantic relationships: A developmental perspective. Journal of Personality and Social Psychology, 92, 355-367. © 2007 American Psychological Association. Reprinted with permission.}

In sum, children are unlikely to develop normally if they never have the opportunity to form an attachment or if their relationships with adults are repeatedly severed. By contrast, a secure attachment during infancy and early childhood has positive implications for later intellectual, social, and emotional development and for the quality of later relationships. Yet you must avoid concluding that infants who are securely attached to their caregivers are forever blessed—or that infants who are insecurely attached to their caregivers are doomed. Early attachments may have few long-term consequences if they change later—and they quite often do. Factors such as a decline in a mother’s sensitivity, the departure of a father from the home, and negative life events in the family can change a secure attachment into an insecure one—just as improvements in a mother’s sensitivity or a family’s circumstances can convert an insecure attachment into a secure one (Roisman & Booth-LaForce, 2014; and see Beijersbergen et al., 2012). As Bowlby appreciated, then, internal working models are just that: working models, subject to revision in response to later social experiences (Sroufe et al., 2005).

All things considered, the Bowlby–Ainsworth attachment theory is well supported. Studies of the long-term consequences of early attachment support Bowlby’s claim that internal working models formed early in life shape later relationships and development. Still, many of us learn new social skills and different attitudes toward relationships in our later interactions not only with parents but also with peers, friends, lovers, and spouses. It is time, then, to supplement this description of parent–child relations in infancy with a look at the “second world of childhood”—the world of peer relations.

Intrigued by parallels between an infant’s attachment to a parent figure and a young adult’s love for a romantic partner, researchers like Phillip Shaver and his colleagues began studying adult romantic relationships from the perspective of attachment theory (Fraley & Shaver, 2008; Mikulincer & Shaver, 2007). Many, though not all, love relationships qualify as attachment relationships; the intimacy component of Sternberg’s triangular theory of love suggests a caring attachment between romantic partners. And signs of attachment anxiety early in a dating relationship are at least as important as passion or sexual desire in predicting whether a serious relationship will evolve (Eastwick & Finkel, 2008). Like the infant who is attached to a parent, the adult who is in love experiences a strong emotional bond to her partner, wants to be close, takes comfort from the bond, is upset by separations, and finds the partner irreplaceable, as illustrated by the story of star-crossed lovers Mike and Marcy at the start of the chapter. Adults often have other attachments, of course: to their children, their parents, and sometimes their best friends.

Figure 14.8 shows how the internal working models that Bowlby theorized we start to construct early in life may affect the quality of our romantic relationships (Bartholomew & Horowitz, 1991; Crowell, Fraley, & Shaver, 1999). It describes the four adult attachment styles that result when view of the self is either positive or negative and view of other people is either positive or negative. These four adult attachment styles can also be described in terms of two dimensions of attachment (Roisman, 2009): anxiety (extent of concern about whether partners will be emotionally available and responsive) and avoidance (extent of discomfort being intimate with and depending on a partner).

Figure 14.8 Adult attachment styles based on internal working models of self and other people arising from early experiences in relationships. These four types of attachment also differ in terms of attachment anxiety (fear of abandonment) and avoidance (discomfort being intimate and dependent) dimensions.

^{Source: Adapted with permission from Bartholomew, K., & Horowitz, L. M. (1991). Attachment styles among young adults: A test of a four-category model. Journal of Personality and Social Psychology, 61, 226-244.1991 American Psychological Association.}

Adults with a secure working model and adult attachment style feel good about both themselves and others; they are not afraid of entering intimate relationships or of being abandoned once they do. People with a preoccupied internal working model have a positive view of other people but feel unlovable. Resembling resistantly attached infants, they crave closeness to others as a means of validating their self-worth, tend to become overly dependent on their partners, and are highly fearful of abandonment; thus, they are high in attachment anxiety but low in avoidance.

Adults with a dismissing style of attachment often have a positive view of self but do not trust other people and dismiss the importance of close relationships. Like avoidantly attached infants, they defend themselves against hurt by not expressing their need for love. They downplay the importance of relationships, find it hard to trust or depend upon partners, feel that others want them to be more intimate than they wish to be, and keep partners at a distance. That is, they are low in attachment anxiety but high in avoidance. Bowlby (1973) described dismissing or avoidant individuals as “compulsively self-reliant.” Finally, adults with a fearful internal working model show both high anxiety and high avoidance, resembling infants with disorganized attachments. They take a dim view of both themselves and other people and display a confusing, unpredictable mix of neediness and fear of closeness. You may wish to see if you can identify the adult attachment styles expressed by the statements in Engagement 14.1.

In a pioneering study, Cindy Hazan and Phillip Shaver (1987) demonstrated that adults’ styles of attachment are related to the quality of their romantic relationships (see also Creasey & Jarvis, 2009). For example, adults with a secure attachment style experience a good deal of trust and many positive emotions in their current love relationships, and their relationships tend to last longer than those of adults with insecure attachment styles. Avoidant lovers fear intimacy, whereas resistant individuals tend to be obsessed with their partners and jealous. When engaged and married partners discuss problems in their relationships, those with a secure attachment style calmly share their feelings and thoughts; avoidant-style adults show physiological signs of shutting down or inhibiting their true feelings; and resistant-style adults become highly emotionally aroused (Roisman, 2007).

The quality of the parent–child relationship that an adult experienced earlier in life predicts both adult attachment style and romantic relationship quality (Fraley et al., 2013; Grossmann et al., 2002a; Howe, 2011; Raby et al., 2015). Attachments to friends in childhood and adolescence help account for the link between the quality of early parent–child attachments and later romantic attachments—and they contribute something extra of their own, too (Fraley & Roisman, 2015). Sensitive partners can make a difference too—for example, turning an avoidant/dismissing partner into a more trusting and cooperative one by expressing confidence in him and appreciating his efforts (Farrell et al., 2016; Simpson & Overall, 2014). So, as Bowlby theorized, internal working models of self and others formed on the basis of parent–child interactions—but modified by later experiences—color the quality of later relationships. Application 14.1 illustrates how attachment theory is being applied to improve both parent–child and couple relationships.

Adults’ attachment styles predict a lot more about adults than the quality of their romantic relationships, though (see Cassidy & Shaver, 2016). A secure internal working model provides a secure base for exploration, predicting the extent to which adults have the confidence and curiosity to explore and master their environments (Elliot & Reis, 2003; Luke, Sedikides, & Carnelley, 2012). Securely attached adults enjoy their work and are good employees in a variety of ways, preoccupied (resistantly attached) adults may need a lot of emotional support and grumble about not being valued enough, and dismissing (avoidantly attached) adults may bury themselves in work and do little socializing with coworkers (Hazan & Shaver, 1990; and see Paetzold, 2015).

Internal working models and attachment styles are also related to an adult’s capacity for caregiving—most importantly, to being a sensitive and responsive parent. Mothers and fathers who had secure relations with their parents tend to interact more sensitively with their children and form more secure attachments with them than parents whose early attachments were insecure (Roisman & Groh, 2011; Shlafer et al., 2015). Attachment styles also predict the quality of caregiving adults provide to their aging parents (Morse et al., 2011).

Finally, attachment styles are related to well-being and both physical and mental health in later life (Howe, 2011; Waldinger et al., 2015). In a study of 60- to 99-year-olds, Victor Cicirelli (2010) concluded that 50% of the older adults studied could be categorized as secure, most of the rest (38%) dismissing-avoidant, and only 12% preoccupied or fearful. As it turns out, older adults who have either a secure or dismissing-avoidant attachment style tend to be happier with their lives than those with either a preoccupied or fearful attachment style (Magai et al., 2001; Bodner & Cohen- Fridel, 2010). Indeed, across cultures middle-aged and older adults appear to be less anxious than young adults about relationships but more avoidant or dismissing (Chopik & Edelstein, 2014). Whether to cope with the loss of a spouse, cope with an imperfect marriage, or optimize independence and minimize dependency on others, then, many older adults seem to become quite self-reliant in their old age and are happy that way.

We can conclude that a secure attachment style is most likely to be associated with good functioning and high well-being throughout the life span (Van Assche et al., 2013; Waldinger et al., 2015). Secure, preoccupied, dismissing, and fearful adult attachment styles have significant implications for adult development— for quality of romantic relationships, exploratory and mastery behavior at work, parenting and caregiving, and adjustment in later life.