JUDE CASSIDY

ABSTRACT This paper outlines some of the ways in which secure attachment is associated with the capacity to participate in successful intimate relationships.

The paper begins with the discussion of 4 key abilities required for intimacy: the ability to seek care, the ability to give care, the ability to feel comfortable with an autonomous self and the ability to negotiate. Bowlby's attachment theory (1969/1982, 1973, 1980) and related research are presented as a framework for understanding the development of these abilities.

Next, attachment, intimacy and sexuality are discussed. In the final section, attachment, intimacy and truth are considered.

Emotions are part of our genetic heritage. Fish swim, birds fly and people feel. Sometimes we're happy, sometimes we aren't; but sometimes in our life we're sure to feel anger and fear, sadness and joy, greed and guilt, lust and scorn, delight & disgust.

While we aren't free to choose the emotions that arise in us, we're free to choose how and when to express them, provided we know what they are. That's the crux of the problem. Many people have been educated out of knowing what their feelings are.

When they hated, they were told it was only dislike. When they were afraid, they were told there was nothing to be afraid of. When they felt pain, they were advised to be brave and smile. Many of our popular songs tell us 'Pretend you're happy when you aren't.'

What is suggested in the place of this pretense? Truth. Emotional education can help children to know what they feel. It's more important for a child to know what he feels than why he feels it. When he knows clearly what his feelings are, he is less likely to feel ambivalent inside. (Haim Ginott, 1965, p. 34)

It's likely that that most people view their attachment relationships as their most intimate relationships. Yet surprisingly, examination of the developmental origins of intimacy hasn't been a focus of attachment theorists.

Although the American Psychiatric Association's Diagnostic & statistical manual of mental disorders (DSM-IV; 1994) contains no diagnostic category uniquely devoted to difficulties with intimacy (such difficulty isn't in itself psychopathology), there are links between difficulties with intimacy and a number of disorders.

Furthermore, understanding intimacy is important to clinicians because of the clear evidence that problems with intimacy contribute to both physiological and health problems as well as to psychological problems (e.g. Berman & Margolin, 1992; Fisher & Stricker, 1982; Loevinger, 1976; Pennebaker, 1990).

According to Webster's New Collegiate Dictionary, intimate means 'belonging to or characterizing one's deepest nature' (1971, 7th ed., p. 444). As Reiss and Patrick (1996) pointed out, the Latin intimus means 'innermost' & intimare means 'to make the innermost known'.

Thus, discussion of intimacy addresses one's deepest nature, one's innermost self, the core of one's being, the truth about who one really is - what the poet Yeats called 'heart-revealing intimacy'. Intimacy, therefore, is making one's innermost known, sharing one's core, one's truth, one's heart, with another and accepting, tolerating the core, the truth, of another.

It's being able to tell both the good and the bad parts of oneself, to tell of anger, ambivalence, love and to accept both the good and the bad parts of another, to accept anger, ambivalence, love. It's to share the self: one's excitements, longings, fears and neediness and to hear of these in another.

This paper begins with discussion of 4 key abilities required for intimacy and of how Bowlby's attachment theory (1969/1982, 1973, 1980) and related research contribute to an understanding of the development of these abilities. My proposition is that secure attachment facilitates these abilities. I'll first describe both theory and data in relation to this proposition.

Next, I discuss attachment, intimacy and sexuality. In the final section, I focus on attachment, intimacy and truth.

The ability to seek care relates to what attachment theorists call the 'attachment behavioral system', a biologically based system of behaviors, most likely to be activated in times of threat, that results in an individual's gaining or maintaining proximity to an attachment figure (Bowlby, 1969/1982; see Cassidy, 1999a, for a summary).

The attachment system is thought to have evolved because individuals who protested separation and sought an attachment figure when threatened were more likely to survive to pass on the genes for this tendency.

Although the attachment system has been studied mostly in infants and mothers, Bowlby (1979a) made it clear that attachment is important 'from the cradle to the grave' and that people of all ages do best when they have a trusted attachment figure, or a few such figures, to whom they can turn in times of trouble.

The reason that seeking care is important for intimacy is that, in the words of poet Robert Hass (1998),

'In this life, the heart is going to be injured'.

Life, for all humans, involves times when the innermost core is filled with fear, sadness, anger, grief and a person will want care because, according to Bowlby, humans are biologically predisposed to want care at such times.

Intimacy means sharing those feelings and that wish for care. Thus, for intimacy, the care-seeking system - the attachment system - has to be functioning well. A person must be able to turn to others, appropriately selected others, effectively, in times of trouble.

Secure attachment

What helps the attachment system function well? Two factors, both related to trust, are important. There are extensive data that secure infants have had experiences with their parents that would contribute to their developing this sort of trust.

First, the trust that others are available and sensitively responsive is important. A number of studies report that secure infants, more than insecure infants, have mothers who are sensitively responsive, particularly when the infants are upset and signal for comfort.

These mothers have been found to be more accepting, co-operative, available, comforting and tender than other mothers (e.g. Ainsworth, Blehar, Waters, & Wall, 1978; Belsky, Rovine, & Taylor, 1984; Egeland & Farber, 1984; NICHD Early Child Care Research Network, 1997).

In fact, a recent meta analysis examining 66 studies and over 4,000 dyads found what in meta-analytic terms is considered a 'moderately strong' association between maternal sensitivity and secure infant attachment (de Wolfe & van Ijzendoorn, 1997).

According to Bowlby, based on the experiences with the mother, the child develops a mental representation (Bowlby calls this a representational model, or an internal working model [IWMJ) of the mother.

Secure children are thought to develop representations of the parent as loving, responsive and sensitive. There are an increasing number of studies that support this proposition. Main, Kaplan and Cassidy (1985) reported the first work examining attachment and children's mental representations, with a longitudinal sample followed from infancy to age 6.

Attachment was assessed in infancy with Ainsworth's Strange Situation procedure (Ainsworth et al., 1978).

At age 6, children's mental representations related to attachment were examined. Children were asked to discuss photographs of other children undergoing increasingly stressful separations from their parents (e.g. parents bringing the child to the first day of school, parents leaving for a weekend, parents leaving for a two-week trip).

At 6, the stories of children who had been securely attached as babies were more often viewed as reflecting representations of the mother as responsive.

A series of studies has followed. Several researchers have examined attachment representations in pre-school and early school-aged children (e. g., Bretherton, Ridgeway, & Cassidy, 1990; Slough & Greenberg, 1990; Shouldice & Stevenson-Hinde, 1992; Solomon, George, & De Jong, 1995).

The methods of these studies have varied. Some involve doll stories, some photographs. In all cases, the attachment system was activated (the child was hurt, frightened; there was a separation). In all the studies, secure children were more likely than insecure children to have positive representations of the mother as available, responsive, comforting (see Solomon & George, 1999b, for a review).

Bowlby (1973) proposed that through the process of generalization secure children come to have more positive representations of others in general. A second aspect of trust important for a well-functioning attachment system is trust in the self as lovable.

According to Bowlby, closely intertwined with the IWM of the attachment figure is the IWM of the self. If, for instance, a child is loved and valued, that child will come to view himself or herself as lovable and valuable. If, however, a child is dismissed or not loved, that child will come to view herself or himself as not lovable and worth little.

This notion - that people learn about themselves from seeing how others view them is a widely held view within several theories, beginning with the earliest theorizing about the self (Baldwin, 1897; Cooley, 1902; Mead, 1934; Sullivan, 1953).

Data from several laboratories support this notion that secure  attachment is associated with positive representations of the self. The first evidence came from Sroufe and Egeland's longitudinal Minnesota Mother-Child Project (Sroufe, 1983).

Attachment was assessed in infancy in a large sample of poor children; in pre-school, teachers used 3 measures to rate children's self esteem. Pre-schoolers who had been securely attached to their mothers in infancy had higher self-esteem than children who had been insecurely attached.

In my own work (Cassidy, 1988), I used doll stories to examine 6-year-olds' representations of the self within the relationship with the mother. Attachment was assessed at age 6 with Main & Cassidy's (1988) reunion procedure.

The representations of secure children were more positive than those of insecure children: secure children had representations of the self as valued and worthy of care. Security scale scores were also significantly positively correlated with global self-esteem scores (as assessed both concurrently & 3 years later; Cassidy, 1988, 1999b). (See Verschueren, Marcoen, & Schoefs, 1996, for similar findings with a sample of children from the Netherlands.)

Thus, there are converging data that securely attached children have had positive experiences when they turned to others for care. And I propose that these experiences foster their capacity for intimacy by making them comfortable and confident in seeking care and by contributing to positive mental representations of others as caring and of themselves as worthy of care.

Secure individuals thereby bring a set of expectations into new relationships - expectations that others are accepting of them & their imperfections - and through a variety of self-fulfilling mechanisms, these positive expectations contribute to intimacy.

But what of children whose experiences of turning to others for care haven't been positive? Attachment researchers have identified 3 patterns of insecure attachment. I now discuss each of these in relation to capacities for seeking care.

Insecure / avoidant attachment

In Mary Ainsworth's pioneering naturalistic observation studies of mothers and infants, a group of babies was identified hose bids for comfort were rejected (approximately 25%; Ainsworth, 1967; Ainsworth et al., 1978).

The mothers of these babies were also uncomfortable with close bodily contact. Main & Solomon (1986) proposed that infants in a caregiving environment in which activation of their attachment system consistently led to painful rejection might develop a strategy in which their attachment system was activated as little as possible.

Such a strategy would be adaptive to their circumstances. (For discussion of avoidance as a defensive process, see also Ainsworth et al., 1978; Cassidy & Kobak, 1988; and Main, 1981a, 1990.)

There are both infants and adults who appear to suppress activation of the attachment system - who have trouble seeking care. Laboratory studies of babies separated from their mothers have revealed that some babies don't seek the mother for comfort or even interaction on reunion as most infants do.

These infants instead actively avoid the mother and become focused on toy play. Bowlby (1980; see also Ainsworth et al., 1978; & Main, 1981a) proposed that sight of the mother might activate the attachment system - which ordinarily leads the infant to engage in bids for contact and comfort - but because the past expression of such bids has been met with painful rejection, the baby has learned to turn defensively toward play activity.

The notion that this play activity is different from genuine play is supported by physiological data. Whereas the heart rate of babies truly interested in play typically decreases, the heart rate of these babies doesn't decrease, suggesting a lack of true focus on the play (Spangler & Grossmann, 1993).

Thus, it seems that these infants aren't shifting attention to the toys, but rather away from the mother. Bowlby (1969/1982) borrowed the term 'diversionary activity' from the ethologists to describe this behavior. Evidence that avoidant children have difficulty seeking care also comes from several additional studies of young children.

In one study examining infant behavior in the Strange Situation, observations of the communication patterns of mother-infant dyads revealed that avoidant infants communicated directly with their mothers only when they (the infants) were feeling well.

When distressed, these infants tended not to directly signal the mother and they didn't seek bodily contact (Grossmann, Grossmann, & Schwan, 1986). In another study of infants during the Strange Situation, avoidant infants didn't make visual contact with the mother when they were aroused, but only in situations with low emotional distress (Spangler & Grossmann, 1993).

Findings from this study converge well with findings from a third Strange Situation study in which insecure/avoidant infants, even though they were just as likely as some secure infants to be distressed during separation from mother, were less likely to seek comfort from (or even interaction with) the mother on reunion, but engaged instead in self-soothing behaviors (Braungart & Stifter, 1991).

In another study, avoidant children similarly masked negative affect, this time in a social situation - a tower-building game - with an adult stranger (Lutkenhaus, Grossmann, & Grossmann, 1985). Children's affective reactions to winning and losing this competitive building game were analyzed.

The avoidant children manifested sadness about losing during the game, but not when the game was over and the experimenter was available for social communication. In fact, during the social communicative exchange, there was a tendency for the avoidant children to replace sadness with smiling.

In contrast, securely attached children showed their sadness after the game when the adult was more available. Thus, even though avoidant children felt and displayed sadness, they did so only when there was no child-adult eye contact, thereby precluding any comfort or reassurance the adult might offer.

As Bowlby's claim that representations are largely experience-based would lead one to predict, the attachment-related representations of avoidant children aren't as positive as those of secure children. In one study, 3-year olds were presented with hypothetical vignettes in which a peer caused something negative to happen to the same-sex child story protagonist.

When asked what the mother in the story would do in response to this situation, children who had been avoidant infants were significantly less likely than other children to describe a supportive maternal response (Cassidy, Kirsh, Scolton, & Parke, 1996).

Findings from another study in which avoidant children had difficulty remembering stories (in a laboratory task) involving mothers who were responsive to distress also suggests that these children may lack representations of their mothers as supportive (Kirsh & Cassidy, 1997).

Representations related to separation and reunion also reflect less optimal experiences. In Main et al.'s (1985) study asking 6-year-olds about photographs of separation, children who had been avoidant infants were unable to suggest how children might cope with separation (see also Shouldice & Stevenson-Hinde, 1992).

In another study, a similar pattern emerged for avoidant 6-year-olds: fears about separation weren't expressed directly, and reunions were hampered by psychological unavailability of family members or by delay and distraction (Solomon et al., 1995).

The representational model of the self of avoidant children has also been found to be problematic. I've used a doll stories procedure to assess 'children's representation of the self in relation to  attachment' (Cassidy, 1988).

The stories of children classified insecure / avoidant to mother at age 6 reflected representations of the self as lacking a supportive relationship with the mother:

The doll protagonist was isolated and/or rejected & the importance of relationships was denied; the existence of conflict was denied in stressful situations, as was the need for help; and when there was a successful resolution, it was brought about entirely by the child.

In the same study, avoidant children responded to a puppet interview with descriptions of themselves as perfect in every way, even when repeatedly pressed for possibly diverse descriptors. This claim of perfection is viewed as a form of defensive idealization of the self by a child who fears rejection should any imperfection be found.

There are also data showing that adults with an insecure / avoidant romantic attachment style have trouble seeking care. Simpson, Rholes & Nelligan (1992) asked dating heterosexual couples to come to the laboratory.

Participants were told that the female partner of the couple would soon be subjected to a stressful experimental procedure and were surreptitiously observed while waiting for the procedure to begin.

Findings revealed that unlike secure women, insecure / avoidant women failed to seek care (defined as failing to share their concerns with their partners) and in fact withdrew from their partners as they became more anxious.

In another laboratory study, Collins & B. C. Feeney (2000) asked one member of a dating couple to disclose a personal problem or worry to his/her partner.

Avoidant attachment predicted ineffective (i.e. indirect) support-seeking. Similarly, in a naturalistic observational study of couples separating at the airport, Fraley & Shaver (1998) found that insecure/avoidant women were less likely than secure women to express their anxiety and seek comfort from their partners; avoidant women were, instead, more likely to pull away or withdraw from their partners.

A number of additional self-report studies also have found that avoidant adults are less likely than secure adults to seek support in response to stress.

For instance, in a study about responses to the missile attacks on Israel during the Gulf War, Mikulincer & his colleagues (Mikulincer, Florian, & Weller, 1993) found that avoidant adults were less likely to seek support from others, but instead used a strategy of distancing themselves from the threat (by trying to 'forget the whole thing') (see also Mikulincer & Florian, 1995; Ognibene & Collins, 1998; Pierce & Lydon, 1998).

Insecure / ambivalent attachment

Ainsworth (Ainsworth et al., 1978) observed that the mothers of insecure/ambivalent infants were inconsistent: sometimes loving and responsive, but only when they could manage, not in response to the infant's signals. An adaptive strategy for an infant whose mother was sometimes responsive, but other times too preoccupied, too overwhelmed, or too inept to respond would be staying near (Main & Solomon, 1986).

Because the infant can't count on the mother to monitor its needs, clinging and monitoring closely her availability becomes a good strategy, so that if need for the mother does arise, the infant will have quick access to her. Bretherton (1985) described this as the infant's having to take on more than its share of the burden of maintaining the connection.

This pattern has been described in terms of hyper vigilance and hyper activation of the attachment system. It is the child's lack of confidence in the caregiver's availability if needed that is thought to contribute to the child's hyper care seeking, the aim of which is to gain quick access to the mother.

The Strange Situation behavior of infants classified as ambivalent is characterized by extreme distress on separation and difficulty in calming on reunion; these infants display angry, resistant behavior toward the parent (Ainsworth et al., 1978).

This heightened negative emotionality can be viewed as a component of the child's strategy to gain the mother's attention  (Cassidy, 1994; Main & Solomon, 1986). The negative emotionality of the insecure / ambivalent child may be exaggerated and chronic because the child recognizes that to relax and allow herself or himself to be soothed by the presence of the attachment figure is to run the risk of then losing contact with the inconsistently available parent.

A history of exaggerated negative emotionality may explain findings revealing that these children have trouble maintaining a boundary between another person's distress and their own (Kestenbaum, Farber, & Sroufe, 1989).

An additional strategy may involve fearfulness in response to relatively benign stimuli. Through exaggerated fearfulness, the infant increases the likelihood of gaining the attention of a frequently unavailable caregiver should true danger arise (Main & Hesse, 1990).

However, the critical research needed to demonstrate a heightening of  negative affect - or indeed a heightened activation of the attachment system - by ambivalent infants doesn't exist. With the avoidant pattern, consistent evidence of minimization of negative affect does exist; avoidant infants and adults can feel negatively aroused while suppressing expression of such feelings (albeit with varying degrees of success; Dozier & Kobak, 1992; Fraley & Shaver, 1997; Grossmann et al., 1986; Lutkenhaus et al., 1985; Malatesta, Culver, Tesman, & Shepard, 1989).

Comparable evidence in relation to the ambivalent pattern would consist of a demonstration that the ambivalent individuals show greater negative reactivity than they actually feel. The rationalization of 'heightening' hasn't yet been attempted.

A child with the experiences typical of ambivalent infants might  develop a representation of the mother as inconsistently available and of the self as able to gain care only by sending exaggerated signals of need.

Studies examining the representations of young children have not yielded a consistent picture for the ambivalent group (which may be due in part to the small number of children typically classified in this group).

In one study, whose findings can be interpreted as reflecting a representation of the mother as relatively unavailable, the parent-child reunions of ambivalent 6 year-olds were characterized by 'delay and distraction' (Solomon et al., 1995). (See Cassidy & Berlin, 1994, for a review of both child and parent behavior associated with this pattern.)

In adult romantic attachments, this hyper activation is manifested as a seeming insatiability for closeness. Ambivalent adults describe  desires to merge with a partner (Hazan & Shaver, 1987); they portray themselves in relationships in ways classified as  'preoccupied' (a. A. Feeney & Noller, 1990); and they're particularly upset by relationship breakups (O. A. Feeney & Noller, 1992).

The heightened desire for closeness reflects an impairment of the attachment system that would be likely to interfere with intimacy, because to expect someone else to fill all one's needs, to be utterly dependent on another, to be threatened by another's natural  autonomy strivings, is bound to lead to ambivalence and resentment in both partners.

Insecure / disorganized attachment

A third more recently identified group of insecure infants and children is becoming better understood following increased research involving high-risk samples: the insecure / disorganized group (Cassidy & Mohr, in press; Hesse & Main, 2000; Lyons-Ruth & Jacobvitz, 1999; Main & Hesse, 1990; Solomon & George, 1999a; van Ijzendoorn, Schuengel, & Bakerman-Kranenburg, 1999).1

These children have had experiences of maternal behavior that is so frightening or unpredictable that they couldn't develop an organized, strategic response to it and so the attachment system is behaviorally disorganized.

It may be these insecure / disorganized children who have the most severe difficulties related to seeking care. Because their mothers are frightening to them, they've been placed in a very difficult position.

Frightening behavior by a parent activates simultaneous inevitably competing tendencies: to flee to the parent as a haven of safety and to flee from the parent as a source of alarm. In this paradoxical situation no organized behavioral strategy is available. Neither proximity seeking nor proximity-avoiding is a solution. The baby must seek care from a person who frightens him or her.

The baby is in a terrible position that can't be resolved, a position evident in the baby's behavioral response to reunion in the Strange Situation: freezing, disorientation, disorganization (Main & Hesse, 1990; see Main & Solomon, 1986, for a description of the behavioral indices of disorganization in infants).

The mental representations that children classified  insecure / disorganized have both of themselves and their mothers are hostile, violent, incoherent, frightening.

For example, in a study in which I interviewed children about themselves (Cassidy, 1988), I asked a 6-year-old what she 'hoped happened to her when she grew up'. She responded that she hoped she would 'get hit by a car'.

In a narrative procedure within the same study, I told children a story about a mother and child having a minor conflict about the dinner menu and asked what the mother would do next. One boy from this group answered, 'She would throw the boy in the garbage dump.'

The story responses of another disorganized child involved furniture flying across the room; in another, all family members died violently (see also Bretherton et al., 1990; Hodges & Steele, 2000; Solomon et a1., 1995).

Summary

Secure attachment is associated with positive care-seeking in both children and adults. In Ainsworth's Strange Situation, when secure babies are distressed, they're direct and clear in their care-seeking. They seek contact with little or no avoidance or resistance.

They're soon calmed and return to play. If secure babies aren't particularly distressed on separation, they're less likely to seek contact; they may be satisfied with proximity or with interaction across a distance.

Furthermore, secure attachment is associated with children's mental representations of the mother as supportive and helpful and of the self as worthy of care.

This doesn't appear to be the case for insecurely attached children. Although the 3 types of insecure attachment differ in many ways, each is likely to interfere with successful seeking of care and thus to impair the capacity for intimacy.

In adulthood, there's also evidence that security is associated with the ability to seek care. Furthermore, there's evidence that adult care-seeking relates to adults' own (retrospectively self-reponed) childhood experiences with their parents (A. Feeney & Noller, 1990; Hazan & Shaver, 1987; Mikulincer et al., 1993).

People generally have sex with those with whom they are intimate. People generally become intimate with those with whom they have sex. Sex is also linked to attachment starting in adolescence.

Examination of the data on timing of first sexual activity in adult relationships (on average, following 7 dates) and the data on the timing of coming to view an adult partner as an attachment figure (on average, after two years; Hazan & Zeifman, 1994), suggests that over the course of relationship formation, people have sex with people to whom they are not yet attached.

Increasing evidence, however, suggests that sexual behavior fosters the tendency for the two people to become attached to each other.

The physiology of attachment and sexuality

Evidence has emerged, primarily from research with non-human mammals, that oxytocin, a posterior pituitary peptide, may be a mechanism through which sexual activity facilitates adult attachment. Oxytocin is related to the development of three important bonds: the infant's attachment to the mother, the maternal bond to the infant, and the adult sexual bond.

For instance, evidence that oxytocin release is actively involved in the infant's social interaction with the mother comes from research with rats.

Oxytocin release seems to be associated with the calming that infants display during social contact, seems to greatly reduce the ultrasonic vocalizations that infant rats typically emit when separated from their mothers, and is required for the development of some pre-attachment processes, such as a baby's recognition of the mother's odor (Nelson & Panksepp, 1996; Singh & Hofer, 1978; see Carter, 1998, for a review).

Oxytocin may also play a role in maternal bonds to infants. When females in many mammalian species give birth, high circulating levels of estrogen trigger the proliferation of oxytocin receptors in many forebrain areas. Changes in oxytocin receptors may be related to induction of maternal behavior: administration of oxytocin to virgin female rats results in a sudden surge of maternal behavior; blocking oxytocin in new rat mothers interferes with maternal behavior (Nelson & Panksepp, 1998).

Thus, in these two intimate, powerful, emotionally significant,  individual-specific bonds - infant to mother and mother to infant - oxytocin seems to playa role in bond formation. If oxytocin also plays a role in attachment formation in adult romantic partners, sex may playa role because for several mammalian species, including humans, a large surge in oxytocin accompanies orgasm for both males and females (Carter, 1992).

The brain opiod theory of social attachment has received much empirical support and is also relevant (e.g. Panksepp, Siviy, & Normansell, 1985). Social contact can induce the release of opiods, and opiods result in powerful reduction of separation distress.

Opiods are rewarding: stimuli associated with them come to be greatly preferred, and this preference is slow to extinguish. According to Hazan and Zeifman (1999): Opiod conditioning would be expected to be the result of repeated anxiety and/or tension alleviating interactions.

Exchanges of this kind are a common feature of both infant-caregiver and adult romantic relationships. When a parent comforts a crying infant, the parent becomes associated ... with the alleviation of distress. Similarly, through repeated comforting exchanges, including the release of tension brought about by sexual climax, a lover comes to be associated with stress reduction and calming.

Relationships that develop into attachment bonds appear to be those in which heightened physiological arousal is repeatedly attenuated by the same person and in a context of close bodily contact. As such, attachment may involve the conditioning of an individual's opiod system to the stimulus ofa specific other. (italics added; p. 350)

The sexual system is, of course, biologically adaptive, just as the attachment system is; its activation typically ensures the creation of offspring who carry the individuals' genes forward into future generations. Yet what would explain the evolutionary processes that linked the sexual system and the attachment system?

Why would it be biologically, evolutionarily advantageous - that is, contributing to humans' reproductive fitness - for sexual activity to foster the process of becoming attached in adulthood?

To become attached to a sexual partner is adaptive because the attachment system then serves to keep the couple together, which, biologically, is important because it means that two people rather than one can protect the resulting offspring and ensure their survival.

It may well be the case that the sexual system serves to keep partners together long enough for their attachment systems to become organized around each other. Once the sexual system is not such a powerful force in keeping a couple together, the attachment system can exert considerable force in uniting the couple.

Given the complexities of sex within a relationship, and given the close ties to the attachment system - and particularly because the attachment system develops first and influences representational models - it is reasonable to expect that individual differences in attachment would relate to individual differences in sexual behavior.

Main (1990) has noted that because humans have evolved with the capacity to adjust to environmental variation, individual differences in attachment quality can be viewed in part as adaptations to the particular caregiving environment in which the individual finds himself or herself.

The same can be said about individual differences in sexual behavior: Such differences can be viewed in part as environmental adaptation (see Buss & Schmitt, 1993; Daly & Wilson, 1988).

Furthermore, it seems reasonable to assume that evolutionary pressures would contribute to a process whereby, given a stable environment, what an individual learns about attachment from early relationships would prove useful (that is, would enhance the individual's reproductive fitness) when making decisions about later attachment and sexual relationships.

What empirical evidence documents a link between the attachment and the sexual behavior systems? Does, for instance, a child's attachment history influence his or her later (adolescent and adult) sexual behavioral system?

Studies from a number of laboratories have revealed such influence among primates (Harlow & Harlow, 1965). In humans, Belsky and his colleagues described ways in which early attachment experiences may contribute to the nature and onset of subsequent sexual activity (Belsky, Steinberg, & Draper, 1991).

For example, Sroufe and his colleagues reported links between secure infant-mother attachment and later (pre-adolescent) gender boundary maintenance, which is thought to be a precursor of later competent sexual behavior (Sroufe, Bennett, Englund, & Urban, 1993).

In another study using adults' retrospective reports, greater maternal rejection in childhood was associated with adult sexual promiscuity (Brennan, Shaver, & Tobey, 1991).

What about reciprocal influences in adulthood? Is there evidence of an association between the attachment and the sexual systems? There are as yet few studies. In the most comprehensive study, Hazan and her colleagues found associations of self-reported attachment style and the frequency and enjoyment of various sexual behaviors in a sample of 100 adults (Hazan, Zeifman, & Middleton, 1994).

Secure adults were more likely to be involved in mutually initiated sexual activity and to enjoy physical contact than were other adults. This sexual activity was in the context of a primary relationship:

Secure adults were less likely than others to be involved in one-night stands or to engage in sex outside the primary relationship.

Anxious / ambivalent females reported greater involvement in exhibitionism, voyeurism and dominance/ bondage, whereas anxious/ambivalent males reported sexual reticence (see also Feeney, Noller, & Patty, 1993, who reported similar findings about anxious/ambivalent males).

Avoidant adults reported less enjoyment of physical contact, and their sexual behaviors were more likely to be those with sex without love. A similar picture of avoidant adults emerged from two other studies:

Feeney and her colleagues (1993) found them to have more accepting attitudes toward casual sex than others, and Brennan and Shaver (1995) found them more likely to engage in one-night stands and to agree with the notion that sex without love is pleasurable.

Stability and change

Thus far, I have built a model where early experiences with secure attachment increase later capacities for intimacy. I also propose that if a person has not had a history of secure attachment but has been able to forge secure attachments later in life and/or in some way rework representational models, the capacity for intimacy will be enhanced.

Schore (1994) argued that the parts of the brain most central to early attachment processes are 'the most plastic areas of the cortex', allowing, to an extent as yet unknown, for continuing reorganization throughout life.

Attempts to understand this process - why and when reorganization is possible and why and when it is not - will offer fascinating and important opportunities for addressing key developmental and clinical questions.

It may be that the time of pair-bonding - related to procreation of offspring and an important time for reproductive fitness - is a time when attachment related neural pathways are particularly open to change. Nelson and Panksepp (1998) reviewed the evidence that gonadal steroids exert strong regulatory influence on opiods, oxytocin and vasopressin, and concluded that this influence may be one mechanism of change:

The onset of puberty and large-scale increases in the synthesis and secretion of gonadal steroids could exert widespread changes and shifts in emphasis throughout the proposed affiliative circuitry of the brain.

This circuit would also undergo functional changes during other periods of time when gonadal steroids were elevated such as during mating and pregnancy ... Many studies now indicate that both early and later modifications of brain affiliative systems do occur, and it  may be time to empirically reassess the role of critical or sensitive periods in the manifestations of affiliative circuits with the brain and psychobiological dispositions of mammals. (p. 444)

Thus it may well be that new, adult attachments, in association with physiological processes related to sexual behavior, may facilitate new capacities for intimacy and secure attachment where few existed previously.

In the past, theorists (e.g. Bowlby, 1979a; Main & Goldwyn, 1998) have regarded the re-workings of mental representations that sometimes occur in adolescence and adulthood as resulting from cognitive development, from a change in representational processes and the development of formal operations.

It is useful to consider whether it may be - as Nelson and Panksepp (1998) proposed - that physiological changes associated with elevated gonadal steroids during puberty, mating, and pregnancy, also play a role in facilitating a reorganization of brain  representations and brain circuitry related to processing attachment-related information.