mothers with a dismissing attachment. You can see an example of the pictures used and the brain imaging data in Figure 4.5.
Genetic research has also implicated the neurotransmitter serotonin in attachment. Serotonin helps maintain stable mood and affect, and dysfunction is associated with depression and anxiety. With regard to romantic attachment, one study concluded that limited serotonin produces altered perceptions of relationships (Bilderbeck et al., 2011). Bartz (2011) posited that serotonin depletion has a similar effect to fearful attachments (i.e., preoccupied) and a temporary reduction in serotonin induces a temporary fearful attachment state.
FIGURE 4.5 Secure Mothers Showed Greater Activation in Areas of the Brain Associated with Reward.
Source: Strathearn L., et al. (2009). Reproduced with permissions of Springer Nature.
Hormones
Attachment research has predominantly focused on two hormones: oxytocin and cortisol. Oxytocin is a critical hormone in lactation, nurturance, and trust. Higher oxytocin levels are associated with maternal sensitivity to infants (Feldman, Weller, Zagoory‐Sharon, & Levine, 2007), quality of play and contact with infants (Feldman, Gordon, Schneiderman, Weisman, & Zagoory‐Sharon, 2010), and maternal trust in adolescents (Venta et al., 2017). Further, cortisol, a hormone released by the HPA axis in response to stress, has been found to differ between secure, disorganized, and insecure children (Spangler & Grossman, 1993). Similarly, Schieche and Spangler (2005) found that securely attached and behaviorally inhibited infants experienced declines in cortisol after a challenging task, whereas avoidant infants experienced a decrease in cortisol early on, followed by an increase in cortisol after the task. While the design and findings of these studies are complex, they lend some support for a cortisol‐coping hypothesis (Spangler & Grossman, 1993).
Temperament
Vaughn, Bost, and Van IJzendoorn (2008) posit that while attachment and temperament domains overlap, they are discrete: attachment exists first in a caregiving relationship and is internalized later, whereas temperament is always an attribute of the individual. Indeed, secure versus insecure groups do not show statistically significant differences with regard to emotional reactivity (Marshall & Fox, 2005), behavioral inhibition (Stevenson‐Hinde & Marshall, 1999; Burgess, Marshall, Rubin, & Fox, 2003), or negative emotionality (Pauli‐Pott Haverkock, Pott, & Beckmann, 2007) in most studies though some have found reduced negative reactivity among securely attached children (Vaughn et al., 2008; van IJzendoorn et al., 2004). Further, some research suggests that the interaction between attachment and temperament is important in predicting behavior. For example, Bohlin, Hagekull, and Andersson (2005) reported that, for securely attached children, high behavioral inhibition was associated with higher social competence whereas, for insecurely attached children, high behavioral inhibition was associated with poorer social competence.
Cognitive Factors
Intelligence and language
Because secure attachments allow young children to venture out to explore their world, they underlie the capacity to interact with and learn from our environment. Indeed, a meta‐analysis by van IJzendoorn and colleagues (1995) concluded that insecure attachment is associated with both lower cognitive performance and poorer language competence. Securely attached infants perform better than insecurely attached infants on tests of cognitive development (Egeland & Farber, 1984; van Bakel & Riksen‐Walraven, 2004) and this pattern extends to intelligence testing in kindergarten (van IJzendoorn & Vliet‐Visser, 1988). In third and fourth grades, children who were more securely attached had (among other things) parents who provided more encouragement and assistance with school and were able to form better relationships and be more cooperative and self‐regulated in school (West, Matthews, & Kerns, 2013). Reciprocal models in which a child’s cognitive ability influences their development of secure attachments have also been supported (e.g., Stievenart, Roskam, Meunier, & van de Moortele, 2011) suggesting the link between cognitive development and attachment is likely to be bidirectional.
Social cognition
Our early attachment relations shape the way that we see, remember, and interact with the social world later on. For instance, studies have shown that securely attached adolescents attend to positive feedback about the self and seek it out from non‐attachment related targets (Cassidy, Ziv, Mehta, & Feeney, 2003); perceive attachment experiences and memories more favorably (Dykas, Woodhouse, Ehrlich, & Cassidy, 2010); and report more prosocial social‐information processing (Granot & Mayseless, 2012). In contrast, insecure adolescents process social information in a negatively biased way (Dykas, Cassidy, & Woodhouse, 2009).
Attentive caregivers are likely to treat their children as psychological agents, implicitly guiding the child to develop mental state reasoning abilities (Fonagy & Allison, 2012). For securely attached children, this process results in accurate social cognition, or reasoning about others’ underlying mental states (Dykas & Cassidy, 2011). This relation is evident into adolescence and beyond. For instance, Hünefeldt, Laghi, Ortu, and Belardinelli (2013) reported that attachment anxiety was associated with less accurate identification of mental states in the eyes of others. Likewise, attachment security has been associated with more frequent and more accurate social cognition (Ammaniti et al., 1999). Among adolescents with psychiatric diagnosis, Sharp and colleagues (2016) tied over‐attribution of mental states to decreased attachment security, and Venta and Sharp (2015) documented associations between attachment disorganization and poor social cognition.
Contextual Risk Factors
Parental psychopathology
Attachment security is more likely among the children of parents who are psychologically healthy. For instance, higher prenatal anxiety (Del Carmen et al., 1993); higher maternal stress (Atkinson et al., 2000); lower maternal psychological adjustment (NICHD Early Child Care Network, 1997); higher paternal depression (Booth‐LaForce et al., 2014); and higher parental depression, anxiety, and personality disorder (Hobson, Patrick, Crandell, Garcia‐Perez, & Lee, 2005) have been linked to insecure child attachments. Still, this area of research has yielded inconsistent findings, suggesting that protective factors are at play. It is most likely that parental psychopathology predisposes parents to parenting practices that are less conducive to secure attachments for their children.
Parent behaviors and attachments
Maladaptive parent behaviors, like child maltreatment, have been consistently associated with attachment insecurity—particularly attachment disorganization—with rates of insecure attachment as high as 90% among infants who are the victims of maltreatment (Cicchetti, Rogosch, & Toth, 2006). Moderate to heavy alcohol (O’Connor, Sigman, & Brill, 1987) and drug use (Melnick, Finger, Hans, Patrick, & Lyons‐Ruth, 2008) have been associated with high rates of infant insecure attachment as well. More generally, Main and Hesse (2006) suggested that “frightened or frightening” parental behaviors place a child at risk of disorganized attachment (Lyons‐Ruth & Jacobvitz, 2008). Main and Hesse (1990) posit that these behaviors emerge from a parent’s own traumatic experiences and create a conflict for the child about whether the parent is a source of distress or a solution to distress (Main & Hesse, 1990). Indeed, research suggests that a parent’s own attachment history is a relevant, albeit distal, risk factor for insecure attachment. Indeed, this is a model known as the “transmission