Bowlby's Attachment Theory Foundation
John Bowlby's attachment theory (1969-1988) posited that infants are biologically predisposed to form selective attachments to caregivers for survival and security. The infant's attachment behaviors (proximity-seeking, crying, clinging) and the caregiver's sensitivity and responsiveness create a bond serving as a secure base for exploration and as a haven during distress.
Bowlby (1973) hypothesized that repeated interactions create an 'internal working model' (IWM)—cognitive-affective representation of self, others, and relationships that guides expectations and behaviors in future relationships. This model is established by 18-24 months and shows remarkable stability across the lifespan (Fraley & Spieker 2003).
The theory integrates evolutionary biology (attachment ensures infant proximity to protectors, increasing survival), ethology (imprinting in birds, separation distress in primates), and cognitive psychology (mental representations guide behavior). Empirical support: infants who experience sensitive caregiving show lower cortisol (stress hormone) and stronger exploratory behavior, consistent with secure base function (Spangler & Grossmann 1993).
Ainsworth's Strange Situation and Attachment Types
Mary Ainsworth's Strange Situation procedure (1978, 1979) is a structured laboratory protocol lasting 21 minutes involving separation from and reunion with the caregiver. The procedure measures attachment security through infant behavior during reunion: (1) Secure attachment (B)—infant seeks proximity, readily comforted, returns to exploration; shown by ~55-65% of samples.
(2) Anxious-resistant (C)—infant resists comfort (angry, passive), shows difficulty settling; shown by ~8-10%. (3) Avoidant (A)—infant ignores or avoids caregiver upon reunion, suppresses distress; shown by ~20-25%.
Ainsworth demonstrated that these patterns reflected antecedent differences in maternal sensitivity: secure mothers responded promptly to infant cues, anxious mothers were inconsistent, avoidant mothers rejected contact. Remarkably, attachment classifications from infant Strange Situation predict relationship quality 20 years later in adulthood (Hamilton 2000).
Meta-analysis (Verhage et al. 2016, N=2,816) shows infant attachment security correlates with better social competence (r=0 38), fewer behavioral problems (r=-0 34), and better peer relationships (r=0
39). However, attachment security accounts for only 10-15% of long-term outcome variance, with parenting quality and socioeconomic factors explaining additional variance.
Main & Solomon's Disorganized Attachment
Mary Main and Judith Solomon (1986, 1990) identified a fourth attachment pattern not captured by Ainsworth's three types: disorganized/disoriented (D) attachment. Disorganized infants show contradictory or odd behaviors (approaching caregiver then freezing, showing fearful expressions toward the secure base, role-reversing).
Disorganized attachment is associated with frightening or frightened parental behavior, parental loss or trauma, or parental substance abuse. Prevalence is 10-15% in low-risk samples, 70% in maltreated infants (Lyons-Ruth et al.
2005). Predictive validity is strong: disorganized attachment predicts aggression, internalizing problems, and dissociative symptoms in childhood (Hesse & Main 2006). In adulthood, childhood disorganized attachment predicts unresolved loss/trauma status in the Adult Attachment Interview (AAI), emotional dysregulation, and interpersonal violence risk.
The mechanism likely involves fear of the attachment figure (paradox: the person meant to provide safety triggers fear), creating irresolvable cognitive-affective conflict.
Fraley's Adult Attachment and ECR-R
R. Chris Fraley et al. (2000) developed the Experiences in Close Relationships (ECR) scale to measure adult romantic attachment based on Bowlby's theory. The full ECR contains 36 items; the revised ECR-R (Fraley et al.
2006) contains 36 items assessing two continuous dimensions: (1) Anxiety dimension (12 items)—fear of abandonment, need for reassurance, preoccupation with relationship ('Do you worry that your partner won't care about you as much as you care about them?' ).
Anxiety correlates with anxious-resistant infant attachment (r=0 42). (2) Avoidance dimension (12 items)—discomfort with closeness and dependence, emotional distance ('Do you feel comfortable sharing your private thoughts and feelings with your partner?'
). Avoidance correlates with avoidant infant attachment (r=0 38). Four attachment styles emerge from combinations: secure (low anxiety, low avoidance), preoccupied (high anxiety, low avoidance), dismissing (low anxiety, high avoidance), fearful-avoidant (high anxiety, high avoidance).
The ECR-R shows strong psychometric properties (α=0 71-0 86 across dimensions). A large meta-analysis (N>100,000; Mikulincer & Shaver 2007) shows secure attachment predicts relationship satisfaction (r=0
43), trust (r=0 51), and longevity. Anxious attachment predicts conflict and jealousy (r=0 35-0 48). Avoidant attachment predicts sexual dissatisfaction and commitment avoidance (r=0 38-0 52).
Adult Attachment Interview and Intergenerational Transmission
Main's Adult Attachment Interview (AAI; 1985, 2000) is a 60-minute semi-structured interview asking adults to recall childhood attachment experiences and reflect on their influences. Responses are coded for coherence, consistency, and emotion management, yielding four classifications paralleling infant patterns: secure-autonomous, dismissing, preoccupied, and unresolved-disorganized.
Remarkably, parental AAI classification predicts infant Strange Situation classification at 78% accuracy (Hesse 2008, meta-analysis of 20 studies with N=1,666). This finding demonstrates intergenerational transmission of attachment: parents with secure internal working models create secure attachment relationships with their infants.
Mechanisms include parental sensitivity (mediated pathway; Hesse 2008) and direct transmission of trauma representations (parents with unresolved trauma may re-enact frightening behaviors). Interventions targeting parental attachment awareness (video feedback of parent-infant interaction) significantly improve infant attachment security and reduce behavioral problems (Juffer et al. 2005).
Attachment and Adult Outcomes
Adult attachment dimensions predict health, mental health, and relationship quality. Secure attachment is associated with lower cortisol at rest and during stress (Kidd et al. 2011), better emotion regulation, and higher relationship satisfaction.
Anxious attachment predicts anxiety disorders (r=0 37), depression (r=0 32), and suicidal ideation (r=0 28). Avoidant attachment predicts sexual dysfunction (r=0 35) and commitment avoidance.
A prospective study by Sprecher et al. (1998) followed couples for 14 years; secure attachment at baseline predicted relationship stability and satisfaction. The theory integrates with neuroscience: oxytocin and opioid systems mediate secure attachment bonding; chronic stress activation (high cortisol) disrupts secure base function.