The latent structure of the adult attachment interview: Large sample evidence from the collaboration on attachment transmission synthesis

Celebrating more than 26,000 adult attachment interviews: mapping the main adult attachment classifications on personal, social, and clinical status

Since the development of the Adult Attachment Interview (AAI) in 1985, more than 26,000 AAIs have been administered, coded, and reported, representing 170 (wo-)man-years of work. We used multinomial tests and analyses of correspondence to compare the AAI distributions in various cultural and age groups, in mothers, fathers, high-risk, and clinical samples with the combined samples of North American non-clinical, non-risk mothers (22% dismissing, 53% secure, 8% preoccupied, and 17% unresolved loss or other trauma). Males were more often classified as dismissing and less frequently classified as secure compared to females (except adoptive fathers), and females were more frequently classified as unresolved (but not more often preoccupied) compared to males. A combination of high scores on the unresolved and insecure-preoccupied dimensions was shared by borderline personality disorder, autism spectrum disorders, and gender dysphoria, while combined high scores on the unresolved and insecure-dismissing dimensions characterized anxiety problems, obsessive-compulsive and thought disorders.

Evidence of a developmental shift in the nature of attachment representations: a longitudinal taxometric investigation of secure base script knowledge from middle childhood into adolescence

Prior research suggests that secure base script knowledge is categorically distributed in middle childhood but becomes dimensionally distributed from late adolescence onward, potentially indicating a developmental shift in the nature of secure base script knowledge. Secure base script knowledge may initially be sparse, giving rise to categorical individual differences, while increased relational experiences later in development might contribute to more elaborated secure base script knowledge and dimensional individual differences. However, the cross-sectional nature of prior research limits inferences about developmental changes. To address this, we conducted a three-year, three-wave longitudinal study with a Western European sample transitioning from middle childhood to adolescence. At Wave 1 (n = 599, Mage = 10.30), secure base script knowledge was categorically distributed. By Wave 2 (n = 435, Mage = 11.30), distribution was ambiguous, and by Wave 3 (n = 370, Mage = 12.09), individual differences were dimensional. These results suggest a developmental shift in secure base script knowledge during the transition into adolescence.

Longitudinal Associations Between Attachment Representations Coded in the Adult Attachment Interview in Late Adolescence and Perceptions of Romantic Relationship Adjustment in Adulthood

Increasingly, researchers have operationalized Adult Attachment Interview (AAI)-derived attachment representations as reflecting individual differences in secure base script knowledge (AAIsbs)-the degree to which individuals show awareness of the temporal-causal schema that summarizes the basic features of seeking and receiving effective support from caregivers when in distress. In a series of pre-registered analyses, we used AAI transcripts recently re-coded for AAIsbs and leveraged a new follow-up assessment of the NICHD Study of Early Child Care and Youth Development cohort at around age 30 years (479 currently partnered participants; 59% female; 82% White/non-Hispanic) to assess and compare the links between AAIsbs and traditional AAI coding measures at around age 18 years and self-reported romantic relationship quality in adulthood. Higher AAIsbs predicted better dyadic adjustment scores in adulthood (r = 0.17) and this association remained significant controlling for other AAI-derived coding measures, as well as sociodemographic and cognitive functioning covariates. Findings extend previous evidence pointing to the predictive significance of AAIsbs for multiple adult functioning domains.

Adult attachment assessed via the ASA and AAI: Empirical convergence and links with autonomic physiological responding during attachment assessments

This study examined the empirical convergence of Attachment Script Assessment (ASA) deactivation, hyperactivation, and anomalous scripts with conceptually corresponding attachment patterns assessed via the Adult Attachment Interview (AAI), and the significance of ASA dimensions for autonomic physiological reactivity during adult attachment assessments. Young adults' (50% male; Mage = 19 years; 80% White/European American) ASA deactivation, hyperactivation, and anomalous content were significantly associated with AAI dismissing (r = .26-.38), preoccupied (r = .31-.35), and unresolved (r = .37) states of mind, respectively. ASA hyperactivation and anomalous content were associated with heightened RSA reactivity to the AAI and ASA, aligning with expectations that these attachment patterns capture the tendency to heighten expressions of negative, traumatic experiences. ASA deactivation was associated with smaller increases in electrodermal activity to the ASA-indicative of less sympathetic arousal-converging with the tendency of individuals higher in deactivation to avoid discussing attachment themes in the ASA.

The predictive significance of attachment script assessment hyperactivation and deactivation: evidence of associations with romantic relationship functioning

This study examined the significance of a novel coding system for evaluating hyperactivation, deactivation, and anomalous content in the Attachment Script Assessment for romantic relationship functioning. In a sample of 208 couples (69% White, Mage 28.7 years), we tested whether ASA hyperactivation and deactivation were associated with theoretically relevant correlates, including observed behavior, parasympathetic reactivity, self-reported affective reactivity to conflict, and relationship satisfaction. Exploratory analyses examined associations of secure base script knowledge (SBSK) and anomalous content with these outcomes. ASA hyperactivation and deactivation were associated with behavioral, physiological, and self-reported functioning in theory-consistent ways. Anomalous content was not associated with romantic functioning. SBSK was associated with satisfaction, but this was not robust to covariates. Findings support the predictive validity of the hyperactivation and deactivation dimensions and suggest that these scales complement SBSK, enabling researchers to assess a wider range of behavioral and physiological indicators associated with distinctive forms of attachment insecurity.

Deactivation, hyperactivation, and anomalous content in the attachment script assessment: stability over time and significance for parenting behavior and physiology

This study examined the stability of Attachment Script Assessment (ASA) deactivation, hyperactivation, and anomalous content and their significance for parenting outcomes in mothers (Mage = 31 years; 78% White/European American) and 6-month-old infants. Comparable to ASA secure base script knowledge (SBSK), mothers' ASA deactivation, hyperactivation, and anomalous content were significantly, moderately stable over two years (r's = .40 - .43). Mothers' ASA hyperactivation and anomalous content were associated with greater maternal intrusiveness, whereas ASA deactivation was associated with greater detachment and less intrusiveness. Only ASA anomalous content was associated with lower maternal sensitivity. Mothers' ASA deactivation was associated with less dynamic change in respiratory sinus arrhythmia during the Still-Face Procedure-reflective of limited mobilization of physiological resources to support responding to infants. Findings support the validity of ASA deactivation, hyperactivation, and anomalous content scripts, and demonstrate their utility in examining adult attachment stability and predictive significance for parent-child outcomes.

Free from your experiences to grow: belief in free will moderates the relationship between attachment avoidance and personal growth initiative

BACKGROUND

Attachment theory proposes that attachment security facilitates personal growth. However, attachment security origins in relationship history, and thus, how people treat their experiences may influence the outcomes of attachment security. People differ in the degree in believing that human beings have free will, and belief in free will may influence the relationship between experiences and outcomes. The present cross-sectional study investigated the relationships between attachment security, belief in free will, and personal growth initiative.

METHODS

We used the cross-sectional data of 346 Chinese college students for data analysis, including correlational analyses, regression, and moderation analyses. The nine-item Chinese version of the Experiences in Close Relationships-Relationship Structures Scale, the sixteen-item Chinese version of the Personal Growth Initiative Scale-II, and the seven-item Free Will subscale of the Chinese version of the Free Will and Determinism Plus Scale were utilized.

RESULTS

Results showed attachment avoidance and belief in free will, not attachment anxiety, was associated with personal growth initiative. Belief in free will moderated the association between attachment avoidance and personal growth initiative. When the centered score of belief in free will was higher than 0.64, attachment avoidance was no longer associated with personal growth initiative. 85.84% of our data were below this Johnson-Neyman significance region, and 14.16% were above. In other words, only those who scored higher than 0.64 on free will beliefs were able to pursue personal growth despite their high attachment avoidance.

CONCLUSIONS

These findings suggest that when believing in free will, avoidantly attached people may believe in their ability to pursue personal growth and think their future has more possibilities, not influenced by other factors like social support, which they think they lack.

Failure to resolve loss and compromised mentalizing in female inpatients with major depressive disorder

Attachment and mentalizing are central concepts in research on the etiology, course, and treatment of depression. The goal of this cross-sectional study was to clarify the unique value of these constructs in characterizing the presence, severity, and chronicity of depression. We examined 50 female inpatients suffering from Major Depressive Disorder (MDD) in comparison to 47 matched healthy controls regarding their attachment states of mind, mentalizing capacities, and clinical variables indicating depression severity and chronicity (e.g. illness duration, number of hospitalizations). In the group of depressed patients, unresolved attachment with regard to loss was significantly overrepresented. Dimensionally, patients were more disorganized and more insecure, whereas there was no difference on the dismissing-preoccupied dimension between the two groups. Mentalizing was significantly lower in patients than in healthy controls. Logistic regression analysis revealed attachment insecurity, mentalizing deficits, and unresolved loss to be incrementally relevant to predict MDD. Correlations with clinical parameters in the group of depressed patients showed positive associations between mentalizing deficits, attachment insecurity, and variables indicating illness chronicity. Our findings highlight the relevance of the inability to resolve or reappraise loss experiences in depressive states.

Affective temperaments and depressive symptoms: The mediating role of attachment

BACKGROUND

Studies showed that affective temperaments and attachment are associated with depressive symptoms, and that they bi-directionally influence each-other. The aim of this study is to explore mechanisms underlying the relationship between the affective temperaments (i.e., depressive, cyclothymic, irritable, hyperthymic and anxious), interview-based attachment, and depressive symptoms.

METHODS

A sample of 61 adolescents and young adults outpatients were asked to complete the Temperament Evaluation of Memphis, Pisa, and San Diego Autoquestionnaire (TEMPS-A) and the Hamilton Rating Scale for Depression (HAM-D). Attachment was assessed through the Adult Attachment Interview (AAI) and employing a dimensional approach to obtain continuous measures. Mediation models were performed with the affective temperaments as predictor, depressive symptoms as the criterion variable, and attachment dimensions as mediators.

RESULTS

Findings showed significant direct effects between all the affective temperaments and depressive symptoms. Only the cyclothymic (β = 0.22; SE = 0.1; 95% IC = 0.05, 0.42) and irritable (β = 0.21; SE = 0.09; 95% IC = 0.04, 0.4) temperaments showed an indirect effect on depressive symptoms through secure-insecure attachment. Dismissing attachment did not predict either the affective temperaments nor depressive symptoms. Preoccupied attachment significantly predicts depressive symptoms and, when controlling for it, the hyperthymic temperament no longer directly associates with depressive symptoms.

LIMITATIONS

The cross-sectional study design limit conclusion about causation and directionality.

CONCLUSIONS

Secure attachment could be a protective factor for depressive symptoms for individuals with a cyclothymic or irritable temperament. Differently, the hyperthymic temperament loses its protective role toward depressive symptoms when accounting for preoccupied attachment.

A Lifespan Development Theory of Insecure Attachment and Internalizing Symptoms: Integrating Meta-Analytic Evidence via a Testable Evolutionary Mis/Match Hypothesis

Attachment scholars have long argued that insecure attachment patterns are associated with vulnerability to internalizing symptoms, such as depression and anxiety symptoms. However, accumulating evidence from the past four decades, summarized in four large meta-analyses evaluating the link between insecure attachment subtypes and internalizing symptoms, provide divergent evidence for this claim. This divergent evidence may be accounted for, at least in part, by the developmental period under examination. Specifically, children with histories of deactivating (i.e., insecure/avoidant) but not hyperactivating (i.e., insecure/resistant) attachment patterns in infancy and early childhood showed elevated internalizing symptoms. In contrast, adolescents and adults with hyperactivating (i.e., insecure/preoccupied) but not deactivating (i.e., insecure/dismissing) attachment classifications showed elevated internalizing symptoms. In this paper, we summarize findings from four large meta-analyses and highlight the divergent meta-analytic findings that emerge across different developmental periods. We first present several potential methodological issues that may have contributed to these divergent findings. Then, we leverage clinical, developmental, and evolutionary perspectives to propose a testable lifespan development theory of attachment and internalizing symptoms that integrates findings across meta-analyses. According to this theory, subtypes of insecure attachment patterns may be differentially linked to internalizing symptoms depending on their mis/match with the developmentally appropriate orientation tendency toward caregivers (in childhood) or away from them (i.e., toward greater independence in post-childhood). Lastly, we offer future research directions to test this theory.

It takes a village: a call for engaging attachment with adjunct disciplines to clarify "in-house" clinical conundrums

Researchers have long discussed the rich clinical applications of attachment theory. Specifically, clinicians have been compelled by the idea that insecure attachment, broadly speaking, may bring about stressful (real or perceived) interpersonal experiences that increases risk for internalizing symptoms. However, recent meta-analyses examining the links between attachment representations and internalizing symptoms have challenged this assumption, highlighting the importance of considering insecure attachment subtypes in understanding one's vulnerability for internalizing symptoms. Here we expand on this special issue's target papers and propose that, despite an increase in citation impact of clinically relevant attachment research, there are still core theoretical and methodological questions left unanswered. We highlight three clinical conundrums: (1) hyperactivating, but not deactivating, attachment is linked to increased internalizing symptoms in adolescence and adulthood; (2) the magnitude of the associations between insecure attachment subtypes and internalizing symptoms varies depending on the developmental period; and (3) self-reported, but not narrative-based, deactivating attachment is associated with increased internalizing symptoms. We call for engagement with adjunct academic disciplines to elucidate these issues. These clinical conundrums have important ethical implications regarding how we understand insecure attachment and necessitate close theoretical and empirical attention before attachment findings can truly inform clinical practice.