Conference Agenda

Although molecular genetic research initially revealed exciting results concerning the role of specific genetic polymorphisms tied to the dopaminergic and serotonergic systems in moderating the association between maternal sensitivity and infant attachment, in what was to become common in the psychological sciences, these initial findings from small samples failed to replicate in larger-sample investigations. Despite this cautionary evidence, a recent study shifted attention to yet another genetic marker of the opioid system, OPRM1 rs1799971, and provided evidence for its role in moderating associations between maternal insensitivity and child resistant and avoidant attachments assessed via the Modified Strange Situation procedure (MSSP; Tchalova et al., 2024). This preregistered replication report attempted to directly replicate Tchalova et al. (2024) by leveraging data from the NICHD Study of Early Child Care and Youth Development (SECCYD). In addition, we extended Tchalova et al. (2024) to examine the robustness of findings to other early observational assessments of attachment (i.e., Strange Situation Procedure [SSP], Attachment Q-Sort) and to more comprehensive assessments of maternal sensitivity utilizing data from the SECCYD and the Generation R Study, a comparably large study of Dutch children. Despite leveraging the two largest genetically-informed studies of attachment to date, we were unable to replicate the pattern of findings reported by Tchalova and colleagues (2024), providing yet another replication failure in the candidate gene literature. The only significant gene-by-environment interaction involving attachment patterns emerged for disorganized attachment during the SSP in the SECCYD, the one pattern for which Tchalova et al. (2024) reported no significant effects. Notably, however, this finding did not replicate in the Generation R Study. Analyses based on the Attachment Q-Sort also failed to support the original findings. These results highlight the need for genetically-informed research designs derived from current best practices to better understand the potential genetic contributions to child attachment.

Abstract

Individual differences in attachment security may be inborn or shaped by the social environment. Previous findings in infancy and early childhood point to a substantial role of the environment. In contrast, a large twin study in early adolescence showed considerable heritability. In this pre-registered study, we examined the heritability of attachment in middle childhood, hypothesizing moderate heritability in this age group. We included 415 same-sex twin pairs (52% girls, 58% monozygotic) between 8 and 11 years old (M = 9.59, SD = 0.79). Participants were recruited from an experimental cohort-sequential study including two age-overlapping longitudinal cohorts. Secure Base Script Knowledge was assessed with the Middle Childhood Attachment Script Assessment. Zygosity of the twins was determined using DNA samples. A significant common environment component (40%) was found in the absence of additive heritability. This result aligns with findings in infancy and early childhood and diverges from results in early adolescence. The gene-environment correlation hypothesis suggesting that older children more actively shape their experiences in social contexts may offer a plausible explanation for heritability of attachment in adolescence, but in middle childhood this mechanism did not yet tip the balance toward a larger role for additive genetics.

Short description

We examined the heritability of attachment in middle childhood in a group of 415 same-sex twin pairs (52% girls, 58% monozygotic) between 8 and 11 years old (M = 9.59, SD = 0.79). Secure Base Script Knowledge was assessed with the Middle Childhood Attachment Script Assessment. 40% of the variance in Secure Base Script Knowledge was explained by common environmental factors. Genetic factors played no significant role.

Attachment theory posits that the major contributions to attachment security and insecurity derive from the quality of caregiving; it is intrinsically a theory of environmental causation. Studies of infant and early childhood attachment shows that these assumptions are well founded, with strong shared environmental influences observed for attachment security and minimal evidence of genetic influences (Bokhorst et al., 2003; Roisman & Fraley, 2008). However, in middle childhood and adolescence evidence suggests that the influences driving attachment security shift towards genetic and non-shared environmental factors (Fearon et al., 2014; Runze et al., 2025). One explanation for this change centres on the role of evocative gene-environment correlation – parental interactions being shaped by genetic factors resident in the child or young person. In this talk, we report findings from a comparatively large twin study (N=592 twin pairs), where we examined the association between attachment, as measured using the Child Attachment Interview, and parental sensitivity, as measured in an 8-minute parent-adolescent interaction. Genetic modelling indicated that approximately 30% of the variance in observed sensitivity is attributable to genetic factors in the young person – direct evidence of evocative gene-environment correlation. Bivariate genetic modelling showed that common genetic factors accounted for a significant proportion of the phenotypic association between attachment and caregiving sensitivity. We conclude that gene-environment correlation plays an important role in attachment dynamics in adolescence.

Prior research has generated non-trivial associations between secure base script knowledge and later indicators of psychosocial adjustment from early childhood through early adulthood (e.g., Dagan et al., 2024; Li et al., 2024; Posada et al., 2019; Ruiz et al., 2019). Yet, our understanding of these associations in the later life course is sparse. Further, much of the extant literature on secure base script knowledge has been conducted using research designs where genetic and environmental contributions to these associations are confounded. In this presentation, we take a lifespan and genetically-informed approach to understanding how secure base script knowledge is associated with psychosocial adjustment. We aim to estimate the relative contributions of genetic and environmental influences to the associations between adult attachment security, as indexed by secure-base script knowledge, and indicators of mental health and interpersonal functioning in late adulthood. We leverage a cohort of same-sex twins (N = 1342, Mage=70.38, SDage=5.42) from the Multi-Ethnic Twin Registry, drawn from participants born in Minnesota and North Carolina. Participants completed the Attachment Script Assessment (ASA) as well as self-report measures of mental health and interpersonal functioning (Short-Form 36 Health Survey, Inventory of Depressive and Anxiety Symptoms, UCLA Loneliness Scale, and Network of Relationships Inventory). We will use bivariate Cholesky models to determine to what extent the variation within each variable of interest, as well as the covariation between ASA scores and each measure of adjustment, are genetically and environmentally mediated. Initial analyses indicate that secure base script knowledge is modestly explained by genetics and substantially accounted for by non-shared environmental influences. Findings will be discussed with respect to how the secure base script functions in the later life course and the role of genetic and environmental influences in understanding the etiology of secure base script knowledge and its psychosocial correlates.