Development and change in attachment: A multiwave assessment of attachment and its correlates across childhood and adolescence.

This research examines the contextual factors that facilitate development and change in attachment during later childhood, adolescence, and early adulthood using a longitudinal cohort design involving 690 children (7–19 years old) and their parents. At each wave, a variety of interpersonal variables (e.g., parent–child stress) were measured. We examined alternative developmental processes (i.e., long-term, catalytic, and short-term processes) that have not been previously distinguished in attachment research. Preregistered analyses revealed that nondevelopmental processes can explain the associations between almost all of the interpersonal variables of interest and attachment security, suggesting that previous research using traditional longitudinal methods may have misattributed nondevelopmental processes for developmental ones. For example, we found that friendship quality, although prospectively associated with attachment both in prior work and in the current study, was not developmentally associated with attachment. However, after controlling for nondevelopmental sources of covariation, we identified a number of developmental processes that may help explain change in attachment. For example, we found that initial levels of parental depression, as well as growth in parent–child stress, were related to growth in adolescent insecurity over 3 years. We also examined 12 genetic variants studied in previous research and found that they were not related to average levels or changes in attachment. These results highlight how distinguishing unique kinds of developmental processes allows for a more comprehensive understanding of attachment.