The Neural Substrates of Attachment in Children and Parents

The Neural Substrates of Attachment in Children and Parents

General Introduction

While the literature on the neural basis of attachment in adults is steadily growing and first findings on the same mechanisms in adolescents are starting to emerge, it remains largely unknown how individual differences in attachment influence brain activation in parents and children. Although nice research by Lane Strathearn and colleagues (see here and here) showed that mothers’ brain responses to images of own (versus unknown) children are related to individual differences in maternal attachment insecurity – particularly attachment avoidance –, the literature on the neural substrates of attachment in parents, and especially fathers, is very scarce. Furthermore, there is no literature available yet on how an avoidant and/or anxious attachment orientation may influence brain responses in children themselves.

To close this gap, Dr Pascal Vrticka has started a new line of research in collaboration with the CIBSR at Stanford University in children aged 8 to 12 years, and with the University of Vienna and MPI CBS Leipzig in parents – both fathers and mothers – of children aged 5-6 years (see CARE Studies for the latter). Experiments mainly focus on fMRI in single participants (children; fathers and mothers) and attachment in all of them is measured using validated age-appropriate self-reports and/or semi-structured narrative interview techniques.


The Neural Substrates of Attachment in Children

In a first study conducted some years ago at Stanford University, 8-12 year-old children performed two social emotional tasks in the fMRI scanner: 1) a social feedback processing task (already previously employed in adults and adolescents), and 2) a newly designed own versus mother face morph task. Attachment in children to their parents was measured with the child version of the self-report questionnaire Experiences in Close Relationships revised (ECR-RC). 

Adapted from Miller et al. (2020)


Our results revealed no associations between brain activity and individual differences in attachment for the social feedback processing task. However, we observed a relationship between child attachment anxiety and brain activity during the own versus mother face morph task in the left anterior temporal pole (ATP). While activity was strongest for the (unmorphed) own and mother face in infants scoring low in attachment anxiety, it was strongest for the morphed faces in infants scoring high in attachment anxiety. In other words, greater attachment anxiety was related to a shift from heightened processing of self and mother faces to morphed faces. These findings suggest that some children may exhibit too much neural sensitivity to overlapping self–mother representations, and are interesting from the perspective that attachment security is related to the development of an autonomous self.

The corresponding paper is now freely available: Miller, J., Shrestha, S., Reiss, A. L., Vrtička, P. (2020). Neural Bases of Social Feedback Processing and Self-Other Distinction in Late Childhood: The Role of Attachment and Age. Cognitive, Affective, and Behavioral Neuroscience, Volume 20, Pages 503-520. https://doi.org/10.3758/s13415-020-00781-wOPEN ACCESS.


The Neural Substrates of Attachment and Caregiving in Parents

Parents (both mothers and fathers) from a different study are also performing two social emotional tasks in the fMRI scanner, comprising an emotional facial expression processing task from own versus unknown children, as well as a virtual ball-tossing game with the own versus an unknown child. Attachment in adults is measured with the self-report questionnaire Experiences in Close Relationships revised (ECR-R) as well as the Adult Attachment Interview (AAI). Video-ratings of a parent-child interaction task (during fNIRS hyperscanning – see here) and parent self-reports pertaining to parent-child relationship quality and caregiving are also being collected. These data are acquired in association with the D-CARE and M-CARE projects (see here).

More information will be made available soon.