Neural correlates of face familiarity in institutionalised children and links to attachment disordered behaviour

Watch this video abstract from Paula Oliveira on her JCPP paper ‘Neural correlates of face familiarity in institutionalised children and links to attachment disordered behaviour’.

Authors; Paula S. Oliveira, Pasco Fearon, Jay Belsky, Ana R. Mesquita, Adriana Sampaio, Diego Pinal, Isabel Soares

First published: 01 December 2022

Open Access paper

Paula S. Oliveira
Paula S. Oliveira

I am a Senior Research Fellow at the Anna Freud National Centre for Children and Families, and most of my research and teaching is done in partnership with University College London. I work across several different research projects, with a focus on attachment difficulties, including attachment disorders, and trauma in children who are or have been looked after or are at risk of entering care. In my research I use different methods ranging from observations and interviews to event-related potentials (EEG). More recently I became particularly interested in the development and evaluation of intervention programmes for children with experiences of care. I also work as a psychologist in a clinical service offering an attachment-based parenting intervention to adoptive, kinship and foster parents.

Other resources

  • Featured paper ‘Neural correlates of face familiarity in institutionalised children and links to attachment disordered behaviour’


[00:00:18.060] Paula Oliveira: This is an abstract video about a paper that my colleagues and I published in the Journal of Child Psychology and Psychiatry in late 2022, with the title “Neural Correlates of Face Familiarity in Institutionalised Children and Links to Attachment Disordered Behaviour.”  Being looked after in institutional or residential care in early life is known to compromise multiple aspects of children’s development.  This includes attachment difficulties and atypical patterns of relating to caregivers and unfamiliar adults that are captured by the labels of reactive attachment disorder and disinhibited social engagement disorder.

Children that present reactive attachment disorder symptoms show marked difficulties in seeking or responding to comfort from caregivers when hurt or distressed.  They can be very withdrawn and emotionally dysregulated.  On the other hand, children presenting disinhibited social engagement disorder show a pattern of indiscriminate friendliness, with little reticence in approaching and interacting with unfamiliar adults and venturing away in unfamiliar settings.

There is a wealth of evidence associating these two patterns of attachment and social problems with social deprivation in early caregiving experiences, particularly being institutionalised or being exposed to repeated changes in caregivers.  But we don’t know much about how these adverse experiences of where the caregiving neglect become biologically embedded and how they relate to neural processes.

In this study, our participants were preschool aged children living in institutional care in Western Europe, in Portugal, and a control group of same aged children continuously raised by their biological families, in Portugal, as well.  We use event-related potentials, or ERPs, to look at children’s neural processing, while viewing the face of their main caregiver and the face of a female stranger.  After cleaning and pre-processing the data, the resulting sample was 100 children who were, on average, four and a half years old.

In this study, we sought to replicate previous findings of the general effects of institutionalisation on the neural processing of faces in settings that were quite different, in important ways, from institutions of previous research.  And we also aimed to investigate the role of variation in the institutional experience and variation in attachment disorder symptoms in children’s neural processing of faces.  And we focused on posterior face sensitive ERP components, namely the P1, the N170 and the P400.

Our analysis plan had two parts.  In the first part, in order to investigate whether neural correlates of face familiarity differ as a function of rearing experiences among institutionalised children, we compared them to a group of family reared controls.  And in the second part we assessed, just among the institutionalised children, whether variation in social or attachment disorder behaviour, so reactive attachment disorder or disinhibited social engagement disorder symptoms, as well as variation in institutional caregiving experiences, predicted face processing patterns.

Moving onto results.  First, we then compared the two groups, the institutionalised group and the comparison home reared group.  We found group differences in the N170 component, where the institutional group showed smaller amplitudes in this component, compared to the control group.  And this group effect held even when taking chronological age and developmental age into account, which suggests that it was not just a product of group differences in developmental level.

Results for the other two components we looked at were not as robust, so I’m not discussing them in this video.  But our finding of a blunted N170 in the currently institutionalised Portuguese children provides, for the first time, evidence of differences in face sensitive perceptual processes in children who enter relatively higher quality institutions later in development and for largely different reasons than those in most previous research.

We also noted face familiarity effects, and we found that institutionalised children showed the same ability that controls did in discriminating their caregiver from the stranger’s face in the 1 – N170 amplitudes.  Specifically, the N170 was larger for the caregivers than the stranger’s face.  Nevertheless, other more nuanced face discrimination effects were seen in the controls that were not seen in the institutionalised group.

Then we looked at the institutionalised group within effects – within group effects.  And first, looking at attachment disorder symptoms, predicting neural responses, we found that RAD symptoms did not predict any ERP components.  But we did find that higher levels of disinhibited social engagement disorder symptoms predicted smaller P1 amplitudes, irrespective of face, which suggests that it could also reflect a more general difficulty with attention.  And higher disinhibited social engagement disorder symptoms also predicted smaller differences in the P400 amplitude to each face and the smaller P400 amplitude in a response to the stranger’s face, specifically.

This result with the P400, which is a component implicated in face processing, as well as novelty detection, suggests that this component is indexing some face sensitive alteration in neural processing that is relevant for our understanding of the indiscriminately social behaviour pattern.  These findings are also in line with our prediction of a link between disinhibited social engagement disorder and alterations in the underpinnings of face familiarity processing and are consistent with previous EEG and fMRI findings.

Now, looking at variables of the experience of institutional care.  We found that age at admission to the institution and length of time institutionalised did not contribute to explain children’s ERPs.  In this study, it was not possible to explore the existence of care desensitive periods because of the age at which children enter the institution, but these findings add to evidence from other studies with post-institutionalised children that there doesn’t seem to be a linear dose effect of institutional aggravation on a child’s development.

In contrast, variation in individualised care in our study did predict children’s neural responses in this study.  Specifically, living in a setting with worse ratios of children to caregivers, that is more children per adult, predicted smaller P400 amplitudes to either face stimuli.  This finding is important, because it provides additional evidence to what we already know, that institutional deprivation, resulting from lack of one-to-one interactions, is a key factor contributing to children’s poor socioemotional functioning, by limiting opportunities for individualised attention and consistent relatedness from a stable caregiver.

Conclusion: In this study, we wanted to contribute to identifying neural mechanisms involved in children’s socioemotional difficulties associated with early caregiving adversity, specifically being raised in institutional care.  Our findings add to the literature showing that social neglect, particularly reduced individualised care, has a deleterious impact on children’s neural processing of faces.

The current study also demonstrated for the first time that a pattern of reduced ERP amplitudes and reduced neural discrimination between faces is associated with disinhibited social engagement disorder in institutionalised children, but not reactive attachment disorder symptoms.

Thank you for listening, and do take a look at our paper if you’d like to find out more about this study.

Add a comment

Your email address will not be published. Required fields are marked *