Assessing disorganized attachment representations: a systematic psychometric review and meta-analysis of the Manchester Child Attachment Story Task

The use and potential of artificial intelligence for supporting clinical observation of child behaviour

BACKGROUND

Observation of child behaviour provides valuable clinical information but often requires rigorous, tedious, repetitive and time expensive protocols. For this reason, tests requiring significant time for administration and rating are rarely used in clinical practice, however useful and effective they are. This article shows that Artificial Intelligence (AI), designed to capture and store the human ability to perform standardised tasks consistently, can alleviate this problem.

CASE STUDY

We demonstrate how an AI-powered version of the Manchester Child Attachment Story Task can identify, with over 80% concordance, children with insecure attachment aged between 5 and 9 years.

DISCUSSION

We discuss ethical issues to be considered if AI technology is to become a useful part of child mental health assessment and recommend practical next steps for the field.

Probing the content of affective semantic memory following caregiving-related early adversity

Cognitive science has demonstrated that we construct knowledge about the world by abstracting patterns from routinely encountered experiences and storing them as semantic memories. This preregistered study tested the hypothesis that caregiving-related early adversities (crEAs) shape affective semantic memories to reflect the content of those adverse interpersonal-affective experiences. We also tested the hypothesis that because affective semantic memories may continue to evolve in response to later-occurring positive experiences, child-perceived attachment security will inform their content. The sample comprised 160 children (ages 6-12 at Visit 1; 87F/73 M), 66% of whom experienced crEAs (n = 105). At Visit 1, crEA exposure prior to study enrollment was operationalized as parental-reports endorsing a history of crEAs (abuse/neglect, permanent/significant parent-child separation); while child-reports assessed concurrent attachment security. A false memory task was administered online ∼2.5 years later (Visit 2) to probe the content of affective semantic memories-specifically attachment schemas. Results showed that crEA exposure (vs. no exposure) was associated with a higher likelihood of falsely endorsing insecure (vs. secure) schema scenes. Attachment security moderated the association between crEA exposure and insecure schema-based false recognition. Findings suggest that interpersonal-affective semantic schemas include representations of parent-child interactions that may capture the quality of one's own attachment experiences and that these representations shape how children remember attachment-relevant narrative events. Findings are also consistent with the hypothesis that these affective semantic memories can be modified by later experiences. Moving forward, the approach taken in this study provides a means of operationalizing Bowlby's notion of internal working models within a cognitive neuroscience framework. RESEARCH HIGHLIGHTS: Affective semantic memories representing insecure schema knowledge (child needs + needs-not-met) may be more salient, elaborated, and persistent among youths exposed to early caregiving adversity. All youths, irrespective of early caregiving adversity exposure, may possess affective semantic memories that represent knowledge of secure schemas (child needs + needs-met). Establishing secure relationships with parents following early-occurring caregiving adversity may attenuate the expression of insecure semantic memories, suggesting potential malleability. Affective semantic memories include schema representations of parent-child interactions that may capture the quality of one's own attachment experiences and shape how youths remember attachment-relevant events.

Preliminary Danish Norms for the Odense Child Trauma Screening (OCTS)

The Odense Child Trauma Screening (OCTS) is Danish story stem screening tool applicable for assessment of preschoolers and young children in risk of being traumatized. Having shown initial evidence of validation, Danish norms are needed to strengthen the clinical assessment with the OCTS by serving as a baseline comparison for assessment of potentially traumatized children. We tested 169 children from the Danish general population aged 4-8 with the OCTS and investigated sex and age differences in play-based behavior and narrative representations. Caregivers reported electronically on child demographic information, psychosocial functioning, and history of trauma exposure using The Strengths and Difficulties Questionnaire (SDQ) and The Diagnostic Infant and Preschool Assessment (DIPA) trauma list. Across the 145 scores of the OCTS coding scheme, significant sex and age differences were only found in five and sixteen scores respectively. In the five codes where significant sex differences were observed, boys' normative scores were higher. No significant sex differences were found in the partial story scores or the OCTS total score. Three significant age differences in partial story and OCTS total scores emerged with 4-year-olds scoring higher than 6-8-year-olds. We further found 13 significant age differences in code scores with higher scores among the youngest of the two groups in question suggesting that scores tend to decrease along older age. Few significant sex and age differences were found in children's OCTS play-based behavior and narrative representations. Indicative of few sex and age biases of the OCTS coding scheme and stories, results suggest that the OCTS can be applied across the intended target group of children aged 4 to 8 years. As higher scores were found in the younger age groups, clinicians should be attentive to age in certain codes of the OCTS coding scheme in their assessment of children in clinical practice. The preliminary normative scores must be interpreted and clinically applied with caution due to our non-representative sample and lack of analyses on factors potentially influencing children's responses to the OCTS (e.g., developmental, contextual, cultural factors).

Supplementary Information

The online version contains supplementary material available at 10.1007/s40653-024-00616-7.

Can the family drawing be a useful tool for assessing attachment representations in children? A systematic review and meta-analysis

A systematic review and meta-analysis were conducted to evaluate the quality and validity of Family Drawings (FD) with an Attachment-Based Coding System in assessing attachment representations among pre-school and school-age children. A literature search in notable databases identified 645 records, of which 20 were eligible after screening and quality assessment. Results showed: 1) ABCD attachment distribution in community children was: 48% secure, 20% avoidant, 21% ambivalent, 11% disorganized. Security prevailed both in classifications and Fury et al.' scales. 2) No significant differences according to the cultural background; 3) At-risk/clinical children showed higher insecurity than community ones using scales; 4) Girls were more secure than boys. In conclusion, FD may be a culture-fair method to assess attachment representations in children. Global scales seem more reliable than ABCD classifications for discriminating at-risk and clinical children, but further studies on these groups are needed.

Attachment representations among school-age children with intellectual disability

BACKGROUND

Research data documenting a high risk of insecure and disorganized attachment among children with intellectual disability (ID) in infancy and early childhood raises the question of mutual influences between ID and attachment in later childhood.

AIMS

The objectives of the present study were to examine attachment among school-age children with ID and whether attachment varies according to level of intellectual functioning, adaptative functioning, and presence of a genetic syndrome (i.e. Down syndrome).

METHODS

Attachment among 54 children with ID aged 8-12 years (30 with Down Syndrome, 24 with non-specific ID) was assessed using the Attachment Story Completion Task, and compared with that of 108 typically developing children, 54 of the same chronological age and 54 of the same mental age.

OUTCOMES

Results show (1) less security among children with ID than among same-age controls (2) more disorganization among children with ID compared to the two control groups, (3) a link between attachment disorganization and level of adaptive functioning among children with ID and (4) no difference in attachment between children with DS and children with non-specific ID.

CONCLUSIONS

Children with ID remain vulnerable to disorganization during late childhood. More research is needed to understand the factors underlying disorganized attachment representations.

The School Attachment Monitor-A novel computational tool for assessment of attachment in middle childhood

Background

Attachment research has been limited by the lack of quick and easy measures. We report development and validation of the School Attachment Monitor (SAM), a novel measure for largescale assessment of attachment in children aged 5–9, in the general population. SAM offers automatic presentation, on computer, of story-stems based on the Manchester Child Attachment Story Task (MCAST), without the need for trained administrators. SAM is delivered by novel software which interacts with child participants, starting with warm-up activities to familiarise them with the task. Children’s story completion is video recorded and augmented by ‘smart dolls’ that the child can hold and manipulate, with movement sensors for data collection. The design of SAM was informed by children of users’ age range to establish their task understanding and incorporate their innovative ideas for improving SAM software.

Methods

130 5–9 year old children were recruited from mainstream primary schools. In Phase 1, sixty-one children completed both SAM and MCAST. Inter-rater reliability and rating concordance was compared between SAM and MCAST. In Phase 2, a further 44 children completed SAM complete and, including those children completing SAM in Phase 1 (total n = 105), a machine learning algorithm was developed using a “majority vote” procedure where, for each child, 500 non-overlapping video frames contribute to the decision.

Results

Using manual rating, SAM-MCAST concordance was excellent (89% secure versus insecure; 97% organised versus disorganised; 86% four-way). Comparison of human ratings of SAM versus the machine learning algorithm showed over 80% concordance.

Conclusions

We have developed a new tool for measuring attachment at the population level, which has good reliability compared to a validated attachment measure and has the potential for automatic rating–opening the door to measurement of attachment in large populations.

Can Psychodynamically Oriented Early Prevention for "Children-at-Risk" in Urban Areas With High Social Problem Density Strengthen Their Developmental Potential? A Cluster Randomized Trial of Two Kindergarten-Based Prevention Programs

Children who live on the margins of society are disadvantaged in achieving their developmental potential because of the lack of a necessary stable environment and nurturing care. Many early prevention programs aim at mitigating such effects, but often the evaluation of their long-term effect is missing. The aim of the study presented here was to evaluate such long-term effects in two prevention programs for children-at-risk growing up in deprived social environments focusing on child attachment representation as the primary outcome as well as on self-reflective capacities of teachers taking care of these children. The latter was a key component for promoting resilient behavior in children. Five hundred and twenty-six children aged 36 to 60 months at risk due to immigration status, low family socio-economic status and child behavior were examined in a cluster-randomized study comparing two preventions, the psychodynamic, attachment-based holistic approach EARLY STEPS (ES) with the classroom based FAUSTLOS (FA) for their efficacy. Primary outcome was the child attachment representation measured by the Manchester Child Attachment Story Task (MCAST). Secondary outcomes were derived from (a) the Caregiver-Teacher Report Form (C-TRF: problem behaviors, including anxiety/depressive symptoms, emotional-reactive and somatic problems, social withdrawal, aggressive behavior, and attention deficit), from (b) the Strength and Difficulties Questionnaire (SDQ, parent version: resilience and wellbeing) and (c) Self-Reflective Scales for teachers (SRS: self-reflective capacities of teachers). Compared to baseline, attachment and behavioral problems improved in both programs. ES led to more secure and more organized attachment representations (medium effect sizes). Aggressive behavior and externalizing problems were reduced in the FA group compared with ES, particularly in boys (medium effect sizes). Self-reflective capacities of the teachers increased only in the ES group. High correlation between children's attachment type with the number of social risk factors and the increase of problematic social behavior strongly indicate that an increase in teachers' self-reflective capacities helps to change children's attachment patterns which thus strengthens the resilience of these children-at-risk [An ethical vote from LPPKJP 2009-02-25 was obtained and the trial registered; Clinical trial registration information: The trial was registered 14.02.2012 (DRKS00003500; https://www.drks.de)].

"Surprise" and the Bayesian Brain: Implications for Psychotherapy Theory and Practice

The free energy principle (FEP) has gained widespread interest and growing acceptance as a new paradigm of brain function, but has had little impact on the theory and practice of psychotherapy. The aim of this paper is to redress this. Brains rely on Bayesian inference during which “bottom-up” sensations are matched with “top-down” predictions. Discrepancies result in “prediction error.” The brain abhors informational “surprise,” which is minimized by (1) action enhancing the statistical likelihood of sensory samples, (2) revising inferences in the light of experience, updating “priors” to reality-aligned “posteriors,” and (3) optimizing the complexity of our generative models of a capricious world. In all three, free energy is converted to bound energy. In psychopathology energy either remains unbound, as in trauma and inhibition of agency, or manifests restricted, anachronistic “top-down” narratives. Psychotherapy fosters client agency, linguistic and practical. Temporary uncoupling bottom-up from top-down automatism and fostering scrutinized simulations sets a number of salutary processes in train. Mentalising enriches Bayesian inference, enabling experience and feeling states to be “metabolized” and assimilated. “Free association” enhances more inclusive sensory sampling, while dream analysis foregrounds salient emotional themes as “attractors.” FEP parallels with psychoanalytic theory are outlined, including Freud’s unpublished project, Bion’s “contact barrier” concept, the Fonagy/Target model of sexuality, Laplanche’s therapist as “enigmatic signifier,” and the role of projective identification. The therapy stimulates patients to become aware of and revise the priors’ they bring to interpersonal experience. In the therapeutic “duet for one,” the energy binding skills and non-partisan stance of the analyst help sufferers face trauma without being overwhelmed by psychic entropy. Overall, the FEP provides a sound theoretical basis for psychotherapy practice, training, and research.