Summary: Artificial intelligence technology was able to accurately predict attachment in young children.
Source: University of Glasgow
For the first time, researchers have used machine learning to successfully measure attachment in children – the vital human bond that humans first develop as infants to their caregivers.
In new multi-disciplinary research, led by the University of Glasgow and published in PLOS ONE, the study team present a quick and easy way to measure attachment through a computer game, that has the potential to be used in largescale public health monitoring.
Attachment is a term used to describe the lasting emotional bond and feeling of connectedness between human beings. The attachment between a young child and their primary caregiver is known to be vitally important for emotional development – disruption to or loss of this bond can affect a child emotionally and psychologically into adulthood, impacting future relationships.
Insecure attachment in children is not necessarily abnormal, and is often an emotional adaptation to less than optimal environmental circumstances. However, insecure attachment is associated with increased risk of psychopathology of various types. As a result, there may be potential for SAM, in the future, to form part of an early school-based screening programme to identify children at risk of psychopathology.
Assessing attachment can be time consuming and complex, as attachment behaviours can, at present, only be observed and assessed by trained professionals when a child feels they are in a stressful situation.
Now, a research collaboration team of child mental health and computing science experts has developed and validated the School Attachment Monitor (SAM) – a computer programme that uses machine learning and smart sensors to accurately assess attachment in children. SAM is delivered by novel software which interacts with child participants, starting with warm-up activities to familiarise them with the task. Children are invited to play with ‘smart dolls’ while interacting with a story on the computer, and data on their attachment patterns are captured through video recording and movement sensors in the smart dolls.
Researchers found that the SAM technology was able to accurately measure a child’s attachment, when compared with the ratings of trained professionals looking at the same child’s data.
Professor Helen Minnis, professor of child and adolescent psychiatry at the University of Glasgow and lead author of the study, said: “Our study shows that using modern sensors and machine learning technology, it has been possible to develop the School Attachment Monitor (SAM) that works well with young children, and most importantly, gives an accurate classification of attachment security versus insecurity compared to manual ratings.
“The aim of our research programme, from its inception over a decade ago, was to develop a quick and easy measure of attachment that can be used in largescale public health monitoring or epidemiology. SAM has the potential to achieve that aim, although further research will be needed to examine its performance in a range of populations.”
SAM offers automatic presentation, on computer, of story-stems based on the Manchester Child Attachment Story Task (MCAST). The study recruited 130 5-9 year old children from mainstream primary schools to participate.
The study, ‘The School Attachment Monitor – a novel computational tool for assessment of attachment in middle childhood’ is published in PLOS ONE.
Funding: The work was funded by the Engineering and Physical Sciences Research Council (EPSRC).
About this AI research news
Source: University of Glasgow Contact: Ali Howard – University of Glasgow Image: The image is credited to University of Glasgow
The School Attachment Monitor – a novel computational tool for assessment of attachment in middle childhood
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.
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.
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.
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.