Site navigation

How can Machine Learning Measure Attachment in Children?

Ross Kelly


Machine learning
Researchers found the technology was able to accurately measure a child’s attachment levels.

Machine learning has been used to provide a deeper insight into attachment in children, a lasting bond that infants first develop with their caregivers.

As part of a study led by the University of Glasgow, researchers have developed a fast and efficient way of measuring attachment through a computer game.

Long-term, the research team believes the project findings have the potential to be used in public health monitoring.

Attachment between a young child and their primary caregiver is regarded as vitally important for emotional development. Disruption of this bond can have adverse emotional and psychological effects on children.

Measuring attachment, however, can be a time consuming and complex process. This, researchers say, is due to the fact that attachment behaviours can only be observed and assessed by trained professionals.

To streamline the process, researchers worked alongside child mental health experts to develop the School Attachment Monitor, or ‘SAM’.

Professor Helen Minnis, professor of child and adolescent psychiatry at the University of Glasgow and lead author of the study, said: “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 large-scale public health monitoring or epidemiology.”

The computer programme uses machine learning and smart sensors to accurately assess attachment in children.

According to the research team, SAM interacts in real-time with child participants – initially 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 while data on their attachment patterns are captured through video recording and movement sensors in the smart dolls.

Researchers found that SAM was able to accurately measure a child’s attachment when compared with the ratings of trained professionals examining the same child’s data.


Professor Minnis commented: “Our study shows that by 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.”

Although SAM is an exciting prospect for public health monitoring, Professor Minnis said that further research will be required to explore the system’s performance “in a range of populations”.

The study, ‘The School Attachment Monitor – a novel computational tool for assessment of attachment in middle childhood’ is published in PLOS ONE. The work was funded by the Engineering and Physical Sciences Research Council (EPSRC).

Ross Kelly

Staff Writer

Latest News

Cybersecurity Finance
Cybersecurity Editor's Picks
Climate Editor's Picks Energy Featured
%d bloggers like this: