Children’s television producers in the UK are building an artificial intelligence tool to predict how video clips affect the developing brains of three- to six-year-olds. The problem is that no one knows what “age-appropriate” content actually looks like for young children. UK law now requires children’s media to comply with the Age-Appropriate Design Code, but the guidelines are vague. Developmental scientists understand children’s cognitive abilities in the lab, and media creators have strong intuitions about pacing and style, but neither side has tested those intuitions against real brain development. This project bridges that gap. If the tool works, it could give animators and producers a concrete way to check whether their content supports or disrupts the executive functions—self-control and learning skills—that mature rapidly between ages three and six. Parents and children could also use it to choose media that fits a child’s developmental stage. Beyond the immediate industry, the project aims to establish a UK-wide network linking neuroscience, AI, and children’s animation, creating new methods for studying how screen media shapes young minds.
View original technical description
Modern childhood is mediated by screens. Children learn, play, and socialize through digital media, and fears of negative impacts on development during a critical period of neuroplasticity has led early-years agencies (e.g. UK Chief Medical Officers, American Academy of Pediatrics, WHO) to recommend limits on children's screen time and restrict it to 'high quality' content. As the UK children's media industry now faces the legal threat of not complying with the Age-Appropriate Design Code (i.e. the Children's Code; part of UK GDPR law), creatives are challenged with figuring out how to ensure their content is appropriate for the developmental stage of their users without clear guidance on how to do so. This is a provocative problem which cannot be solved by a single discipline. Developmental scientists can advise on the neurocognitive capacities of children, but little is known about how these skills apply to media. Media theorists understand the practice of children's TV creation, but these theories are not empirically validated. Ironically, the experts who have the greatest ability to gauge media's "age appropriateness" are the creatives who are now subject to the Children's Code. The UK children's media industry has been at the pioneering edge of international children's TV and animation for decades. Child- and adult-directed content differs radically in style (e.g. animation vs. live-action, slow vs. fast pace) and embedded learning (e.g. phonics, numeracy, diversity), but the impact of these approaches is not empirically validated. A route to understanding "age appropriateness" may be in working with creatives to formalise, test and concretize their intuitions in the form of objective computational tools. These tools could be used to check whether content is developmentally appropriate for their target viewers, for parents and children to make informed content selection, and the creation of new forms of tailored media experiences that nurture child development. This is the vision of the Animating Minds project. Animating Minds brings together a unique team of interdisciplinary researchers from children's animation practice, media theory, developmental psychology, neuroscience, and artificial intelligence (AI). Our aim is to build an AI tool that can predict a video clip's potential impact on children's neurocognition, specifically their executive functions, across different developmental stages - 3 to 6 years, a key period of neurocognitive development and a transitional age in terms of TV programming. Executive functions (EF) are cognitive skills essential for self-control and learning that mature across this age-span and have been shown to be impacted by screen media. We will accomplish the overall aim through four objectives: O1) Formalise creative insights from children's media creators about how they tailor their animated content to particular age groups; O2) Develop a machine learning (ML) classifier of children's animation age ratings; O3) Extend the ML model to classify behavioural EF ratings of animations across a diverse sample of 3- and 6-year-olds; O4) Validate creative intuitions and ML model predictions by comparing the impact of manipulated clips on 3- and 6-year-olds neurocognitive EF markers. Accomplishing these objectives will require transdisciplinary collaboration not usually seen in within-council projects. The interwoven nature of the objectives and the team has the potential to establish new methods, theories, and computational tools which will yield benefits across disciplines. The Animating Minds project will establish a UK-wide network of expertise in applied neurocognitive development fostering radically new creativity in children's media.
Plain English summaries and category classifications on this site are generated by AI and may not perfectly reflect the original research.
Is something wrong? Let us know