The Assessment of Fine Motor Skill Development through Sketch Features and Their Connection to Spatial Cognition Skills to Support School and STEM Learning Readiness

Abstract

The American Academy of Pediatrics says that 1 in 6 children has some form of developmental delay or disability but most of these delays are not identified or treated until after starting school. During their formative years before school, young children are developing sensory and motor experiences and fine motor skills that are a critical part of healthy development. The impact of these skills goes beyond the physical; for instance, children’s fine motor skills are linked not only to drawing ability but also to cognitive, social-emotional, self-regulatory, and academic development. As such, it is important to evaluate motor development from an early age, but assessments from teachers and parents can be affected by inexperience or bias. Direct assessments with pediatricians are preferred. However, these can be expensive and time-consuming, and pediatricians may be overburdened with many patients. Since these delays can impact later development, school readiness, and critical skills like reading and math that are linked to STEM achievement, we seek a technological solution to equip parents and teachers with better, unbiased screening tools that may be able to identify developmental delay early, lessening the burden on pediatricians and ensuring fewer children slip through the cracks by finding those most in need of help.

In this work, we discuss a study conducted with 70 preschoolers centered on drawing activities inspired by developmental assessments. We utilize concepts of sketch recognition to analyze these drawings for motor- and age-related features. Based on these findings, we construct machine learning models to differentiate children’s ages and fine motor levels. We present a hybrid model based on assessing circle, square, and triangle shapes that has the potential to digitize fine motor integration tasks found in existing assessments, at least as an assistive tool. We consider additional factors such as pencil grip and technology experience, as well as implications relating to school-readiness and STEM learning. In the final sections, we explore broader impacts of this work and connect from child-centered computer interactions to more general human-computer interaction, emphasizing the critical role of universal and accessible design for building technological solutions for the benefit of all.