The Key Artificial Intelligence Technologies in Early Childhood Education: A Review
Yi Honghu, Liu Ting, Lan Gongjin
TL;DR
The paper surveys the current landscape of AI in early childhood education, with a focus on AI-based robotic agents and interventions for children with autism spectrum disorder. It combines a qualitative review of representative robots (e.g., NAO, KASPAR, Keepon, Paro, Probo) with a comprehensive bibliometric analysis (via Scopus) to map publication trends, venues, and key contributors from 2010 to 2022. It identifies core challenges—data scarcity and quality, explainability, and educator readiness—while outlining trends toward integrating state-of-the-art AI (including large language models) and VR/vision/NLP tools to enhance learning and social development. The work provides a structured taxonomy, actionable insights, and recommendations to guide future research and practice in AI-enabled early education, serving both newcomers and experienced researchers.
Abstract
Artificial Intelligence (AI) technologies have been applied in various domains, including early childhood education (ECE). Integration of AI educational technology is a recent significant trend in ECE. Currently, there are more and more studies of AI in ECE. To date, there is a lack of survey articles that discuss the studies of AI in ECE. In this paper, we provide an up-to-date and in-depth overview of the key AI technologies in ECE that provides a historical perspective, summarizes the representative works, outlines open questions, discusses the trends and challenges through a detailed bibliometric analysis, and provides insightful recommendations for future research. We mainly discuss the studies that apply AI-based robots and AI technologies to ECE, including improving the social interaction of children with an autism spectrum disorder. This paper significantly contributes to provide an up-to-date and in-depth survey that is suitable as introductory material for beginners to AI in ECE, as well as supplementary material for advanced users.
