Karunaratna, SulaknaVargas-Diaz, DanielKim, JisunWang, JennyChoi, KoeunLee, Sang Won2024-12-032024-12-032024-11-11https://hdl.handle.net/10919/123723In recent years, the number of AI voice agent applications designed to help young children learn math has increased. However, the impact of interactivity within these applications on children’s learning and engagement remains unexplored. While current apps may employ various levels of interactions, such as visual, haptic, sound, and animation, the efficacy of these interactions in facilitating children’s learning remains uncertain. This research investigates how varying levels of interactivity in touch-based interfaces, combined with an AI voice agent, affect the learning of counting skills in children aged 2 to 4 years.We examine three conditions: baseline (no demonstration), animated demonstration, and interactive demonstration. By examining how these different levels of interactivity influence children’s engagement with math apps, this study seeks to enhance our understanding of effective design strategies for educational technology targeting early childhood education. The findings of this research hold the potential to inform the development of interfaces for math games that leverage both touch-based interactions and AI voice assistants to support young children’s learning of foundational mathematical concepts.application/pdfenCreative Commons Attribution 4.0 InternationalArticle - Refereed2024-12-01The author(s)https://doi.org/10.1145/3678884.3681886