Artificial intelligence enhances children's science learning from television shows.

Ying Xu,Kunlei He,Julian Levine,Daniel Ritchie,Zexuan Pan,Andres Bustamante,Mark Warschauer
DOI: https://doi.org/10.1037/edu0000889
2024-09-28
Journal of Educational Psychology
Abstract:Educational television programs are important learning resources for young children, especially those from underresourced households. These programs' potential can be amplified if children are given the opportunity to meaningfully interact with media characters during their video watching. In this project, we partnered with PBS KIDS to develop interactive science-focused videos in which the main character, powered by artificial intelligence, engaged in dialogic interactions with children by asking them questions and providing responsive feedback. The children who watched the interactive videos performed better on a science posttest than children who watched the broadcast version of the video (without any interaction) or a pseudointeractive version (in which the media character asks children the same questions and gives generic feedback after a fixed amount of time). The artificial intelligence character's responsiveness positively influenced both the quantity and quality of children's verbalizations during video watching, compared to the pseudointeractive condition. This article sheds light on the feasibility and effectiveness of using conversational technologies to support active learning in children through educational programs. (PsycInfo Database Record (c) 2024 APA, all rights reserved)
psychology, educational
What problem does this paper attempt to address?