eaSEL: Promoting Social-Emotional Learning and Parent-Child Interaction through AI-Mediated Content Consumption
Project Overview
The document explores the integration of generative AI in education, specifically how it enhances social-emotional learning (SEL) for children aged 5-8 through a system called eaSEL. This innovative approach combines AI-generated content with SEL curricula, allowing children to engage with video content independently while receiving tailored activities and conversation prompts for parents. The findings reveal that eaSEL effectively fosters emotional reflection in children and promotes meaningful discussions between parents and their children, thereby supporting the development of critical social-emotional skills. By leveraging AI's capabilities, the system not only enriches children's learning experiences but also strengthens parent-child interactions, highlighting the potential of generative AI to positively impact educational outcomes in the realm of social-emotional development.
Key Applications
AI-mediated social-emotional learning activities
Context: Used in parent-child interactions to facilitate discussions about emotions and social skills, specifically during in-home media consumption by children aged 5-8 and their parents. This includes identifying SEL moments in children's videos and generating related activities and conversation starters.
Implementation: A prototype system that utilizes generative AI to detect social-emotional learning moments in children's videos. It generates prompts and conversation starters for children and parents to reflect on and discuss these moments, enhancing engagement and understanding.
Outcomes: ["Increased use of emotional language in children's retellings of videos", 'Enhanced parent-child engagement around SEL themes', "Improved children's understanding of social-emotional skills", 'Encouraged active participation from parents']
Challenges: ['Language comprehension issues for children', 'Potential for inappropriate content in source videos', 'Varying parental knowledge of SEL', 'Ensuring age-appropriate language and content', 'Parents may find it difficult to engage consistently']
Implementation Barriers
Content Appropriateness
Some generated activities may relate to themes in videos that parents find inappropriate for children.
Proposed Solutions: Implement content moderation or filtering mechanisms to ensure generated activities align with child-appropriate content.
Language Complexity
Generated prompts may include vocabulary that is too advanced for the target age group (5-8 years old).
Proposed Solutions: Develop child-centric language models trained on age-appropriate content to ensure suitable language use.
Parental Involvement
Parents may not have sufficient knowledge about SEL to engage effectively with their children using the system, and may not consistently engage with the AI-generated content or discussions.
Proposed Solutions: Provide parents with explicit SEL information and guidance through the system to enhance their engagement, and encourage regular interaction through reminders or gamified elements to enhance participation.
Technical Barrier
The complexity of integrating AI tools into existing educational frameworks.
Proposed Solutions: Providing user-friendly interfaces and clear guidelines for implementation.
Project Team
Jocelyn Shen
Researcher
Jennifer King Chen
Researcher
Leah Findlater
Researcher
Griffin Dietz Smith
Researcher
Contact Information
For information about the paper, please contact the authors.
Authors: Jocelyn Shen, Jennifer King Chen, Leah Findlater, Griffin Dietz Smith
Source Publication: View Original PaperLink opens in a new window
Project Contact: Dr. Jianhua Yang
LLM Model Version: gpt-4o-mini-2024-07-18
Analysis Provider: Openai