Scholarlink

<p>As Generative Artificial Intelligence (GenAI) becomes increasingly integrated into educational environments, most implementations remain cognitively adaptive- personalizing content based on performance or preferences- while overlooking the vital role of learner emotion. This paper introduces a novel dual-adaptive framework for Emotion-Aware GenAI Tutors (EAGAITs) that combines real-time cognitive modeling with affective sensing to deliver personalized, emotionally intelligent instruction. Drawing on interdisciplinary foundations from educational psychology, affective computing, and AI design, the framework proposes a system that detects learner emotions through multimodal inputs (for example, sentiment, facial expressions, and behavioral patterns) and utilizes Large Language Models (LLMs) to adapt both the complexity and tone of feedback. Through real-world use cases across K-12, higher education, and MOOCs, this paper illustrates how EAGAITs enhance engagement, persistence, and learning outcomes. The proposed framework aims to inspire the development of human-centered AI systems that understand what learners know and how they feel- bridging cognition and emotion in the next generation of personalized education.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Building emotionally intelligent generative AI tutors for education

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年2月5日

查看全文
  • Pradeep Kumar Dadabada

摘要

As Generative Artificial Intelligence (GenAI) becomes increasingly integrated into educational environments, most implementations remain cognitively adaptive- personalizing content based on performance or preferences- while overlooking the vital role of learner emotion. This paper introduces a novel dual-adaptive framework for Emotion-Aware GenAI Tutors (EAGAITs) that combines real-time cognitive modeling with affective sensing to deliver personalized, emotionally intelligent instruction. Drawing on interdisciplinary foundations from educational psychology, affective computing, and AI design, the framework proposes a system that detects learner emotions through multimodal inputs (for example, sentiment, facial expressions, and behavioral patterns) and utilizes Large Language Models (LLMs) to adapt both the complexity and tone of feedback. Through real-world use cases across K-12, higher education, and MOOCs, this paper illustrates how EAGAITs enhance engagement, persistence, and learning outcomes. The proposed framework aims to inspire the development of human-centered AI systems that understand what learners know and how they feel- bridging cognition and emotion in the next generation of personalized education.