Personalizing explanations of AI-driven hints to users' cognitive abilities: an empirical evaluation
Project Overview
The document examines the role of generative AI in education, focusing on a study that explores the personalization of explanations provided by an Intelligent Tutoring System (ITS) to support students, particularly those with low Need for Cognition and Conscientiousness. The findings indicate that personalized explanations effectively enhance user engagement, improve comprehension of instructional hints, and result in notable learning gains. However, there are mixed reactions from users; while many benefited from the tailored support, some found the explanations to be distracting or intrusive, highlighting the need for a balanced approach to personalization in educational AI applications. Overall, the study underscores the potential of generative AI to tailor educational experiences but also points to the importance of user perception in the effectiveness of such technologies.
Key Applications
ACSP (Adaptive CSP) applet
Context: Intelligent Tutoring System for teaching constraint satisfaction problems to university students.
Implementation: Personalized explanations were delivered alongside hints, with features to increase access and time spent on explanations.
Outcomes: Increased interaction with explanations, improved understanding of hints, and higher learning gains.
Challenges: Some users found explanations too verbose and the interface could be perceived as intrusive.
Implementation Barriers
User Experience
Explanations were perceived as distracting and confusing by some users, leading to lower satisfaction.
Proposed Solutions: Adjustments to the explanation delivery method and content could help mitigate this issue.
Project Team
Vedant Bahel
Researcher
Harshinee Sriram
Researcher
Cristina Conati
Researcher
Contact Information
For information about the paper, please contact the authors.
Authors: Vedant Bahel, Harshinee Sriram, Cristina Conati
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