MathWiz, an intelligent tutoring system integrating culturally responsive teaching and AI, shows promise for addressing critical math proficiency gaps in underserved communities.

Objective: The main goal of this study is to develop and implement MathWiz, an intelligent tutoring system that integrates culturally responsive teaching, mastery learning, and artificial intelligence to revolutionize math education and address the alarming math proficiency gaps in Baltimore City Public Schools and beyond.

Methods: The researchers developed MathWiz with four core components:

1. A personalization module that collects user data to build a digital model representation of each student
2. An adaptive learning engine that evaluates proficiency and customizes learning paths
3. A dynamic lesson generation engine that creates culturally relevant and skill-appropriate lessons
4. A 3D avatar math companion that serves as an interactive tutor

MathWiz incorporates culturally responsive teaching by tailoring content to students' cultural backgrounds and interests. It employs a mastery learning approach where students must demonstrate proficiency in concepts before advancing to more complex topics. The system features multimodal interaction through 3D avatars that reflect diverse cultural identities, particularly representative of Baltimore City Public Schools demographics (70% African-American, 20% Hispanic/Latino).

Key Findings:

• In 2023, Baltimore City Public Schools test results revealed that none of 1,736 students from 13 schools achieved proficiency on the Maryland state math exam, highlighting the critical need for intervention.
• Less than a quarter of Maryland K-12 students are proficient in math, and nationally, only 26% of 12th-grade students score at or above proficiency in mathematics.
• Previous intelligent tutoring systems like ActiveMath, Lexue100, AnimalWatch, Wayang Outpost, ALEKS, and Carnegie Learning's Cognitive Tutor have shown promise but often lack cultural responsiveness or comprehensive adaptive features.
• MathWiz addresses these gaps by providing personalized, culturally relevant content that adapts to students' individual learning needs and cultural contexts.
• The system hypothesizes that integrating culturally relevant content with mastery learning will enhance math proficiency, engagement, and attitudes toward mathematics.

Implications: This research contributes to the field of AI in education by:

• Demonstrating how AI can be leveraged to create more equitable and inclusive learning environments
• Showcasing the integration of culturally responsive teaching with intelligent tutoring systems
• Providing a framework for addressing systemic inequities in mathematics education through technology
• Offering a scalable solution for improving math proficiency in underserved communities
• Establishing a model for how personalization and cultural relevance can enhance student engagement and learning outcomes

Limitations:

• The system is still in development, and its effectiveness has not yet been fully evaluated
• The research does not provide empirical evidence of MathWiz's impact on student achievement
• There are ethical concerns surrounding data collection, privacy, and potential bias that need to be addressed
• The current implementation may be limited to specific mathematical concepts and grade levels
• The study does not address potential challenges in implementation across diverse educational settings

Future Directions: The paper outlines a three-phase approach for future research:

1. Co-design workshops with students and teachers to refine features and strategies
2. Usability studies to evaluate system interface and logic
3. Pilot testing to assess preliminary effectiveness

Future work will involve 50-200 middle and high school students, math teachers, and parents from Baltimore City Public Schools. Data will be collected through pre- and post-assessments, user profiles, interactive lessons, and surveys to further personalize learning, broaden content diversity, and strengthen MathWiz's ability to close educational gaps.

Title and Authors: "Revolutionizing Culturally Relevant Math Education Through AI and Co-Designing Strategies" by Naja A. Mack, Clyde W. Tandjong, Michael B. Adeleke, Elijah Ballou, Amyra Harry, and Jaunel Panton from the Computer Science Department at Morgan State University, Baltimore, Maryland, USA.

Published On: 2025 (for the CHI EA '25 conference) Published By: Extended Abstracts of the CHI Conference on Human Factors in Computing Systems (CHI EA '25), to be held April 26-May 01, 2025, in Yokohama, Japan.