Researchers at the Technical University of Munich and the University of Cologne are using AI to provide personalized math support for students by analyzing eye movements. Learn about the breakthrough’s potential impact on education.
A new AI-based learning system developed by researchers at the Technical University of Munich (TUM) and the University of Cologne is set to revolutionize math education by providing individualized support to schoolchildren.
The innovative system, described in a study published in the journal Educational Studies in Mathematics, identifies students’ strengths and weaknesses in mathematics through a webcam that tracks their eye movements. This allows for the generation of personalized problem-solving hints and exercises, significantly enhancing the teacher’s ability to offer tailored support.
“The AI system classifies the patterns,” Achim Lilienthal, a TUM robotics professor, said in a news release.
Utilizing a standard webcam, the system captures eye movement patterns that reveal which areas of the screen draw the student’s attention most frequently. These visual patterns are then converted into heat maps, with red areas signifying frequent focus and green areas indicating brief glances. By analyzing these heat maps, the software customizes learning content for each pupil.
Maike Schindler, professor of mathematics in inclusive and special education contexts at the University of Cologne, emphasized the novelty and importance of the technology.
“Tracking eye movements in a single system using a webcam, recognizing learning strategies via patterns and offering individual support, and finally creating automated support reports for teachers is completely new,” Schindler added.
Her decade-long collaboration with Lilienthal has been pivotal in bringing this technology to fruition.
Funded by the German Federal Ministry of Education and Research (BMBF), the KI-ALF research project recently completed the development of this innovative tool. The project targets students experiencing significant difficulties in learning mathematics, although the potential for customization stretches to high-achieving pupils as well.
The system works by presenting children with a variety of math tasks, such as counting dots in progressively challenging configurations. The eye tracking data reveals which students adapt quickly to these challenges and which ones require additional support.
“Tasks involving visually presented, digital learning materials are particularly suitable for this approach,” added Schindler.
Lilienthal, leveraging his experience in robotics research, overcame the cost barrier associated with high-end eye tracking technology. By incorporating AI to adjust for the lower precision of standard webcams, the researchers developed an affordable yet effective tool for schools.
“Today it makes no difference to our application whether we work with our webcams or high-end eye trackers,” he added.
The Wulfen Comprehensive School in Dorsten, North Rhine-Westphalia, is the first institution to adopt this AI-based learning system. A significant proportion of students at Wulfen had been identified with arithmetic difficulties. Now, the AI-powered system enables the school to support far more students concurrently compared to traditional methods, which often limit personalized attention to one-on-one sessions with teachers.
This breakthrough is especially pertinent in an era where educational resources, including qualified teachers, are in short supply.
Schindler praised the system’s potential during these challenging times.
“Especially in times of scarce resources and teacher shortages, our system for promoting basic math skills is simply an excellent support for schools,” she concluded.
Source: Technical University of Munich