Hector Research Institute of Education Sciences and Psychology

AEyeCoL

Advanced Understanding of Eye Gaze in Co-Located Collaborative Learning

Our Goal

This project aims to understand gaze interactions cooperative learning (CL) scenarios, using mobile eye-tracking technology and machine learning techniques. By analyzing gaze interactions, the project seeks to uncover the social and cognitive processes that drive effective collaboration.

The Challenge

Collaborative learning is part of everyday life in schools and universities. Yet one aspect has hardly been researched: the role of eye contact. Why do we seek others’ gaze when learning together, and how does this influence the learning process? This project will work exploratively with two samples from schools and universities to further develop these foundations and generate practical insights for education on the role of eye contact in group learning.

 

Our Questions
  • What are the key gaze indicators?
  • How can computer vision methods be used to automatically extract gaze indicators from eye-tracking data and replace manual annotation?
  • Can group interactions and team dynamics be automatically assessed with machine learning based on gaze indicators?
  • Can methods and tools developed for higher education be adapted and applied in school learning environments?

  • How do CL-task quality and instructional design influence CL-gaze indicators and group dynamics?

     

Background

Collaborative Learning

The PISA 2022 report highlighted collaborative learning (CL) as one of the key components of effective learning strategies for sustained lifelong learning. CL is an active learning process in which individuals work together to address problems and build common understanding, and in that way, co-construct knowledge collaboratively. CL is a widely used form of instruction in schools and higher education.

The Role of Gaze in Collaborative Learning

In CL environments, eye gaze serves as a crucial non-verbal cue that provides insights into coordination, mutual understanding, and engagement between participants, making it a valuable indicator for assessing interpersonal dynamics and cognitive alignment. In social interactions, gaze serves as a crucial indicator, signaling the direction of attention and conveying information on affective and mental states. Consequently, the most widely used gaze indicator for collaboration is joint visual attention (JVA). Research indicates that higher levels of JVA, namely a closer coupling of collaborators' eye gaze potentially representing shared cognition, is positively related to collaboration quality. 

Timeline

August 1, 2025
Project Start
August 1, 2028
Final Report

Project participants

This project is funded by third-party funds from the German Research Foundation.