Research visit to Switzerland
Last week, I returned from my three-week research visit to the Institute for Research in Open, Distance and eLearning (IFeL) at the Swiss Distance University of Applied Sciences (FFHS), which also holds the UNESCO Chair on Personalized and Adaptive Learning in Distance Education. This visit was made entirely possible thanks to an expenditure grant from Suomen Kulttuurirahasto (The Finnish Cultural Foundation). I am very grateful to Egon Werlen and the entire IFeL team for their amazing hospitality and engaging in fruitful discussions.
The visit began with a presentation of my doctoral project, where I received valuable feedback and a host of questions that will directly shape my work. As a result of my visit, we will continue to collaborate on my second PhD paper. Thanks to the feedback I received, I will now incorporate entropy analysis to capture emotional dynamics more effectively. This analysis will provide a measure of the stability and complexity of emotional patterns over time, allowing us to detect changes and shifts in emotion dynamics. This will enable me to formulate a testable hypothesis for a larger dataset my third PhD paper is going to be based on. We also briefly discussed potential avenues for the third paper, potentially leveraging machine learning methods like Recurrent Neural Networks (RNNs). I also had the opportunity to test the latest version of FaceReader. This allowed me to compare results on my dataset with those from iMotions, leading to an interesting discussion about the advantages and disadvantages of both softwares for emotion recognition. I dedicated time to refining the theoretical framework of my dissertation, working on my manuscript, and addressing various technical challenges, such as data synchronization issues, that arise during data collection and analysis.
I participated in multiple workshops and became familiar with the past and current projects done at IFeL, including those in UNESCO cooperation with South African, Tanzanian, and other partners.
For instance, I gained detailed insight into other methods for studying real-time cognition and affect dynamics, such as mouse tracking. I was excited to learn more about its implementation in online learning through the Swiss-South African research group’s SAMBA project (Skills and Attitudes through Machine-Based Adaptive attention guidance). Mouse clicking can serve as a low-cost and deployable proxy for cognitive (i.e., attention) as well as affective (i.e., emotions like boredom) processes during learning. We discussed the raw data structure, various mouse tracking features, and how this technique relates to other process-oriented measures like eye-tracking, facial recognition, and subjective reports.
I also learned how machine learning is being applied in the context of math assessment through DeepScan, a system that provides automatic feedback on students’ graphs in math-related courses. While natural language processing effectively handles written elements, graphic elements often pose a challenge. DeepScan allows for the automatic assessment of these graphic elements and will link them to written answers. The goal of this system is not to replace teacher assessment, but to provide educators overwhelmed with grading with initial assessment so they can distribute their limited attentional resources more effectively.
Another exciting area of research I explored was virtual reality in the learning context. I had a chance to play around with a VR prototype developed within the LAVIR project (Learning Analytics in Virtual Reality). This prototype, which includes a 3D heart model, quizzes, and posters, is designed to be tested with pre-service teachers in South Africa to explore what learning analytics activities in a virtual classroom might provide.
In line with IFeL’s strong focus on self-regulated learning and adaptive learning, I also got familiar with self-control tasks implemented in their online courses. Finally, the team is actively engaged in developing and validating a robust framework to evaluate teachers' digital and AI competences.
The visit was not all work! I also got to explore the wonderful mountain landscapes of Brig and surrounding areas and visit Zurich, Lucerne, and Bern.
Overall, I successfully advanced the methodological part of my doctoral project and gained knowledge about multimodal learning analytics and personalised and adaptive learning. The three weeks at IFeL have set a strong foundation for ongoing and future collaborations.