Optimization of learning strategies in digital environments and platforms

At the University of Santiago de Chile, Dr. Héctor Ponce’s laboratory conducts eye tracking studies to measure the performance of different learning strategies. The results gleaned are becoming crucial in contexts where education takes place in primarily digital environments.

Objective

Computer-based learning helps us to uncover and diversify knowledge. Throughout 2020, due to the Sars-cov2 pandemic, virtual, synchronous, and asynchronous education exploded worldwide bringing all kinds of educational activities to a digital space. It has therefore become more important than ever to measure cognitive processes efficiently in children, their performance, the impact, and the role of different educational and interactive strategies.

The questions posed by Dr. Héctor Ponce require eye tracking methodologies that analyze the cognitive processes and the efficiency of different learning strategies. Eye tracking also helped to adapt and improve interactive interfaces focused on education.

Method

The studies carried out in the lab have focused on virtual interactions such as Drag & Drop, cloze tests and their efficiency, note taking in the text box, graphic organizers in online reading tasks, and other measures of effectiveness associated with response times.

Using screen-based eye tracking, the researchers carried out a series of studies in which they sought to understand the effects of an improved cloze test with drag & drop-type interactions in a digital environment on response time. The performance of a banked cloze test was compared using a conventional interface in paper format.

In another instance, a separate cloze test was analyzed with improvements in the interface using a computer and with the possibility of Drag & Drop. (Reference image No. 2).

In experiment one, 56 fourth-grade students were tested in the conventional group where they responded considerably slower, but just as accurately. They spent more time looking at the word list on the right side of the screen but spent the same time on the text as the group that used the improved interface.

In experiment two, the same pattern of results was replicated with 148 sixth graders. Results were also better for the three versions of the improved interface compared to the conventional interface.

Conclusion

One of the conclusions of this study indicates that the alternative interfaces examined in this set of experiments were able to improve the response execution phase in a cloze test, without excessively affecting the response generation phase.

The accuracy of the responses between the groups in the different experiments (image 3) does not show a significant variation in results. However, the highlighted numbers indicate significantly longer response times (in minutes) for the conventional group, including the time spent in the word list.

Desktop eye trackers, such as the Tobii Pro Nano, Pro Fusion and Pro Spectrum facilitate this type of study, since the stimuli presented on screen allows ease of analysis, and high precision of response time analysis with the mouse and keyboard.

Tobii Pro Lab allows the analysis of different metrics associated with “clicks” and “clicks & releases”. These metrics can be associated with interaction and response times. Since these metrics are related to eye movements and attention, studies of human behavior like these will address a new and innovative perspective. These metrics can be reviewed here.

Publications

The research developed and published by this laboratory has revolved around research in technological education, communication, and development.

To learn more about this and other publications of this research laboratory, you can visit the following links:

An eye movement analysis of highlighting and graphic organizer study aids for learning from expository text - ScienceDirect

https://www.sciencedirect.com/science/article/abs/pii/S0747563213002938?via%3Dihub

https://doi.org/10.1016/j.compedu.2018.06.024

https://link.springer.com/article/10.1007/s11423-020-09740-1

https://journals.sagepub.com/doi/abs/10.1177/0735633120969666