Exploring the Cognitive Dynamics of Learners' Deep Learning in an Online Course

Online learning, as a novel approach to education based on the Internet, has become a critical means of advancing educational informatization. However, problems such as “surface-level learning” and “low-quality engagement” are prone to occur. This study aimed to explore new methods for evaluating online deep learning by analyzing and comparing the epistemic networks of different learners, including their network structures and development trajectories. Specifically, 2,470 pieces of text data from 98 participants in an online course were collected and analyzed using Delphi method, content analysis, and epistemic network analysis. The findings revealed that learners engaged in deeper levels of deep learning during the mid-term phase of the course. Evaluative analysis activities were crucial in enhancing learners' performance. Furthermore, deeper-indicator cognitive activities within the same level had a more significant impact on performance improvement. These findings emphasize the importance of strengthening learners' evaluative analysis activities and deeper-indicator cognitive engagement in online learning, especially at the beginning and end of a course.