The effect of intelligent affective feedback on learner engagement, satisfaction and perceived learning in online learning

Several studies in the literature have demonstrated the interaction between emotions and the learning process, influencing cognitive aspects such as perception, attention, memory, motivation, thinking, and problem-solving. Thus, understanding emotions within the context of learning is crucial for optimizing learning outcomes, personalizing instruction, and fostering learner engagement and motivation. However, detecting emotions in online learning environments presents challenges compared to face-to-face settings due to the absence of nonverbal emotional cues. This difficulty can lead to feelings of loneliness and disconnection among students, potentially resulting in decreased interest and withdrawal from the course. In recent years, efforts have been made to address this gap in online learning platforms through advancements in the field of affective computing. Affective support systems, integrating intelligent support systems and affective computing techniques, aim to regulate negative emotional states and enhance students' emotion regulation skills, thus enhancing the online learning experience. In this study, a system was developed to assess the engagement levels of online students using facial expression analysis and provide affective feedback to students whose engagement level falls below a predefined threshold. The system aimed to improve students' cognitive, affective, behavioral engagement, satisfaction levels, and perceived learning outcomes. The effectiveness of the intelligent affective support system was examined through a nested design, specifically focusing on the English I course at Anadolu University Open Education Faculty. Results from testing the developed prototype application within the scope of the research demonstrated the effectiveness of affective support notifications when participants' interest in the lesson declined and disengagement occurred. Cognitive reappraisal and attentional deployment strategies, among various emotion regulation strategies, were perceived as particularly effective. Affective support notifications positively influenced participants' cognitive engagement by facilitating goal-setting, reflection on learning, and the use of effective learning strategies. The notifications also supported participants' behavioral engagement by increasing their interaction with learning materials, directing attention towards learning materials and encouraging effort in completing activities. Moreover, notifications had a positive impact on emotional engagement by enhancing motivation and interest in the lesson. These findings were further supported by the analysis of system logs data. Furthermore, participants perceived affective support notifications as beneficial to their learning outcomes, highlighting the potential of notifications to enhance focus and motivation. However, suggestions were made to improve the system's effectiveness, including considerations for the frequency of notifications, display settings, and system interface design. This research provides valuable insights and serves as a model for institutions offering distance education to large audiences, such as Anadolu University Open Education Faculty, by demonstrating the effective integration of affective support into online learning environments.