Emotion refers to spontaneous responses to events that are both universally shared and culturally specific (Ekman & Cordaro, 2011). Emotion is a rather complicated concept. In psychology, there are six basic emotions: happiness, sadness, fear, anger, surprise, and disgust (Ekman,1992). However, some of the basic emotions (e.g., disgust) are less observed in learning. Pekrun et al. (2002) examine academic emotions and propose activity-related emotions (e.g., enjoyment and boredom) and outcome-related ones (e.g., hope, anxiety, and shame). Furthermore, they claim that some emotions are caused by the present learning content like confusion, while others are caused by future events like anxiety (Pekrun et al., 2022).

The Circumplex Model of Affect (Russell, 1980) classifies emotions based on two factors: Valence and Arousal. Valence represents the positivity or negativity of an emotion, while arousal indicates the level of activation or intensity. Accordingly, there are four quadrants: High Arousal, Positive Valence; High Arousal, Negative Valence; Low Arousal, Positive Valence; and Low Arousal, Negative Valence. The emotions in the High Arousal, Positive Valence quadrant, like excitement and happiness are about emotional engagement, which is the focus of the study. In addition, emotional change or emotional patterns caused by learning activities or events are more critical than emotion itself in the learning process (Huang et al., 2024).

Engagement is the cognitive and affective efforts learners devoted to learning (Fredricks et al., 2004). Engagement has three common dimensions: behavioral, cognitive, and emotional/effective (Martin & Borup, 2022). Behavioral engagement refers to learners' observable actions, such as attention, curiosity, and concentration. Cognitive engagement highlights the mental effort learners invest in the learning process (Fredricks et al., 2004). Emotional engagement pertains to the affective states that learners experience during the learning process (Skinner & Belmont, 1993). Emotional engagement, the focus of the study, is highly correlated with emotions and engagement but different from them. Emotional engagement particularly refers to learning-related emotions in the quadrant of High Arousal and Positive Valence.

The relationships among the dimensions are complicated. Some researchers claim that emotional engagement affects cognitive engagement and behavioral engagement (Pekrun & Linnenbrink-Garcia, 2012), while others state that emotional engagement is triggered by cognitive engagement and then enhances or impedes the cognitive learning strategy use (Obergriesser & Stoeger, 2020). In addition, some researchers identify that the dimensions overlap with intersections like cognitive-emotional engagement (D’Errico et al., 2018). It seems that the dimensions are highly correlated and dynamically interacting and mutually reinforcing the learning process.

The above review shows that emotional engagement is highly important in the regular learning process. It is even more crucial in the online learning environment due to the perceived high isolation and limited social interaction. However, emotional engagement is less studied in empirical studies (Bedenlier et al., 2020). The recent development of Generative AI (GenAI) shows its great potential to emotionally engage online learners through their interaction with the GenAI chatbot. However, limited empirical studies have been conducted to investigate the affordance of GenAI for learner emotional engagement. Therefore, this study aims to design an interactive GenAI-powered chatbot to emotionally engage online learners in the BSL setting by providing personalized instructional support and emotional support. Specifically, the study aims to answer the following research questions:

1. What features should an interactive GenAI chatbot have to foster online learner emotional engagement?
2. How effectively does the GenAI chatbot facilitate online learner emotional engagement?
3. How do learners perceive the use of a GenAI chatbot to support their emotional engagement?

Context This proposed study will be conducted in the courses offered for adult learners who are taking a Masters’ programme at a university in Singapore. Each course will have approximately 20 learners. It consists of 13 weekly teaching sessions, each lasting 3 hours. About 6 to 8 sessions will be conducted in a blended synchronous learning (BSL) mode, which involves both onsite and online learners simultaneously. Research design This study will follow the educational design research approach, which is a systematic method for addressing complex educational issues while generating both theoretical and practical insight. This approach involves design, development, and implementation of innovative educational interventions within authentic contexts (Plomp, 2023). It is particularly well suited to this study, as the integration of a GenAI chatbot into a BSL environment represents a relatively novel attempt with limited experience in existing research. Consequently, iterations of formative evaluation and reflection are essential to refine the intervention and achieve an optimal solution. More concretely, this study will consist of three major stages: preliminary research, prototyping, and assessment (Wang et al., 2017). During the preliminary research stage, tentative design principles for the design of a GenAI chatbot will be formulated through context analysis, needs analysis, and a literature review. The prototyping stage will involve iterative design, development, implementation, and formative evaluation. The design will specifically focus on pedagogical, social, and technical perspectives (Wang et al., 2017). The development will adopt DeepSeek or ChatGPT's API. The implementation will be conducted in an authentic BSL setting. A formative evaluation will be conducted in each iteration to collect feedback and improve the quality of the chatbot. The prototyping stage will answer the first research question. After the prototyping stage , the finalized chatbot will be further examined through a quasi-experiment in the assessment stage. Assessment The assessment will be conducted with two classes, with one using the GenAI chatbot as an experiment group and the other without using the chatbot as a control group. The assessment stage will answer the remaining two research questions. Multimodal data will be collected and analyzed during the assessment stage. Real-time facial expression data to assess emotional changes as a non-intrusive way will be collected. Additionally, lesson observations, surveys, interviews, and student feedback data will also be collected and analyzed to understand learners’ feedback and perspectives on using the chatbot.

This paper presents a proposal for the first author’s PhD study. The study is presently at the preliminary research stage, and detailed design and implementation have yet to be carried out. Through a literature review, she has established a conceptual framework and identified initial design principles guiding the design of this study. The framework outlines the relationship of key concepts like emotion, engagement, emotional engagement, and GenAI. The literature review identifies that the relationship between cognitive engagement and emotional engagement is highly strong. An increase in cognitive engagement may lead to positive emotional engagement. Therefore, as an intervention to enhance online learner emotional engagement, the GenAI chatbot will provide both emotional support and instructional support. By facilitating learners’ cognitive development, the chatbot has the potential to foster their positive emotions, thereby leading to positive emotional engagement. After the preliminary research phase, the next step will gradually refine the design principles and develop the GenAI chatbot prototype. Furthermore, this study will go through 3 to 4 iterations conducted in an authentic BSL setting to continuously refine the design specifications and principles. After the prototyping phase, a concrete GenAI chatbot will be finally developed, and it will be further examined during the assessment phase to investigate the extent of the chatbot enhancing online learner emotional engagement. Online learner emotional engagement with GenAI is a rather new research area. The purpose of this paper is to share our tentative research ideas and gather constructive feedback from fellow researchers to refine the study design. Additionally, hopefully this proposal will inspire further research in this field.

Bedenlier, S., Bond, M., Buntins, K., Zawacki-Richter, O., & Kerres, M. (2020). Facilitating student engagement through educational technology in higher education: A systematic review in the field of arts and humanities. Australasian Journal of Educational Technology, 36(2), 126–150. https://doi.org/10.14742/ajet.5477 D'errico, F., Paciello, M., De Carolis, B., Vattanid, A. P., Alestra, G., & Anzivino, G. (2018). Cognitive emotions in e-learning processes and their potential relationship with students' academic adjustment. International Journal of Emotional Education, 10(1), 89–111. Ekman, P., & Cordaro, D. (2011). What is meant by calling emotions basic. Emotion Review, 3(4), 364–370. https://doi.org/10.1177/1754073911410740 Ekman, P. (1992). An argument for basic emotions. Cognition and Emotion, 6(3–4), 169–200. https://doi.org/10.1080/02699939208411068 Fredricks, J. A., Blumenfeld, P. C., & Paris, A. H. (2004). School Engagement: Potential of the Concept, State of the Evidence. Review of Educational Research, 74(1), 59–109. https://doi.org/10.3102/00346543074001059 Huang, C., Yu, J., Wu, F., Wang, Y., & Chen, N. (2024). Uncovering emotion sequence patterns in different interaction groups using deep learning and Sequential Pattern Mining. Journal of Computer Assisted Learning, 40(4), 1777–1790. https://doi.org/10.1111/jcal.12977 Martin, F., & Borup, J. (2022). Online learner engagement: Conceptual definitions, research themes, and supportive practices. Educational Psychologist, 57(3), 162–177. https://doi.org/10.1080/00461520.2022.2089147 Obergriesser, S., & Stoeger, H. (2020). Students’ emotions of enjoyment and boredom and their use of cognitive learning strategies – how do they affect one another? Learning and Instruction, 66, 101285. https://doi.org/10.1016/j.learninstruc.2019.101285 Pekrun, R., Goetz, T., Titz, W., & Perry, R. P. (2002). Academic emotions in students’ self-regulated learning and achievement: A program of qualitative and quantitative research. Educational Psychologist, 37(2), 91–105. https://doi.org/10.1207/s15326985ep3702_4 Pekrun, R., & Linnenbrink-Garcia, L. (2012). Academic emotions and student engagement. In S. L. Christenson, A. L. Reschly, & C. Wylie (Eds.), Handbook of Research on Student Engagement (pp.259–282). https://doi.org/10.1007/978-1-4614-2018-7_12 Plomp, T. (2013). Educational design research: An introduction. In T. Plomp & N. Nieveen (Eds.), Educational design research (pp.11-50). Netherlands Institute for Curriculum Development (SLO). Russell, J. A. (1980). A circumplex model of affect. Journal of Personality and Social Psychology, 39(6), 1161–1178. https://doi.org/10.1037/h0077714 Skinner, E. A., & Belmont, M. J. (1993). Motivation in the classroom: Reciprocal effects of teacher behavior and student engagement across the school year. Journal of Educational Psychology, 85(4), 571-581. https://doi.org/10.1037/0022-0663.85.4.571 Wang, Q., Quek, C. L., & Hu, X. (2017). Designing and improving a blended synchronous learning environment: An educational design research. The International Review of Research in Open and Distributed Learning, 18(3). https://doi.org/10.19173/irrodl.v18i3.3034