Decision-guided chatbots and cognitive styles in interdisciplinary learning

Abstract

Research on chatbots has distinct interdisciplinary features and can be generalized to learning in various fields. However, related research is still fragmented across disciplines and applications. This study aimed to develop a decision-guided chatbot for interdisciplinary learning. To investigate the effects of this learning model on learning achievements, learning motivation, collective efficacy, classroom engagement, satisfaction with the learning approach, and cognitive load of learners with different cognitive styles, this study was conducted in an environment education course in a junior high school in northern Taiwan. A total of 71 learners from two classes were recruited in this study; the experimental group, a class of 35 learners, adopted a decision-guided chatbot for learning, while the control group, a class of 36 learners, adopted conventional technology-assisted learning. The results showed that the experimental group significantly outperformed the control group on learning achievements, extrinsic motivation, collective efficacy, cognitive engagement, emotional engagement, and satisfaction with the learning approach. Moreover, the experimental group perceived lower mental efforts. In terms of cognitive styles, analytical learners had significantly higher learning achievements than intuitive learners. In the control group, the analytical learners had higher cognitive engagement than intuitive learners. In the experimental group, analytical learners had significantly lower mental load than intuitive learners. In addition, the analytical learners and intuitive learners in the experimental group respectively perceived higher mental load than those in the control group.