Analysis of a chatbot as a dialogic reading facilitator: its influence on learning interest and learner interactions

Abstract

Educational chatbots are gaining momentum due to their distinctive affordances of interactivity, immediacy, ease of use, and individualized experience. However, a fairly limited body of literature discusses how a chatbot can facilitate collaborative learning among peers in extensive reading contexts to encourage more vibrant interactions supporting further interest development. Therefore, this research aimed to analyze the affordances and limitations of a chatbot to facilitate human–human interactions by incorporating the refined Academically Productive Talk framework for nurturing a learning community, forming accurate knowledge, fostering rigorous thinking, and encouraging affective responses for elementary school learners. Specifically, the purpose of the research was to observe the situational interest of the learners, their interaction patterns, and their social learning behaviors. This research developed a chatbot stored with 64 children’s storybooks to initiate and facilitate peer dialogues. A group of 30 learners were paired up to conduct two chatbot-facilitated dialogic reading activities. A total of 30 discourse logs and students’ feedback on a survey of situational interest were analyzed. The discourse analysis of this research supports the affordances of the chatbot acting as an effective dialogue initiator and discussion facilitator to support both human-chatbot and human–human social learning. The chatbot encourages a diverse interactive dialogic climate, and four interaction patterns were identified. The situational interest of the initial encounter with the chatbot was boosted; however, their interest was unable to be sustained. The implications for the affordances and limitations of educational chatbots are discussed.