Coregulated Learning Between Human and Artificial Pedagogical Agents in the Context of a Multi-Agent Adaptive Hypermedia Environment

• In Event: Innovations in Researching Regulation of Learning in Solo and Collaborative Tasks

Abstract

Objectives: Contemporary research on multi-agent learning environments has focused on self-regulated learning (SRL) while relatively little effort has been made to use co-regulated learning as a guiding theoretical framework. This oversight requires attention given the complex nature self-and other-regulatory processes play when human learners and artificial pedagogical agents interact to support learners’ internalization of SRL processes. For example, learning with a multi-agent hypermedia learning environment such as MetaTutor involves learner interaction with four artificial pedagogical agents. Accordingly, this paper will focus on the challenges and opportunities of our methodological and analytical approaches.

Theoretical Framework: Our assumptions are in line with Winne and Hadwin’s theory of SRL and Azevedo and colleagues’ assumptions regarding SRL as an event. We have also adopted a contemporary model of co-regulation and its associated assumptions in order to examine and identify: (1) learner-initiated and learner-agent regulatory processes fostering students’ internationalization of regulatory processes and learning about challenging domains. (2) emerging interaction patterns of learner and agent regulatory behaviors, including mediation and internalization of SRL processes; (3) whether distributing regulatory expertise amongst multiple agents is effective in enhancing learners’ internalization of SRL; and (4) if there are emerging interactions of learner-agent predictive of learners’ internalization of SRL processes.

Methods: Participants included 42 college students using MetaTutor (an adaptive, multi-agent hypermedia system) to learn about the human circulatory system during a multi-day experiment. Participants were randomly assigned to one of three conditions—control, prompt, or prompt and feedback. In the prompt and feedback condition, students received timely prompts delivered by the embedded pedagogical agents to use SRL processes and then received feedback regarding the accuracy and /or effectiveness of their use of these processes. Those in the prompt condition received the same prompts but not feedback. Lastly, those in the control condition did not receive prompts from the agents. Agent behaviors were triggered by various conditions including learner’s actions, time thresholds, specific events and episodes (e.g., during a sub-goal), and learners’ responses to agents’ prompts and feedback.

Data Sources: Data sources include dialogue between learners and agents, screen recordings, concurrent think-alouds, video recording of the face (for affect detection and classification), quizzes, summaries, log-files, and notes and drawings.

Results: A complex set of qualitative and quantitative data analyses will: (1) illustrate the importance of the nature of learner-initiated and agent-initiated regulatory moves, (2) highlight the qualitative nature of dialogue between learners and different agents, (3) provide evidence of qualitative and quantitative changes in learners’ cognitive and metacognitive processes throughout the two-hour task, (4) demonstrate the impact of agents’ co-regulation on learners’ instructional choices, and students’ adoption of different regulatory processes.

Contribution: This study will be presented within the framework of co-regulated learning and therefore extends current research on multi-agent learning environments. By doing so, we extend the human and computerized theoretical models typically used in this research area. The results will enhance our understanding of the qualitative nature of students’ internalization of SRL by focusing on dialogue and behaviors of both learners and artificial agents during co-regulated learning.

Authors

• Roger Azevedo, McGill University

• Reza Feyzi Behnagh, McGill University

• Jason Matthew Harley, McGill University

• François Bouchet, McGill University