Emotion generation and regulation during co-regulated learning between human and artificial pedagogical agents in the context of Crystal Island, a multi-agent game-based learning environment.
- Roger Azevedo (North Carolina State U., Theme 1)
- James Lester (North Carolina State U., Theme 2)
- Christopher Williams (North Carolina State U.)
- Michelle Taub (North Carolina State U.)
- Nicholas Mudrick (North Carolina State U.)
- Andy Smith (North Carolina State U.)
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 (Hadwin et al., 2011). This oversight needs to be addressed given the complex nature that self-and other-regulatory processes play when human learners and artificial pedagogical agents (APAs) interact to support learners’ internalization of cognitive, affective, and metacognitive (CAM) SRL processes (Azevedo et al., in press). As such, we face several conceptual, theoretical, methodological, analytical, and educational issues.
We propose to modify and test several of the assumptions underlying self-regulation (SRL), co-regulation (CoRL), and socially-shared regulation (SSRL) from the fields of educational, learning, and computational sciences (e.g., Gratch et al., 2009; Hadwin et al., 2011; Marsela et al., 2010; Harley, Taub, Bouchet, & Azevedo, 2012; under review) using Lester and colleagues’ Crystal Island, a multi-agent game-based learning environment for middle grade science (Rowe et al., 2011). We plan on modifying and creating several versions of the Crystal Island environment (e.g., control, affect-only, regulation-only, affect + regulation) with APAs that embody both computational models of affect and theoretical assumptions underlying SRL, CoRL, and SSRL.
Empirically testing the assumptions with a multi-agent game-based learning environment (such as Crystal Island) is critical to advancing our understanding of how approaches to self- and other-regulated learning (i.e., based on conceptions of SRL, CoRL, and SSRL) induce affective processes. These affective processes may lead human learners to generate specific emotions that either impede (e.g., frustration) or foster learners` ability to regulate their cognitive, affective, and metacognitive (CAM) processes and therefore influence their learning outcomes. The project aims to build on existing interdisciplinary literature and extend current conceptual, theoretical, methodological, and analytical techniques regarding the role of emotions, SRL, and multi-agent game-based environments for science learning. This project falls within the objectives of the LEADS grant and directly answers questions related to each individual theme and also across themes.