Several studies have attempted to capture and analyze the intersect of self-regulated learning (SRL) behaviors and agency (i.e., control over one's own actions) during game-based learning. However, limited studies have attempted to theoretically ground or analytically evaluate these constructs in appropriate theoretical assumptions that can discuss and aptly analyze SRL. As such, this paper argues that complex systems theory, which refers to SRL as a system that is self-organizing, interaction dependent, and emergent, should be integrated into theoretical models of SRL and be analyzed using nonlinear dynamical systems theory techniques to fully capture how learners' SRL behaviors can be captured and scaffolded during game-based learning. This paper guides future discussions and empirical research to understand how to better scaffold learners' SRL behaviors using restricted agency during game-based learning by: (1) understanding scaffolding SRL during game-based learning; (2) reviewing studies that review the intersection of SRL, agency, and game-based learning; (3) discussing the limitations within the field; (4) defining and defend SRL according to complex systems theory; and (5) discussing the open challenges in theoretically, methodologically, and analytically applying complex systems theory to SRL.

Self-regulated learning (SRL) is critical for learning across tasks, domains, and contexts. Despite its importance, research shows that not all learners are equally skilled at accurately and dynamically monitoring and regulating their self-regulatory processes. Therefore, learning technologies, such as intelligent tutoring systems (ITSs), have been designed to measure and foster SRL. This paper presents an overview of over 10 years of research on SRL with MetaTutor, a hypermedia-based ITS designed to scaffold college students' SRL while they learn about the human circulatory system. MetaTutor's architecture and instructional features are designed based on models of SRL, empirical evidence on human and computerized tutoring principles of multimedia learning, Artificial Intelligence (AI) in educational systems for metacognition and SRL, and research on SRL from our team and that of other researchers. We present MetaTutor followed by a synthesis of key research findings on the effectiveness of various versions of the system (e.g., adaptive scaffolding vs. no scaffolding of self-regulatory behavior) on learning outcomes. First, we focus on findings from self-reports, learning outcomes, and multimodal data (e.g., log files, eye tracking, facial expressions of emotion, screen recordings) and their contributions to our understanding of SRL with an ITS. Second, we elaborate on the role of embedded pedagogical agents (PAs) as external regulators designed to scaffold learners' cognitive and metacognitive SRL strategy use. Third, we highlight and elaborate on the contributions of multimodal data in measuring and understanding the role of cognitive, affective, metacognitive, and motivational (CAMM) processes. Additionally, we unpack some of the challenges these data pose for designing real-time instructional interventions that scaffold SRL. Fourth, we present existing theoretical, methodological, and analytical challenges and briefly discuss lessons learned and open challenges.

Undergraduate students (N = 82) learned about microbiology with Crystal Island, a game-based learning environment (GBLE), which required participants to interact with instructional materials (i.e., books and research articles, non-player character [NPC] dialogue, posters) spread throughout the game. Participants were randomly assigned to one of two conditions: full agency, where they had complete control over their actions, and partial agency, where they were required to complete an ordered play-through of Crystal Island. As participants learned with Crystal Island, log-file and eye-tracking time series data were collected to pinpoint instances when participants interacted with instructional materials. Hierarchical linear growth models indicated relationships between eye gaze dwell time and (1) the type of representation a learner gathered information from (i.e., large sections of text, poster, or dialogue); (2) the ability of the learner to distinguish relevant from irrelevant information; (3) learning gains; and (4) agency. Auto-recurrence quantification analysis (aRQA) revealed the degree to which repetitive sequences of interactions with instructional material were random or predictable. Through hierarchical modeling, analyses suggested that greater dwell times and learning gains were associated with more predictable sequences of interaction with instructional materials. Results from hierarchical clustering found that participants with restricted agency and more recurrent action sequences had greater learning gains. Implications are provided for how learning unfolds over learners' time in game using a non-linear dynamical systems analysis and the extent to which it can be supported within GBLEs to design advanced learning technologies to scaffold self-regulation during game play.

Open-world games promote engagement by offering players a high degree of autonomy to explore expansive game worlds. Player goal recognition has been widely explored for modeling player behavior in open-world games by dynamically recognizing players' goals using observations of in-game actions and locations. In educational open-world games, in-game reflection tools can help students reflect on their learning and plan their strategies for future gameplay. Data generated from students' written reflections can serve as a source of evidence for modeling player goals. We present a multimodal goal recognition approach that leverages players' written reflections along with game trace log features to predict player goals during gameplay. Results show that both the highest predictive performance and best early prediction performance are achieved by deep learning-based, multimodal goal recognition models that utilize both written reflection and gameplay features as input. These models outperform unimodal deep learning models as well as a random forest baseline. Multimodal goal recognition using natural language reflection data has significant potential to enhance goal recognition model performance, as well as player modeling more generally, to support the creation of engaging and adaptive open-world digital games.

To effectively process complex information within intelligent tutoring systems (ITSs), learners are required to engage in metacognitive monitoring micro-processes (content evaluations [CEs], judgments of learning [JOLs], feelings of knowing [FOKs], and monitoring progress towards goals [MPTGs]). Learners' average monitoring micro-process strategy frequencies were used to examine learning gains using a person-centered approach as they interacted with MetaTutor. Undergraduates (n = 94) engaged in self-initiated and system-facilitated self-regulated learning (SRL) strategies as they studied the human circulatory system with MetaTutor, a hypermedia-based ITS. Using hierarchical clustering, results showed a difference in learning between clusters differing in metacognitive monitoring process usage. Specifically, learners who used both CEs and FOKs for a greater proportion of monitoring strategy usage had significantly greater learning gains than learners who used MPTGs. Implications for monitoring strategy usage across different micro-processes and the development of ITSs to facilitate and scaffold learners' interactions with these micro-processes via prompting are discussed.

Reflection is critical for adolescents' problem solving and learning in game-based learning environments (GBLEs). Yet challenges exist in the literature because most studies lack a theoretical perspective and clear operational definition to inform how and when reflection should be scaf-folded during game-based learning. In this paper, we address these issues by studying the quantity and quality of 120 adolescents' written reflections and their relation to their learning and problem solving with Crystal Island, a GBLE. Specifically, we (1) define reflection and how it relates to skill and knowledge acquisition; (2) review studies examining reflection and its relation to problem solving and learning with emerging technologies; and (3) provide direction for building reflection prompts into GBLEs that are aligned with the learning goals built into the learning session (e.g., learn about microbiology versus successfully solve a problem) to maximize adolescents' reflection, learning, and performance. Overall, our findings emphasize how important it is to examine not only the quantity of reflection but also the depth of written reflection as it relates to specific learning goals. We discuss the implications of using game-learning analytics to guide instructional decision making in the classroom.

Game-based learning environments are designed to provide effective and engaging learning experiences for students. Predictive student models use trace data extracted from students' in-game learning behaviors to unobtrusively generate early assessments of student knowledge and skills, equipping game-based learning environments with the capacity to anticipate student outcomes and proactively deliver adaptive scaffolding or notify instructors. Reflection is a key component of self-regulated learning, and it is critical in effective learning. However, there is currently limited work exploring the utility of reflection for inducing accurate predictive student models. This article presents a predictive student modeling framework that leverages natural language responses to in-game reflection prompts to predict student learning outcomes in a game-based learning environment for middle school microbiology, CRYSTAL ISLAND. With data from a pair of classroom studies involving 118 middle school students, we investigate the accuracy of early prediction models that utilize features extracted from student trace data combined with word embedding-based representations (i.e., GloVe, ELMo) of student reflection responses. We evaluate the accuracy of the predictive models over time using data from incremental segments of each student's interaction with the game-based learning environment, and we compare against models that omit student reflection features. Results reveal that models encoding students' natural language reflections with ELMo word embeddings yield significantly improved accuracy compared to other representations, with the greatest accuracy demonstrated by an ensemble of predictive models. We discuss the implications of these results for the design of game-based learning environments.

The goal of this study was to investigate 65 students' evidence scores of emotions while they engaged in cognitive and metacognitive self-regulated learning processes as they learned about the circulatory system with MetaTutor, a hypermedia-based intelligent tutoring system. We coded for the accuracy of detecting students' cognitive and metacognitive processes, and examined how the computed scores related to mean evidence scores of emotions and overall learning. Results indicated that mean evidence score of surprise negatively predicted the accuracy of making a metacognitive judgment, and mean evidence score of frustration positively predicted the accuracy of taking notes, a cognitive learning strategy. These results have implications for understanding the beneficial role of negative emotions during learning with advanced learning technologies. Future directions include providing students with feedback about the benefits of both positive and negative emotions during learning and how to regulate specific emotions to ensure the most effective learning experience with advanced learning technologies.

Self-regulated learning (SRL) with advanced learning technologies has shown to significantly augment learners' performance across contexts. Yet studies find learners lack sufficient SRL skills to successfully implement strategies (e.g., judgments of learning, note taking, self-testing, etc.). Current research does not fully explain how and why this failure of effective strategy deployment occurs. We used principle component analysis (PCA) on process data (i.e., log files) from 190 undergraduates learning with MetaTutor, a hypermedia-based intelligent tutoring system, to explore underlying patterns in the frequency of strategy deployment occurring with and without pedagogical agent scaffolding to better understand any underlying structures of system- and learner-initiated cognitive and metacognitive SRL strategy use. Results showed that the system's underlying architecture deploys processes corresponding to both the phases of learning and type of effort allocation according to Winne's (2018) Information Processing Theory of SRL. However, learner-initiated processes for those who received scaffolding only displayed strategy deployment that corresponded to the type of effort allocation required of the processes (i.e., more effortful constructionist processes like note-taking versus short canned responses for judgements of learning). Additionally, results suggest all learners deploy strategies based on the familiarity of processes. Regression models using these principle components outperformed raw frequency models for capturing post-test learning performance across all participants.

Reflection plays a critical role in learning by encouraging students to contemplate their knowledge and previous learning experiences to inform their future actions and higher-order thinking, such as reasoning and problem solving. Reflection is particularly important in inquiry-driven learning scenarios where students have the freedom to set goals and regulate their own learning. However, despite the importance of reflection in learning, there are significant theoretical, methodological, and analytical challenges posed by measuring, modeling, and supporting reflection. This paper presents results from a classroom study to investigate middle-school students' reflection during inquiry-driven learning with Crystal Island, a game-based learning environment for middle-school microbiology. To collect evidence of reflection during game-based learning, we used embedded reflection prompts to elicit written reflections during students' interactions with Crystal Island. Results from analysis of data from 105 students highlight relationships between features of students' reflections and learning outcomes related to both science content knowledge and problem solving. We consider implications for building adaptive support in game-based learning environments to foster deep reflection and enhance learning, and we identify key features in students' problem-solving actions and reflections that are predictive of reflection depth. These findings present a foundation for providing adaptive support for reflection during game-based learning.