Search for: All records

Research into student affect detection has historically relied on ground truth measures of emotion that utilize one of three sources of data: (1) self-report data, (2) classroom observations, or (3) sen- sor data that is retrospectively labeled. Although a few studies have compared sensor- and observation-based approaches to student af- fective modeling, less work has explored the relationship between self-report and classroom observations. In this study, we use both recurring self-reports (SR) and classroom observation (BROMP) to measure student emotion during a study involving middle school students interacting with a game-based learning environment for microbiology education. We use supervised machine learning to develop two sets of affect detectors corresponding to SR and BROMP-based measures of student emotion, respectively. We compare the two sets of detectors in terms of their most relevant features, as well as correlations of their output with measures of student learning and interest. Results show that highly predictive features in the SR detectors are different from those selected for BROMP-based detectors. The associations with interest and moti- vation measures show that while SR detectors captured underlying motivations, the BROMP detectors seemed to capture more in-the- moment information about the student’s experience. Evidence sug- gests that there is benefit of using both sources of data to model different components of student affect. 

rior research has shown that digital games can enhance STEM education by providing learners with immersive and authentic scientific experiences. However, optimizing the learning outcomes of students engaged in game-based environments requires aligning the game design with diverse student needs. Therefore, an in-depth understanding of player behavior is crucial for identifying students who need additional support or modifications to the game design. This study applies an Ordered Network Analysis (ONA)—a specific kind of Epistemic Network Analysis (ENA)—to examine the game trace log data of student interactions, to gain insights into how learning gains relate to the different ways that students move through an open-ended virtual world for learning microbiology. Our findings reveal that differences between students with high and low learning gains are mediated by their prior knowledge. Specifically, level of prior knowledge is related to behaviors that resemble wheel-spinning, which warrant the development of future interventions. Results also have implications for discovery with modeling approaches and for enhancing in-game support for learners and improving game design. 

Prior research has shown that digital games can enhance STEM education by providing learners with immersive and authentic scientific experiences. However, optimizing the learning outcomes of students engaged in game-based environments requires aligning the game design with diverse student needs. Therefore, an in-depth understanding of player behavior is crucial for identifying students who need additional support or modifications to the game design. This study applies an Ordered Network Analysis (ONA)—a specific kind of Epistemic Network Analysis (ENA)—to examine the game trace log data of student interactions, to gain insights into how learning gains relate to the different ways that students move through an open-ended virtual world for learning microbiology. Our findings reveal that differences between students with high and low learning gains are mediated by their prior knowledge. Specifically, level of prior knowledge is related to behaviors that resemble wheel-spinning, which warrant the development of future interventions. Results also have implications for discovery with modeling approaches and for enhancing in-game support for learners and improving game design. 

Pedagogical agents offer significant promise for engaging students in learning. In this paper, we investigate students’ conversational interactions with a pedagogical agent in a game-based learning environment for middle school sci- ence education. We utilize word embeddings of student-agent conversations along with features distilled from students’ in-game actions to induce predictive models of student engagement. An evaluation of the models’ accuracy and early prediction performance indicates that features derived from students’ conversa- tions with the pedagogical agent yield the highest accuracy for predicting student engagement. Results also show that combining student problem-solving features and conversation features yields higher performance than a problem solving-only feature set. Overall, the findings suggest that student-agent conversations can greatly enhance student models for game-based learning environments. 

Game-based learning offers rich learning opportunities, but open-ended games make it difficult to identify struggling students. Prior work compares student paths to a single expert’s “golden path.” This effort focuses on efficiency, but additional pathways may be required for learning. We examine data from middle schoolers who played Crystal Island, a learning game for microbiology. Results show higher learning gains for students with exploratory behaviors, with interactions between prior knowledge and frustration. Results have implications for designing adaptive scaffolding for learning and affective regulation. 

Game-based learning offers rich learning opportunities, but open-ended games make it difficult to identify struggling students. Prior work compares student paths to a single expert’s “golden path.” This effort focuses on efficiency, but additional pathways may be required for learning. We examine data from middle schoolers who played Crystal Island, a learning game for microbiology. Results show higher learning gains for students with exploratory behaviors, with interactions between prior knowledge and frustration. Results have implications for designing adaptive scaffolding for learning and affective regulation.