Haley Noh

Haley Noh, Aarna Chowdhary, Jeroen Ooge et al.

Intelligent Tutoring Systems often grant learners shared control over skill and problem selection. Prior work suggests learners exhibit diverse task-selection strategies, such as avoiding challenge, which may interact with mastery learning systems that optimize task selection based on estimated knowledge. Algorithmic constraints on problem selection may help mitigate these effects, but testing such constraints in classrooms is costly. We propose a simulation-based framework to examine how learner task-selection strategies and system constraints shape mastery learning efficiency. Using interaction data from 261 students across two mathematical domains (equation solving and graph interpretation), we simulate strategies such as Weakness Targeting and Interleaving. We evaluate how these strategies affect overpractice as a measure of efficiency. Results show substantial variability across strategies, with risk-averse strategies producing higher levels of overpractice, especially for complex multi-step problems. Targeted system constraints significantly reduce inefficiencies for maladaptive strategies while minimally affecting already efficient strategies. These findings show how simulation grounded in student data can guide the redesign of shared-control tutoring systems before classroom deployment.

Jiayi Zhang, Conrad Borchers, Clayton Cohn et al.

The field of learning analytics has made notable strides in automating the detection of complex learning processes in multimodal data. However, most advancements have focused on individualized problem-solving instead of collaborative, open-ended problem-solving, which may offer both affordances (richer data) and challenges (low cohesion) to behavioral prediction. Here, we extend predictive models to automatically detect socially shared regulation of learning (SSRL) behaviors in collaborative computational modeling environments using embedding-based approaches. We leverage large language models (LLMs) as summarization tools to generate task-aware representations of student dialogue aligned with system logs. These summaries, combined with text-only embeddings, context-enriched embeddings, and log-derived features, were used to train predictive models. Results show that text-only embeddings often achieve stronger performance in detecting SSRL behaviors related to enactment or group dynamics (e.g., off-task behavior or requesting assistance). In contrast, contextual and multimodal features provide complementary benefits for constructs such as planning and reflection. Overall, our findings highlight the promise of embedding-based models for extending learning analytics by enabling scalable detection of SSRL behaviors, ultimately supporting real-time feedback and adaptive scaffolding in collaborative learning environments that teachers value.