Although previous research suggests that programming games can improve accessibility and engagement for novice programmers, they may not provide the timely, structured scaffolding novices need to reason systematically about programming problems. To address this gap, we introduce MazeMate, a large language model (LLM)–based chatbot embedded in a programming game, 3D Maze, that provides on-demand and contextualized support during maze-based programming tasks. MazeMate is designed using computational thinking (CT) as a guiding framework, with scaffolds that explicitly target decomposition, pattern recognition, abstraction, and algorithmic thinking to support systematic problem-solving processes. MazeMate was implemented and evaluated in an undergraduate digital literacy course across three maze-solving tasks of increasing difficulty. All students had access to MazeMate and voluntarily chose whether to use it. We examined students’ interaction counts with MazeMate and the solution efficiency accordingly. We found that students who interacted with MazeMate have achieved higher solution efficiency in the lower-difficulty task than those who did not, whereas in higher-difficulty tasks, increased interaction was not consistently associated with improved performance. These findings informed a set of design implications to guide subsequent iterations of CT-aligned scaffolding for MazeMate. This work contributes empirical evidence on the use of Computational Thinking as a design framework for LLM-based scaffolding in game-based programming environments.

Students with strong Computational Thinking (CT) skills possess a unique ability to analyze problems, devise efficient solutions, and navigate the intricacies of a rapidly evolving digital landscape. Given the conceptual overlapping between CT skills and engineering design competencies, engineering design processes provide students with a context for applying and developing CT skills. However, how to promote students to develop CT skills through pedagogical design in engineering education needs further research, especially in the formal higher education context. To address this gap, we constructed a model and designed a course that supports students in applying CT (i.e., decomposition, pattern recognition, abstraction, algorithm design, and troubleshooting/debugging) skills during multiple engineering design iterations. We collected 13 group design reports from 62 undergraduate students regarding their efforts in designing and solving mazes over three design iterations by applying CT skills. Using mixed methods, we examined what and how CT skills were demonstrated in the group reports, and what changes groups made between design iterations and why. We found that the participants demonstrated five CT skills with differing frequencies and needed more support in troubleshooting. When making changes between design iterations, groups mainly considered enabling users to apply CT skills, avoiding hard coding, adjusting the complexity of the mazes, considering design constraints to meet engineering design requirements, and enhancing user experience. The findings underscore the pressing need to equip students with the ability to navigate and resolve intricacies, particularly in troubleshooting, and groups' abilities to consider various elements when making engineering design decisions.

Reliance on AI describes the behavioral patterns of when and how individuals depend on AI suggestions, and appropriate reliance patterns are necessary to achieve effective human-AI collaboration. Traditional measures often link reliance to decision-making outcomes, which may not be suitable for complex problem-solving tasks where outcomes are not binary (i.e., correct or incorrect) or immediately clear. Therefore, this study aims to develop a scale to measure undergraduate students' behaviors of using Generative AI during problem-solving tasks without directly linking them to specific outcomes. We conducted an exploratory factor analysis on 800 responses collected after students finished one problem-solving activity, which revealed four distinct factors: reflective use, cautious use, thoughtless use, and collaborative use. The overall scale has reached sufficient internal reliability (Cronbach's alpha = .84). Two confirmatory factor analyses (CFAs) were conducted to validate the factors using the remaining 730 responses from this activity and 1173 responses from another problem-solving activity. CFA indices showed adequate model fit for data from both problem-solving tasks, suggesting that the scale can be applied to various human-AI problem-solving tasks. This study offers a validated scale to measure students' reliance behaviors in different human-AI problem-solving activities and provides implications for educators to responsively integrate Generative AI in higher education.

BackgroundSocial media provides new opportunities for teachers to learn, communicate and develop professional relationships. It has been proved to be a valid and helpful resource for teachers' professional learning purposes.ObjectivesWhile previous studies pursued questions like how participants feel, how to support interaction and why participants remain committed, we asked a more fundamental question: what is the structure of a massive informal online professional learning network and what dynamics can we expect regarding participants' commitment?MethodsThis work presents an empirical study of massive informal online professional networks to investigate the dynamics of learning communities and participants' commitment over time. We employed social network analysis and data mining techniques on a longitudinal data set of more than 400,000 tweets published with the hashtag "#edchat."Results and ConclusionsWe found that around 30% of participants remained committed to the informal learning community over time. Meanwhile, as more and more people committed to the online learning community, participants tended to form smaller communities where the internal connection was stronger.TakeawaysIn informal online learning environments, participants can form stable connections. The 30% threshold can be used to measure massive informal online learning networks in terms of commitment or persistence.

Computational Thinking (CT) is a foundational problem-solving skill, and gamified programming environments are a widely adopted approach to cultivating it. While large language models (LLMs) provide on-demand programming support, current applications rarely foster CT development. We present MazeMate, an LLM-powered chatbot embedded in a 3D Maze programming game, designed to deliver adaptive, context-sensitive scaffolds aligned with CT processes in maze solving and maze design. We report on the first classroom implementation with 247 undergraduates. Students rated MazeMate as moderately helpful, with higher perceived usefulness for maze solving than for maze design. Thematic analysis confirmed support for CT processes such as decomposition, abstraction, and algorithmic thinking, while also revealing limitations in supporting maze design, including mismatched suggestions and fabricated algorithmic solutions. These findings demonstrate the potential of LLM-based scaffolding to support CT and underscore directions for design refinement to enhance MazeMate usability in authentic classrooms.

The study examined students' spatial collaborative problem-solving behaviors when engaging in a design task dependent on spatial reasoning. Thirty undergraduate students working alone and collaboratively were tested on performance differences in solving an origami Sonobe cube as an active hands-on spatial problem solving task. Epistemic network analysis and sequential pattern mining were conducted to reveal the relationships among collaborative problem solving behaviors and students' embodied engagement displayed in low - versus high-performing student pairs. The core findings were that successful student pairs (higher scores on their final sketches) engaged more in sketching and gesturing in their collaborative problem-solving process, whereas their less successful peers engaged more in trial-and-error or heuristic experimentation by manipulating the origami units. This study can contribute to developing specific pedagogical strategies that better prepare students for collaboration on tasks that require a high level of spatial cognition.

The increasing use of Generative Artificial Intelligence (GAI) tools in education highlights the need to understand their influence on individuals' thinking processes and agency. This research explored 20 university students' interaction with GAI during programming. Participants completed surveys, recorded their screens during an hour-long programming session, and reflected on their GAI use. To analyse the data, we developed an AI-augmented thinking coding scheme with four dimensions: Question Formulation, Solution Development, Solution Analysis and Evaluation, and Solution Refinement. Participants were categorised into human-led and AI-led groups based on the time ratio of human-generating source code versus copying source code from GAI. T-tests indicated that the human-led group spent significantly more time on Solution Development and Solution Refinement than the AI-led group. Sequential pattern mining revealed distinct patterns of the two groups: the human-led group often refined GAI outputs, while the AI-led group frequently relied on direct answers from GAI. Correlation analyses found that positive attitudes towards AI, critical thinking, and programming self-efficacy positively correlated with Question Formulation; critical thinking was positively related to Solution Refinement; and programming self-efficacy was negatively associated with Solution Analysis and Evaluation. This study enhances understanding of the thinking process in GAI-supported programming.

This research investigates distinct human-generative AI collaboration types and students’ interaction experiences when collaborating with generative AI (i.e., ChatGPT) for problem-solving tasks and how these factors relate to students’ sense of agency and perceived collaborative problem solving. By analyzing the surveys and reflections of 79 undergraduate students, we identified three human-generative AI collaboration types: even contribution, human leads, and AI leads. Notably, our study shows that 77.21% of students perceived they led or had even contributed to collaborative problem-solving when collaborating with ChatGPT. On the other hand, 15.19% of the human participants indicated that the collaborations were led by ChatGPT, indicating a potential tendency for students to rely on ChatGPT. Furthermore, 67.09% of students perceived their interaction experiences with ChatGPT to be positive or mixed. We also found a positive correlation between positive interaction experience and a sense of positive agency. The results of this study contribute to our understanding of the collaboration between students and generative AI and highlight the need to study further why some students let ChatGPT lead collaborative problem-solving and how to enhance their interaction experience through curriculum and technology design.

Fostering conducive epistemic emotions to students' knowledge building (KB) is critical but challenging. Considering the potential role of metacognition and KB in promoting conducive epistemic emotions, we designed a quasi-experiment and examined the influence of collaborative analytics-enhanced reflective assessment (demonstrate metacognition) on conducive epistemic emotions to students' KB. The experimental class included 48 undergraduates engaged in KB inquiries enhanced by collaborative analytics-enhanced reflective assessment. In contrast, the comparison class included 46 undergraduates engaged in KB enhanced by portfolio assessment. Students' online Knowledge Forum discourse was collected and coded regarding the demonstrated epistemic emotions. Comparative analysis of the frequency distribution and lag sequential patterns of epistemic emotions revealed that the collaborative analytics-enhanced reflective assessment design fostered conducive epistemic emotions to KB among the undergraduates. Correlation analysis revealed positive correlations between conducive epistemic emotions, higher-level KB discourse moves, and individual knowledge creation. The findings enrich our understanding of promoting undergraduates' conducive epistemic emotions and uncover the relationships between metacognition, epistemic emotions, and KB practices. It provides implications for developing a reflective and collaborative culture, providing scaffoldings and opportunities for expressing and analyzing epistemic emotions, utilizing prompt sheets and collective reflections to enhance the productive use of analytics, and fostering a conducive emotional experience for students.

The application of virtual Reality (VR) games in education has been explored in recent years with successful outcomes in promoting student engagement and motivation. Escape room, an increasingly popular form of games used for both recreational and educational purposes, has shown to promote creative thinking and support assessment to evaluate students' understanding of a particular subject or skill. In this work, we present a VR escape room game for assisting university level computer science students to learn basic concepts in computer architecture. Situated in a futuristic sci-fi theme, our game challenges students' understanding in (1) binary, decimal, and hexadecimal number representations, (2) memory and register access operations, (3) assembly language, as well as (4) string and instruction representation. Our post-game questionnaires have shown high student immersion as well as positive learning outcomes.

Although the educational potential of social media has been widely acknowledged, it remains unclear whether social media use might present opportunities for the development of math identity. This study drew on the socialization framework of math identity to examine the relationships between social media use for math learning, bicultural identity integration, ethnic-math beliefs and math identity among ethnic minority high school students in China. Structural equation modelling results showed that (a) social media use was positively related to math identity; (b) high bicultural identity integration was positively associated with math identity, while biased ethnic-math beliefs were negatively related to math identity and (c) social media use suppressed the socialization process of math identity, weakening the maladaptive roles of low bicultural identity integration and biased ethnic-math beliefs in shaping math identity. Our findings highlight the beneficial role of social media use in student math learning. Theoretical and practical implications are discussed. Practitioner notes What is already known about this topic What this paper adds Implications for practice and/or policy Social media use for learning is associated with students' academic motivation, engagement and academic achievement. Social media use may have affordances for identity development, especially for ethnic minorities. The formation of ethnic minority students' math identity is intertwined with their math and ethnic socialization. Social media use for math learning was directly and positively related to ethnic minorities' math identity. This study quantitatively confirmed the associations of bicultural identity integration and ethnic-math beliefs with math identity. Social media use suppressed the relations of bicultural identity integration and ethnic-math beliefs with math identity. Promoting social media use for math learning benefits ethnic minority students' math identity. Social media use for math learning could alleviate the maladaptive role of low bicultural identity integration and biased ethnic-math beliefs in math identity development. Ethnic minority students' math identity socialization processes merit attention.

Previous research is replete with evidence that emotions and self-regulation work together to influence learning performance, but distinct trait and state features of emotions and self-regulation are rarely considered. With an analytic sample comprising 9,501 daily diaries from 280 adolescents participating in a self-driven learning program, this study used multilevel modeling to examine how trait and state positive and negative emotions and self-regulation interact to predict adolescents' perceived daily learning progress. Results suggested that daily perceived learning progress was associated with trait and state positive emotions and self-regulation, as well as trait negative emotions. Furthermore, there was a significant positive interaction between state positive emotions and state self-regulation on perceived daily learning progress, such that when adolescents' state self-regulation was higher than usual, their perceived daily learning progress was more sensitive to state positive emotion. Results underscore the importance of enhancing adolescents' self-regulation and positive emotion, and the feasibility of facilitating adolescents' learning even if they are in a state of greater negative emotion.

BackgroundSocial annotation has emerged as a promising educational technology that fosters collaborative reading and discussion of digital resources among learners. While the positive impact of social annotation on students' learning process and performance is widely acknowledged, students' behavioural patterns in social annotation are underexplored.ObjectivesThis study investigated patterns in students' use of annotation and response behaviours in social annotation activities. We also explored how students' performance in the behavioural, cognitive, emotional, and social dimensions varied based on their behavioural patterns.MethodsWe recruited 93 undergraduates who were enrolled in an elective course at a large North American University. Students were tasked with collaboratively annotating the class readings uploaded to Perusall, a social annotation platform, over 7 weeks. We used metaclustering to determine the optimal number of clusters pertaining to student behaviours. We compared the differences among clusters across multiple performance dimensions.Results and ConclusionsTwo distinct clusters were identified and defined as initiators and responders. We found that responders had significantly longer active reading time and exhibited greater social annotation effort compared to initiators. However, initiators received more peer acknowledgement, as evidenced by higher upvotes. No significant difference was found in cognitive insight between initiators and responders, but responders demonstrated significantly higher cognitive discrepancy. Additionally, there were no significant differences in positive and negative tones between initiators and responders; however, responders displayed higher levels of prosocial behaviours than initiators. This study has significant practical implications regarding promoting students' collaborative learning experience in social annotation. What is already known about this topic Annotation and response behaviours are two primary actions in social annotation. Understanding how students navigate through annotations and respond to their peers' contributions is essential for optimizing their learning experience. Social annotation outcomes can be assessed in the behavioural, cognitive, emotional, and social dimensions.This paper revealed students' behavioural patterns in social annotation activities. This study offered a comprehensive understanding of the various dimensions of performance among students with different behavioural tendencies.Implications for practice and/or policy The two student clusters, initiators and responders, revealed distinct engagement patterns in social annotation and informed the design of targeted scaffoldings. Responders were not passive learners since they demonstrated significantly longer active reading time, greater social annotation effort, and higher level of prosocial behaviours. Educators might not need to place significant emphasis on monitoring participants' emotional expressions in social annotation. Social annotation platforms should incorporate features that encourage and reward both initiation and response behaviours.

Purpose offers several important benefits to youth. Thus, it is necessary to understand how a sense of purpose develops in supportive contexts and what psychological resources can help. From 2021 to 2022, this study investigated purpose change among 321 youth (M-age = 16.4 years; 71% female; 25.9% Black, 33.3% Asian, 15.6% Hispanic/Latinx, 13.4% White, 9.7% multiracial) participating in GripTape, a similar to 10-week self-driven learning program. Many youth started with high initial purpose that increased throughout enrollment (Strengthening), whereas others began with slightly lower purpose that remained stable (Maintaining). For each unit increase in baseline agency, youth were 1.6x more likely to be classified as Strengthening. As such, agency may be a resource that helps youth capitalize on certain types of environments.

The way teachers support students' Knowledge Building discourse may influence their subsequent discourse moves and emotions. However, in previous research on Knowledge Building, teacher scaffolding was rarely scrutinized, especially in offline Knowledge Building discourse; neither was how the support is associated with students' discourse moves or emotions. This study addressed these issues by examining how two teachers scaffolded the offline Knowledge Building discourse of a class of students from grade 2 to grade 3 and the associations between teacher scaffolding and students' discourse moves and emotions. Through discourse analysis, we identified a few patterns of teaching scaffolding for Knowledge Building, including giving instructions, asking for elaboration, revoicing, asking for explanations, inviting different ideas, and inviting new directions. Using correlation analyses, we found positive correlations between teacher asking for explanations and students providing partial or elaborated explanations. Teacher inviting new directions was positively correlated with students' questions. There was a negative correlation between teacher asking for elaboration and student curiosity. This study has implications for how teacher support can be designed to foster desired student discourse moves and emotions.

The significance of interdisciplinary learning has been well-recognized by higher education institutions. However, when teaching interdisciplinary learning to junior undergraduate students, their limited disciplinary knowledge and underrepresentation of students from some disciplines can hinder their learning performance. ChatGPT’s ability to engage in human-like conversations and massive knowledge grounded in different disciplines holds promise in enriching undergraduate students with the disciplinary knowledge that they lack. In this exploratory study, we engaged 130 undergraduate students in a three-condition quasi-experiment to examine how ChatGPT influences their demonstrated and perceived interdisciplinary learning quality, as measured by their online posts and surveys, respectively. The content analysis results show that overall, students’ online posts could be coded into four interdisciplinary learning dimensions: diversity, disciplinary grounding, cognitive advancement, and integration. The means of the first three dimensions were close to the middle level (ranging from 0.708 to 0.897, and the middle level is 1), whereas the mean score of integration was relatively small (i.e., 0.229). Students under the ChatGPT condition demonstrated improved disciplinary grounding. Regarding their perceived interdisciplinary learning quality, we did not find significant differences across the three conditions in the pre- or post-surveys. The findings underscore ChatGPT’s ability to enhance students’ disciplinary grounding and the significance of further fostering their integration skills.

BackgroundUndergraduates' collective epistemic agency is critical for their productive collaborative inquiry and knowledge building (KB). However, fostering undergraduates' collective epistemic agency is challenging. Studies have demonstrated the potential of computer-supported collaborative inquiry approaches, such as KB-the focus of this study, and reflective assessment to foster collective epistemic agency.ObjectivesThis study used a quasi-experimental design to examine the impact of reflective assessment on undergraduates' collective epistemic agency and the mechanisms of this impact.MethodsAn experimental class, comprising 40 undergraduates, engaged in KB inquiry enhanced by analytics-assisted reflective assessment, while a comparison class, comprising 41 undergraduates, engaged in portfolio-supported reflective assessment in KB inquiry. Classroom observations, audio recordings of group discussions and reflections, and student artefacts were collected in both conditions.Results and ConclusionsThe experimental class demonstrated significantly higher levels of collective epistemic agency than the comparison class, as indicated by their collective knowledge advancement and epistemic and metacognitive actions. Qualitative thematic analysis revealed that reflective assessment enhanced the undergraduates' collective epistemic agency through their collective monitoring and regulation of inquiry and ideas, negotiation and synthesis of the community's ideas, conceptualization of collective theories and identification of collective inquiry trajectories with an internal reflective structure.Major TakeawaysHigher-level metacognitive actions are critical for productive collaborative inquiry and higher-level epistemic agency that does not often occur naturally and should be promoted by appropriate scaffolding strategies. Students need data from learning analytics and accompanying prompt sheets to help them develop the internal metacognitive structure for agency and learning. What is currently known about this topic?Implications for practice/or policyUndergraduates' collective epistemic agency (CEA) is critical for productive KB.Analytics-assisted reflective assessment (AARA) are promising in fostering CEA.What does this paper add?AARA has positive influences on students' CEAAARA helps students develop CEA through the following mechanisms.AARA supports students' collective regulation and negotiation of collective ideas.AARA enhances conceptualization and deepening of collective trajectories with an internal reflective structure.Higher-level metacognitive actions are critical for collaborative inquiry and CEAAnalytics-assisted RA can enhance students' higher-level metacognitive actions.Students should be provided with learning data and appropriate scaffolding strategies.These data and strategies can develop students' internal metacognitive structure crucial for agency and learning.