Artificial intelligence (AI) tutors have become increasingly popular in learning environments. In this study, we propose an AI agent prototype framework for exploring AI-assisted learning with temporal interaction patterns, multiple outcomes analysis, and behavioral-cognitive learner profiling. Based on three research questions, this study aims to investigate whether early interaction patterns can predict later performance and trust, how multiple outcomes can be traded off with different AI tutor feedback conditions, and if learner profiles can be identified with behavioral and cognitive indicators. An AI tutor agent has been developed to provide various feedback forms to learners, including hints, explanations, examples, and code. A neural policy model and a stochastic simulation framework are used to produce artificial student-AI tutor interaction records, which include response time, attempts, hint requests, correctness, quiz results, improvement, satisfaction, and trust. Temporal features are used to predict later correctness and trust with early interaction patterns, and clustering methods are used to find learner profiles. The results showed that early interaction patterns were predictive of later performance and trust, that student behavior changed over time with AI-based tutoring, and that latent student profiles could be identified based on their behavioral and cognitive differences.
With the proliferation of simple and complex bioinformatics tools, there is the need to teach researchers how to use these tools effectively. To evaluate the potential of cognitive tutoring in the wide-scale adoption of several bioinformatics tools, we designed a simple prototype. We embedded a cognitive tutor, built with the Cognitive Tutor Authoring Tool, on a preexisting platform, the Gene Adjacency Program, developed by the University of Ibadan Bioinformatics group. Our preliminary tests show that researchers who used the platform with the cognitive tutor embedded showed higher levels of c…
Beyond Helpfulness: A Teaching-over-Solving Diagnostic for Measuring Educational Impact in LLM Tutors
Washington University in St. Louis（圣路易斯华盛顿大学）；Department of Operations Research and Engineering Management, Southern Methodist University（南卫理公会大学运筹学与工程管理系）
Million Tutoring Moves (MTM): An Open Multimodal Dataset for the Science of Tutoring
National Tutoring Observatory（国家辅导观测站）；Cornell University（康奈尔大学）；FreshCognate；Massachusetts Institute of Technology（麻省理工学院）；Carnegie Mellon University（卡内基梅隆大学）
From Earthquake Solidarity to Educational Equity: Conceptualizing a Sustainable, Volunteer-Driven P2P Learning Ecosystem at Scale
Gdańsk University of Technology（格但斯克理工大学）；University of Galway（戈尔韦大学）；University of Applied Sciences and Arts Hannover（汉诺威应用科学与艺术大学）；University of Oulu（奥卢大学）
Optimizing Student Ability Assessment: A Hierarchy Constraint-Aware Cognitive Diagnosis Framework for Educational Contexts
University of Science and Technology of China（中国科学技术大学）；Liupanshui Normal University（六盘水师范学院）；Hefei Comprehensive National Science Center（合肥综合性国家科学中心）；National University of Singapore（新加坡国立大学）
Computational thinking is a way of reasoning about the world in terms of data. This mindset channels number crunching toward an ambition to discover knowledge through logic, models and simulations. Here we show how computational cognitive science can be used to reconstruct and analyse the structure of computational thinking mindsets (forma mentis in Latin) through complex networks. As a case study, we investigate cognitive networks tied to key concepts of computational thinking provided by: (i) 159 high school students enrolled in a science curriculum and (ii) 59 researchers in complex systems…
Centre for Artificial Intelligence, University College London（伦敦大学学院人工智能中心）；Jožef Stefan Institute（约瑟夫·斯特凡研究所）；Jožef Stefan International Postgraduate School（约瑟夫·斯特凡国际研究生院）
Educational recommenders have received much less attention in comparison to e-commerce and entertainment-related recommenders, even though efficient intelligent tutors have great potential to improve learning gains. One of the main challenges in advancing this research direction is the scarcity of large, publicly available datasets. In this work, we release a large, novel dataset of learners engaging with educational videos in-the-wild. The dataset, named Personalised Educational Engagement with Knowledge Topics PEEK, is the first publicly available dataset of this nature. The video lectures h…
ZPD-SCA: Unveiling the Blind Spots of LLMs in Assessing Students' Cognitive Abilities
School of Psychology, South China Normal University（南方科技大学心理学院）；Information Hub, Hong Kong University of Science and Technology (Guangzhou)（香港科技大学（广州）信息中心）；School of AI, Guangzhou University（广州大学人工智能学院）；College of Cyber Security, Jinan University（济南大学网络安全学院）
