llm_wiki/raw/教育AI研究/outputs/pdf_contents.json

{
  "Celik 等 - 2026 - Co-constructing adaptive lesson plans with GenAI Pre-service teachers' Intelligent-TPACK and prompt": {
    "pages": 16,
    "author": "Ismail Celik",
    "content": "Co-constructing adaptive lesson plans with GenAI: Pre-service \nteachers ’ Intelligent-TPACK and prompt engineering strategies\nIsmail Celika,*\n, Sini Kontkanenb\n, Jari Laruc\n, Alanur Ahsen Dalyancic\naAcademy Research Fellow, Faculty of Education and Psychology, University of Oulu, FI-90014, Oulu, Finland\nbUniversity of Eastern Finland, Faculty of Philosophy, Joensuu, Finland\ncFaculty of Education and Psychology, University of Oulu, FI-90014, Oulu, Finland\nARTICLE INFO\nKeywords:\nAdaptive learning\nTeacher education\nIntelligent-TPACK\nPrompt engineeringABSTRACT\nGenerative Artificial Intelligence (GenAI) technologies present new opportunities for teachers to \ndesign adaptive and student-centered instruction. However, the educational value of GenAI de-\npends not only on technical usage but also o | the advent of Generative Artificial Intelligence (GenAI) technologies, it has become increasingly feasible for teachers to design more \npersonalized and responsive learning experiences (Hsia et al., 2025 ). This is because GenAI tools can support differentiation, scaf-\nfolding, and learner agency (Yang & Markauskaite, 2025 ). However, teachers have a responsible role in the ethical and pedagogical use \nof GenAI technologies in designing adaptive teaching practices (Celik, 2023 ; Henderson et al., 2025 ; Tagare et al., 2025 ).\nGenAI tools such as ChatGPT, CoPilot, and Google Gemini, which are now widely used by teachers and students, are considered \nintelligent systems (Ariza et al., 2025 ; Guggemos, 2024 ; Hong et al., 2025 ). For teachers to effectively integrate these tools into their \np | In sum, teachers should consider student agency, scaffolding strategies, and technology use as flexible tools when designing \nadaptive learning environments. These three elements work together to ensure that learners are both supported and empowered, while \ninstruction remains responsive to diverse needs. The current study addresses adaptive learning through these concepts by examining \nhow lesson plans incorporate opportunities for student choice, the presence of scaffolding strategies, and the integration of flexible \ntechnological tools.\n2.2. Intelligent-TPACK\nThe Intelligent TPACK framework is a contemporary extension of the TPACK framework developed by Mishra and Koehler (2006). \nIt was proposed in response to the growing integration of AI technologies into educational settings. This "
  },
  "Ferreira和Ineson - 2026 - Embodied learning in teacher education Investigating student-teachers' experiences in engaging with": {
    "pages": 14,
    "author": "Juliene Madureira Ferreira",
    "content": "Research paper\nEmbodied learning in teacher education: Investigating student-teachers ’ \nexperiences in engaging with embodied cognition theories\nJuliene Madureira Ferreiraa,*\n, Gwen Inesonb\naFaculty of Education and Culture, Tampere University, Tampere, Finland\nbDepartment of Education, College of Business, Art and Social Sciences, Brunel University London, London, United Kingdom\nARTICLE INFO\nKeywords:\nEmbodied cognition\nStudent-teachers ’ experiences\nEmbodied concept learning\nEnactive cognition theory\nQualitative methodABSTRACT\nThis study examines how integrating perceptual, sensorimotor, and reflective processes supports the learning of \nenactive cognition theory, demonstrating how student-teachers develop competencies for engaging with theory \nthrough lived, embodied experiences. We em | models that explain learning processes from an embodied perspective) \nand as a pedagogical approach (i.e., pedagogies that promote embodied \nlearning), and developing methods for assessing the construction of \nembodied knowledge.\nWhile there is a continuous interest in embodied learning ap-\nproaches, student-teachers are not necessarily exposed to such \nembodied practices or encouraged to learn about theories that address \nembodiment in cognitive processes during their education and training. \nCurrent literature indicates that, firstly, most research on embodiment \nin learning is conducted in school contexts, targeting children and ad-\nolescents, and focusing on explicating correlations between movement \n(e.g., teachers’ or students’ gesturing) and learning (Hegna and Ørbæk, \n2021 ; Jussli | knowledge (Alexander et al., 2011 ; Rustici, 1997 ), fostering a mean -\ningful approach to enhance competencies and skills applicable across \ndifferent fields of knowledge and educational levels (Gross & Rutland, \n2017 ). Reflection plays a crucial role in transforming experience into \nmeaningful understanding. It is part of the dynamic interplay between \nbody, brain, and environment, serving as the connecting tissue between \ndoing (experiencing) and knowing (Laner, 2021 ). Reflection must also \nbe understood as an embodied process — not merely a cognitive exercise \ndetached from the body, but one that arises through and with bodily \nexperience (Gallagher, 2023 ; Kinsella, 2007 ). It serves as the bridge that \nconnects embodied action with conceptual understanding, enabling \nlearners to ma"
  },
  "Guo 等 - 2025 - One year in the classroom with ChatGPT empirical insights and transformative impacts": {
    "pages": 17,
    "author": "Feng Guo",
    "content": "feduc-10-1574477 May 22, 2025 Time: 18:24 # 1\nTYPE Original Research\nPUBLISHED 27 May 2025\nDOI10.3389/feduc.2025.1574477\nOPEN ACCESS\nEDITED BY\nXiaoxun Sun,\nAustralian Council for Educational Research,\nAustralia\nREVIEWED BY\nAlexa Alice Joubin,\nThe George Washington University,\nUnited States\nRahul Joshi,\nManav Rachna International Institute\nof Research and Studies (MRIIRS), India\nMary Liz Brooks,\nWest Texas A&M University, United States\n*CORRESPONDENCE\nFeng Guo\nfeng-guo@utc.edu\nRECEIVED 20 February 2025\nACCEPTED 02 May 2025\nPUBLISHED 27 May 2025\nCITATION\nGuo F, Li T and Cunningham CJL (2025) One\nyear in the classroom with ChatGPT:\nempirical insights and transformative\nimpacts.\nFront. Educ. 10:1574477.\ndoi: 10.3389/feduc.2025.1574477\nCOPYRIGHT\n© 2025 Guo, Li and Cunningham. This is an\nopen-ac | feduc-10-1574477 May 22, 2025 Time: 18:24 # 2\nGuo et al. 10.3389/feduc.2025.1574477\nAlthough there is growing interest in the use of GAI in education,\nthere is a great need for more empirical studies that discuss\nits adoption and impact (Farrokhnia et al., 2024). In particular,\nsystematic evaluations of the effectiveness of GAI-based classroom\nactivities remain scarce. Such evaluations should carefully consider\nfactors like pedagogical formats, audience demographics, and\ndisciplinary differences. To help address this gap, we present\nthe findings from a year-long experiment conducted in 2023,\nwhere the authors incorporated ChatGPT into college classroom\nteaching across various subjects and formats, involving both\nundergraduate and graduate students. This teaching experiment\nspans two semesters | feduc-10-1574477 May 22, 2025 Time: 18:24 # 3\nGuo et al. 10.3389/feduc.2025.1574477TABLE 1 Empirical studies of use of ChatGPT in education.\nEducation level\nTopics Graduate-level Undergraduate K-12 Unspecified\nTools to Support\nTeachers/studentsAgarwal et al. (2023) Bartoli et al. (2024) Lower et al. (2023) Parker et al. (2023) Scherr et al. (2023)\nTotlis et al. (2023) Ngo et al. (2024) Guleria et al. (2023) Wandelt et al. (2023)Cowling et al.\n(2023) de\nVicente-Y agüe-\nJara et al. (2023)\nGhafouri (2024)\nJeon and Lee\n(2023) Leite\n(2023) Meron\nand Araci (2023)\nPodlasov and\nMatviichuk\n(2023) Shue\net al. (2023) Tlili\net al. (2023)\nVeras et al.\n(2023)Y an (2023)\nImran and\nAlmusharraf\n(2023) Lee et al.\n(2024)\nRuiz-Rojas et al.\n(2023) de Winter\net al. (2023) Niu\nand Xue (2023)\nWang et al.\n(2024) Yo"
  },
  "Lee - 2025 - Virtual internships as alternative work-based learning Examining access, quality, and outcomes for": {
    "pages": 15,
    "author": "Changhee Lee",
    "content": "Virtual internships as alternative work-based learning: Examining \naccess, quality, and outcomes for underserved students☆,☆☆\nChanghee Lee*\nDepartment of Leadership, Policy, and Organizations, Vanderbilt University, Nashville, TN, USA\nABSTRACT\nThis study examines whether virtual internships serve as a viable alternative to in-person work-based learning opportunities, and, if so, for whom \nand in which specific dimensions. Drawing on a large-scale survey (Nˆ11,009) and administrative data collected from 17 U.S. postsecondary in-\nstitutions during the 2020-2021 academic year, the research compares the experiences of historically marginalized students in virtual and in-person \ninternships to those of their non-marginalized counterparts across access, program quality, and outcomes. Findings fr | internships ’ critical role in shaping future career trajectories.\nThe advent of virtual internships —digitally mediated WBL experiences conducted remotely (Hora et al., 2021 )—has sparked \nscholarly discourse on their potential to mitigate structural inequalities in traditional, place-based internships. Proponents argue that \nthe flexibility of digital platforms may democratize access (e.g., Kraft et al., 2019 ; Reid et al., 2023 ), particularly for marginalized \npopulations who face barriers such as relocation costs, unpaid positions, limited professional networks, and unfamiliar workplace \nnorms. Indeed, federal initiatives like Virtual Student Federal Service Internships explicitly target “students who might not otherwise \nbe able to participate in an in-person [federal service] intern | distances and logistical challenges (The White House, 2021 ). Indeed, emerging evidence indicates that students who have faced longer \ncommutes to traditional worksites report higher satisfaction with virtual internship arrangements (Januszewski & Grzeszczak, 2021 ). \nCollectively, these shifts may broaden participation for both students and hosts, expanding the scale and diversity of internship \nopportunities.\nThese accessibility gains may extend into recruitment practices as well. Virtual internships often utilize open, digitized platforms \nthat can theoretically reach a broader applicant pool than conventional campus-based recruiting or referral networks (Jeske & Axtell, \n2014 ), which tend to favor students with privileged social and cultural capital (Rivera, 2016 ). These entrenched n"
  },
  "Riofrío-Luzcando 等 - 2026 - Comparing automaton-based approach with machine learning models for predicting student errors in pro": {
    "pages": 15,
    "author": "Diego Riofrío-Luzcando",
    "content": "Contents  lists available  at ScienceDirect\nExpert  Systems  With  Applications\njournal  homepage:  www.elsevier.com/locate/eswa\nComparing  automaton-based  approach  with  machine  learning  models  for \npredicting  student  errors  in procedural  training  to support  intelligent  \ntutoring  systems  \nDiego  Riofrío-Luzcando\na,∗, Jaime  Ramírez\nb, Marta  Berrocal-Lobo\nc\naQuantitative  Methods Department,  CUNEF Universidad,  Calle Pirineos 55, Madrid, 28040, Madrid, Spain\nbCenter for Biomedical  Technology,  Universidad  Politécnica  de Madrid, Campus de Montegancedo,  Pozuelo de Alarcón, 28223, Madrid, Spain\ncEscuela Técnica Superior de Ingeniería  de Montes, Forestal y del Medio Natural, Universidad  Politécnica  de Madrid, C. de José Antonio Novais, 10, Madrid, 28040, \nMadrid, Spain\na | D. Riofrío-Luzcando  et al.\nThe remainder  of this paper is organized  as follows.  Section  2 re-\nviews relevant  works in the application  of ML to e-learning.  Section  3 \ndescribes  the dataset  used to evaluate  the prediction  models.  Section  4 \nexplains  how the prediction  model would be incorporated  in the tutor-\ning strategy.  Section  5 outlines  the methodology  adopted  to explore  the \npredictive  performance  of the models.  Section  6 presents  the experimen-\ntal results.  Section  7 offers a discussion  of the findings.  Finally,  Section  8 \nconcludes  the paper and proposes  directions  for future research.\n2.  Related  work\nThe related work is divided  into two sections.  Section  2.1 briefly \npresents  some key results of ML applied  to e-learning,  while Section  2.2  | D. Riofrío-Luzcando  et al.\n–Incompatibility  Errors\n–World Errors\n–Other Errors.\nCorrect  events are right actions  according  to the protocol  of the prac-\ntical assignment.  Dependency  or incompatibility  errors depend  on the \nconfiguration  of the virtual laboratory  (detailed  in Rico et al. (2012 )) \nset up by the instructor.  They are related to the right order in which to \nperform  the actions  in the practical  assignment.  World errors refer to \nfailures  in the handling  of 3D objects;  for example,  if a student  tries to \ndrop an object where it should not be dropped.  Finally,  the other error \nevents category  represents  errors that are not pedagogically  relevant;  for \nexample,  if the student  tries to repeat an action that has already  been \nperformed.\nAdditionally,  e"
  },
  "Shi 等 - 2026 - Large language models in education a systematic review of empirical applications, benefits, and cha": {
    "pages": 16,
    "author": "Yuhong Shi",
    "content": " Contents lists available at ScienceDirect\nComputers and Education: Artificial Intelligence\njournal homepage: www.sciencedirect.com/journal/computers-and-education-artificial-intelligence  \nLarge language models in education: a systematic review of empirical \napplications, benefits, and challenges\nYuhong Shi  ∗  iD, Kun Yu, Yifei Dong, Fang Chen\nData Science Institute, Faculty of Engineering and Information Technology, University of Technology Sydney, Ultimo, NSW 2007, Australia\nH I G H L I G H T S\n•  Reviews 88 empirical studies on LLM applications in education, selected from 3344 publications since ChatGPT’s release (Nov 2022–Mar 2025).\n•  Identifies six key LLM applications, with Intelligent Tutoring Systems being the most common.\n•  Empirical evidence shows that LLMs enhance academic p | Y. Shi, K. Yu, Y. Dong et al.\ntheir progress, and reflect on their understanding through personalized \nfeedback and metacognitive prompts ( Fan et al. , 2025 ). Concurrently, \nLLMs align with both Cognitive Load Theory ( Sweller , 1988 ) and the \nZone of Proximal Development (ZPD) ( Vygotsky , 1978 ) through their \nadaptive capabilities. Specifically, Vygotsky ’s ZPD theory conceptual­  \nizes the gap between what learners can accomplish independently and \nwhat they can achieve with guidance, while Sweller ’s Cognitive Load \nTheory posits that learning effectiveness depends on how instructional \ndesign manages the limited capacity of working memory by balancing \nintrinsic, extraneous, and germane cognitive load. LLM-integrated sys­  \ntems adjust response complexity and break down intricate  | Y. Shi, K. Yu, Y. Dong et al.\nTable 1  \nSummary of related systematic review studies on LLMs in education.\nCitation  Domain  Coverage period  Contributions\nChatGPT in English Language Teaching (ELT) ( Adipat , \n2025 )  ELT  2020–2024  ChatGPT’s opportunities, challenges, and ethical \nconsiderations.\nLLM in Higher Education ( Chhina et al. , 2023 ) HE 2018–2023 Benefits and challenges of LLMs in higher \neducation.  \nChatGPT in ELT ( Wang, Hanafi Zaid, et al. , 2024 )  ELT  2023–2024  Opportunities, challenges, and trends in applying \nChatGPT in ELT.  \nOpen-Source LLMs in Education ( Lin et al. , 2024 )  General Education  2023–2024  Open-source LLMs and their suitability for ed­\nucational applications in English-speaking \ncontexts.  \nLLMs in Medical Education ( Lucas et al. , 2024 )  Medica"
  },
  "Sinha - 2026 - Making failure desired during learning – A quasi-experimental study": {
    "pages": 17,
    "author": "Tanmay Sinha",
    "content": "Making failure desired during learning – A \nquasi-experimental study\nTanmay Sinha\nNational Institute of Education, Nanyang Technological University, Singapore 637616\nARTICLE INFO\nKeywords:\nfailure\ngrowth mindset\nutility value\nmixed-methods analysisABSTRACT\nOne hundred and nineteen ninth-grade students engaged in one of two preparatory interventions \n– growth mindset or utility value – aimed at increasing the desirability of failure in learning \nduring a quasi-experimental study. An additional fifty-one students participated in a control \ncondition that offered no such preparation. Everyone then underwent a standard productive \nfailure learning task where they ideated to solve an open-ended math problem prior to receiving a \nformal lecture on the targeted concept. Following mixed-methods an | 2.Theoretical background\n2.1. Productive failure learning context\nProductive failure, conceptualized by Kapur & Bielaczyc (2012) , is a constructivist two-phase learning design that engages novices \nin open-ended problem-solving on a yet-to-be-learned concept before providing canonical instruction. Students activate their prior \nknowledge to typically generate multiple suboptimal solutions, and during that exploratory process, begin recognizing gaps in their \nknowledge, which can make them more prepared to learn from follow-up instruction (Loibl et al., 2017 ). Supportive social norms and \nscaffolds such as motivation to persist without the fear of failing are commonplace during the initial problem-solving phase of \nwell-designed productive failure. Follow-up instruction typically builds o | Previous student-focused growth mindset interventions in math (Bui et al., 2023 ) have exhibited the following limitations – \nspecifically, they have (i) utilized learning materials that are either entirely domain-general (e.g., information on brain function and \nthe strengthening of neural connections through failure) or domain-specific (e.g., how beliefs about math can influence perceptions of \nfailure), (ii) employed direct teaching and/or reading exercises to educate students on these topics, which may be detrimental to those \nwith low success expectations, (iii) focused primarily on quantitative measures to evaluate impact, leaving a gap in qualitatively \nunderstanding how these interventions shape students ’ mindsets towards failure, and (iv) been mainly implemented in American or \nE"
  },
  "Wenzel 等 - 2026 - Designing conversational Agents for adaptive instructional support in business simulation gaming": {
    "pages": 21,
    "author": "Anna Wenzel",
    "content": "Designing conversational Agents for adaptive instructional support in \nbusiness simulation gaming\nAnna Wenzela,*\n, Jan-Martin Geigerb, Andreas Lieninga\naFaculty of Business and Economics, Professorship of Entrepreneurship and Economic Education, TU Dortmund University, Friedrich-W ohler-Weg 6, 44227, Dortmund, \nGermany\nbJunior-Professorship of Innovation and Transfer of Digital Teaching, University of Münster, Roggenmarkt 15, 48143, Münster, Germany\nARTICLE INFO\nKeywords:\nDigital game-based learning\nConversational agent\nArtificial intelligence\nDesign knowledge\nBusiness simulation games\nUniversal design for learningABSTRACT\nAdaptive instructional support that addresses individual learner differences in learning strengths, challenges, and \ninterests is essential for engaging learners in mea | Design for Learning (UDL; Cast, 2018 ), instructional design should \nprovide multiple means of representation, action and expression as well \nas engagement. Adhering to these principles has been shown to enhance \nlearning processes for all learners (Almeqdad et al., 2023 ; Capp, 2017 ) \nand aligns with an equity-by-design perspective that emphasizes pro-\nactively removing learning barriers and preventing digital learning en-\nvironments from reinforcing existing disparities, for example related to \naccess, prior knowledge, skills, and language proficiency (Gottschalk & \nWeise, 2023 ). Providing such equitable learning opportunities demands \ntailored instructional support, including timely guidance, scaffolding, \nand formative feedback, which are essential for engaging learners with \ndiverse | most relevant information; and encouraging germane processing, which \npromotes active engagement and deeper cognitive investment (Mayer & \nMoreno, 2003 ). Therefore, we propose MR1: Contribute to meaningful \ncognitive engagement in DGBL by minimizing extraneous cognitive load \n(MR1.1), managing intrinsic cognitive load (MR1.2) and fostering germane \ncognitive load (MR1.3) .\nWe ground the motivational perspective of DGBL in Self-Determina -\ntion Theory (SDT) (Deci et al., 1991 ; Ryan & Deci, 2000 ). According to \nSDT, intrinsic motivation flourishes in environments that fulfill three \nbasic psychological needs: competence, which is the sense of effectively \nachieving internal or external goals; relatedness, which is the experience \nof secure and meaningful connections with others; and auton"
  }
}