| 545 | 0 | 109 |
| 下载次数 | 被引频次 | 阅读次数 |
高阶能力的发展已成为学生适应未来社会与实现终身发展的核心素养与关键支撑。利用大语言模型支持高阶能力发展被视为一种有效路径。然而，当前单一的大语言模型存在可解释性弱、稳定性不足等内在局限，在系统化、深层次的高阶能力培养方面仍面临挑战。为此，本研究基于ARCS动机理论，提出一种多大语言模型(LLMs)多阶段融合赋能的智能对话式学习框架，将不同大语言模型有机融入学习活动设计与任务实施过程，从多个维度考察LLMs对学生高阶能力发展的促进作用。研究结果表明：(1)LLMs多阶段融合支持下的学习活动能够整体显著提升学生的高阶能力与学习绩效；(2)LLMs介入增强了学生在学习活动中的社会情感能力，并在延迟测试中保持了学习绩效的显著提升；(3)学习动机、任务复杂度与交互强度是预测高阶能力发展的显著因素。此外，本研究从LLMs多阶段融合激发学习动机、强化认知投入、适配任务复杂度以及增强交互强度四个维度，剖析了该框架赋能高阶能力发展的作用机理，为推动教育高质量发展、实现从知识传递向能力塑造的育人范式变革提供了理论支撑与实践进路。
Abstract：The development of higher-order skills has become a core competency and a key foundation for students to adapt to future society and achieve lifelong development. Leveraging large language model(LLM) to support the development of higher-order skills is regarded as an effective pathway. However, single LLM, due to inherent limitations such as weak interpretability and insufficient stability, still face challenges in facilitating systematic and in-depth cultivation of higher-order skills. To address this issue, this study, grounded in the ARCS motivation theory, proposes an intelligent dialogic learning framework empowered by multi-stage integration of multiple LLMs. The framework systematically embeds different LLMs into the design of learning activities and the implementation of tasks, and examines their effects on the development of higher-order skills from multiple dimensions. The results show that:(1) learning activities supported by multi-stage LLMs integration significantly improve students' higher-order skills and learning performance overall;(2) the integration of LLMs enhances students' socio-emotional abilities during learning activities and sustains significant improvement in learning performance in delayed tests; and(3) learning motivation, task complexity, and interaction intensity are significant predictors of the development of higher-order skills. Furthermore, this study analyzes the underlying mechanisms through which the framework facilitates the development of higher-order skills from four dimensions: stimulating learning motivation through multi-stage LLM integration, strengthening cognitive engagement, aligning task allocation with task complexity, and enhancing interaction intensity. This research provides theoretical support and practical pathways for promoting high-quality education and advancing the transformation from knowledge transmission to competency-oriented education.
[1]彭正梅，吴月竹，陈丽莎.高质量发展要求培养高阶能力：对教育强国的共同考察[J].全球教育展望，2024(3):144-160.
[2]孙建文.AI4LS:智能时代学习科学研究新范式[J].华中师范大学学报(人文社会科学版),2025(3):154-162.
[3]李梦兴，刘妍.AI智能体能提升学习效果吗?——问题导向思维视角下学生深度问题解决能力发展效果研究[J].现代教育技术，2025(10):42-51.
[4]王一岩，刘淇，郑永和.对话式人机协同学习：本质内涵与未来图景[J].中国电化教育，2024(11):21-27.
[5]刘永贵，眭雨婷.从支持工具到学习主体：“GenAI-学习者”协作学习系统的模型构建与促进策略[J].现代教育技术，2025(4):53-61.
[6]严奕峰，丁杰，高赢，等.生成式人工智能赋能数字时代育人转型[J].开放教育研究，2024(2):42-48.
[7]Farquhar S,Kossen J,Kuhn L,et al.Detecting hallucinations in large language models using semantic entropy[J].Nature,2024(8017):625-630.
[8]陈向东，卢淑怡，易乐湘.文化冲突：大语言模型教育应用中的张力与调适[J].远程教育杂志，2025(3):3-15+43.
[9]郝俞植，谢丹阳.基于多个 LLM 代理的经济和公共政策分析框架[J].计量经济学报，2025(3):615-630.
[10]王晶莹，周丹华，杨洋，等.科学高阶思维：内涵价值、结构功能与实践进路[J].现代远距离教育，2023(2):11-18.
[11]胡小勇，孙硕，杨文杰，等.人工智能赋能：学习者高阶思维培养何处去[J].中国电化教育，2022(12):84-92.
[12]Liu J,Liu Z,Wang C,et al.Key factors and mechanisms affecting higher-order thinking skills of primary and secondary school students in the smart classroom environment[J].Current Psychology,2024(11):9651-9664.
[13]Lu K,Yang H H,Shi Y,et al.Examining the key influencing factors on college students’ higher-order thinking skills in the smart classroom environment[J].International Journal of Educational Technology in Higher Education,2021(1):1-13.
[14]Lu K,Pang F,Shadiev R.Understanding the mediating effect of learning approach between learning factors and higher order thinking skills in collaborative inquiry-based learning[J].Educational Technology Research and Development,2021(5):2475-2492.
[15]姜强，刘盼，倪静，等.基于时间序列数据驱动的在线学业预测机理模型研究[J].现代远距离教育，2025(1):58-67.
[16]王志军，滕志强，苏晨予.国际生成式人工智能教育应用创新——全球人工智能学习和教育研究联盟之“人工智能赋能学习者会议”综述[J].远程教育杂志，2024(2):65-74.
[17]张治.ChatGPT/生成式人工智能重塑教育的底层逻辑和可能路径[J].华东师范大学学报(教育科学版),2023(7):131-142.
[18]郑永红，王辰飞，张务伟.生成式人工智能教育应用及其规制[J].中国电化教育，2024(5):114-119.
[19]曹培杰，谢阳斌，武卉紫，等.教育大模型的发展现状、创新架构及应用展望[J].现代教育技术，2024(2):5-12.
[20]李艳，许洁，贾程媛，等.大学生生成式人工智能应用现状与思考——基于浙江大学的调查[J].开放教育研究，2024(1):89-98.
[21]Song C,Song Y.Enhancing academic writing skills and motivation:assessing the efficacy of ChatGPT in AI-assisted language learning for EFL students[J].Frontiers in Psychology,2023,14:1260843.
[22]Yang L,Li R.ChatGPT for L2 learning:Current status and implications[J].System,2024,124:103351.
[23]Chen C H,Chang C L.Effectiveness of AI-assisted game-based learning on science learning outcomes,intrinsic motivation,cognitive load,and learning behavior[J].Education and Information Technologies,2024(14):18621-18642.
[24]Keller J M.Development and use of the ARCS model of instructional design[J].Journal of Instructional Development,1987(3):2-10.
[25]李卢一，郑燕林.ARCS模型视角下创客项目设计研究[J].现代远距离教育，2018(2):12-19.
[26]师亚飞，童名文，孙佳，等.混合同步学习环境对学生认知投入的影响机制研究[J].中国远程教育，2021(9):29-38+68+77.
[27]Team K,Bai T,Bai Y,et al.Kimi K2.5:Visual Agentic Intelligence[J].arXiv preprint arXiv:2602.02276,2026.
[28]Wilujeng I.Increasing the Attention,Relevance,Confidence and Satisfaction (ARCS) of Students through Interactive Science Learning Multimedia[J].Research in Learning Technology,2021,29:1-13.
[29]Ryan R M,Deci E L.Self-determination theory and the facilitation of intrinsic motivation,social development,and well-being[J].American psychologist,2000(1):1-11.
[30]Liu A,Feng B,Xue B,et al.Deepseek-v3 technical report[J].arXiv preprint arXiv:2412.19437,2024.
[31]Salehi M,Abolghasemi M.Investigating the effect of the collaborative teaching method of jigsaw on the critical thinking of the sixth grade students of Najaf Abad city[J].Thinking and Children,2024(1):141-173.
[32]Zeng A,Lv X,Hou Z,et al.Glm-5:from vibe coding to agentic engineering[J].arXiv preprint arXiv:2602.15763,2026.
[33]卢雨，吴应辉.七种大模型在国际中文教育知识考核中的实验比较[J].海南师范大学学报(社会科学版),2025(3):77-87.
[34]Kaliisa R,Misiejuk K,López-Pernas S,et al.How does artificial intelligence compare to human feedback?A meta-analysis of performance,feedback perception,and learning dispositions[J].Educational Psychology,2026(1):80-111.
[35]Liebenow L W,Schmidt F T C,Meyer J,et al.Self-assessment accuracy in the age of artificial Intelligence:Differential effects of LLM-generated feedback[J].Computers & Education,2025,237:105385.
[36]黄昌勤，钟益华，王希哲，等.从单智能体到多智能体：大模型智能体支持下的激励型学习活动设计与实证研究[J].华东师范大学学报(教育科学版),2025(5):44-56.
[37]Barak M,Watted A,Haick H.Establishing the validity and reliability of a modified tool for assessing innovative thinking of engineering students[J].Assessment & Evaluation in Higher Education,2020(2):212-223.
[38]Hwang G J,Chen C H.Influences of an inquiry-based ubiquitous gaming design on students’ learning achievements,motivation,behavioral patterns,and tendency towards critical thinking and problem solving[J].British journal of Educational Technology,2017(4):950-971.
[39]王才康，胡中锋，刘勇.一般自我效能感量表的信度和效度研究[J].应用心理学，2001(1):37-40.
[40]Yang T C,Lin Z S.Enhancing elementary school students’ computational thinking and programming learning with graphic organizers[J].Computers & Education,2024,209:104962.
[41]Qi X I A,Weijia L I,Yiming Y,et al.A Systematic Review and Meta-Analysis of the Effectiveness of Generative Artificial Intelligence (GenAI) on Students’ Motivation and Engagement[J].Computers and Education:Artificial Intelligence,2025:100455.
[42]李曼丽，乔伟峰，李睿淼.大语言模型工具能促进高校学生的高阶思维能力发展吗?——基于12所“双一流”大学学生问卷调查的实证分析[J].现代教育技术，2025(1):34-43.
[43]李海峰，王炜.人机争论探究法：一种争论式智能会话机器人支持的学生高阶思维能力培养模式探索[J].电化教育研究，2024(3):106-112+128.
[44]和文斌，赵帅，阿不来提·瓦依提，等.基于生成式人工智能的人机协同学习更能提升学习成效?——基于20项实验和准实验的元分析[J].开放教育研究，2024(5):101-111.
[45]郝祥军，顾小清，柏宏权.人机何以共情：基于人机交往的情感连接与伦理反思[J].现代远距离教育，2025(3):79-87.
基本信息:
DOI：10.13927/j.cnki.yuan.20260430.006
中图分类号:TP18;G434
引用信息:
[1]姜强,解荞旭,刘盼,等.LLMs多阶段融合赋能高阶能力发展的智能对话式学习框架设计及作用机理研究[J].现代远距离教育,2026,No.223(01):35-46.DOI:10.13927/j.cnki.yuan.20260430.006.
基金信息:
国家自然科学基金面上项目“具身认知下大学生深度学习的多维生成式诊断及人智协同干预机制研究”(编号：62577018); 吉林省社会科学基金项目“吉林省高等教育数字化转型路径及对策研究”(编号：2023B88);吉林省社会科学基金项目“数据驱动的吉林省中小学教师数字素养转化效能增值评价研究”(编号：2025B90); 2026年吉林省教育厅社科重点项目“人工智能赋能教师课堂教学成效的增值评价与优化路径研究”(编号：JJKH20260351SK)
2026-04-30
2026-04-30
2026-04-30