| 186 | 0 | 55 |
| 下载次数 | 被引频次 | 阅读次数 |
该研究聚焦当前人才选拔过程中对创造性思维与真实问题解决能力的评估需求,构建了一套基于先进计算技术的智能化测评方法。该方法以心理学理论为基础,融合人工智能大语言模型(Large Language Models, LLM)技术,涵盖题目筛选、数据标注与算法建模等关键环节。研究通过对3226道中学考试题目的系统分析,揭示了现有评价体系在创造力各维度的覆盖不足;基于2867名学生的6575份答卷数据,开发并验证了自动评分算法,结果显示在多个评分维度上的人机一致性相关系数达0.60以上,具备可靠的评分效能。研究发现,学生的认知能力与创造力呈显著正相关(r=0.44, p<0.001),但两者匹配类型多样,印证了单一认知测验难以全面识别创新潜能的现实挑战。本研究不仅在理论上整合了认知心理学与计算语言学的交叉视角,也在实践上为教育评价智能化变革与拔尖创新人才早期识别提供了初步可行的技术路径与实证依据,对推动素质教育评价体系改革具有重要启示意义。
Abstract：This study addresses the critical demand for assessing creative thinking and real-world problem-solving abilities in current talent selection processes by constructing an intelligent assessment method grounded in advanced computational technologies.Integrating psychological theories with Large Language Models(LLMs), the method encompasses key components including item screening, data annotation, and algorithmic modeling.Through a systematic analysis of 3,226 secondary school examination items, this research reveals significant deficiencies in the coverage of various creativity dimensions within existing evaluation frameworks.Furthermore, an automated scoring algorithm was developed and validated based on 6,575 responses from 2,867 students.The results demonstrate reliable scoring efficacy, with human-AI consistency correlation coefficients exceeding 0.60 across multiple scoring dimensions.The study also identifies a significant positive correlation between students' cognitive ability and creativity(r=0.44, p<0.001).However, the diversity in alignment patterns between these two constructs corroborates the practical challenge that single cognitive tests are insufficient for comprehensively identifying innovative potential.Theoretically, this research integrates crossdisciplinary perspectives from cognitive psychology and computational linguistics.Practically, it provides an initial feasible technical pathway and empirical evidence for the intelligent transformation of educational evaluation and the early identification of high-potential innovative talent, offering significant implications for reforming quality-oriented education assessment systems.
[1]庞维国.创造性心理学视角下的创造性培养:目标、原则与策略[J].华东师范大学学报(教育科学版),2022,40(11):25-40.
[2]RUNCO M A,JAEGER G J.The standard definition of creativity[J].Creativity Research Journal,2012,24(1):92-96.
[3]KAUFMAN J, BEGHETTO R. Beyond big and little:The four C model of creativity[J].Review of General Psychology,2009,13(1):1-12.
[4]WALLAS G.The art of thought[M].New York:Harcourt,Brace,1926.
[5]MUMFORD M D,MCINTOSH T.Creative thinking processes:The past and the future[J].The Journal of Creative Behavior,2017,51(4):317-322.
[6]AMABILE T M.Social psychology of creativity:A consensual assessment technique[J].Journal of Personality and Social Psychology,1982,43(5):997-1013.
[7]BAER J,KAUFMAN J C,et al.Extension of the consensual assessment technique to nonparallel creative products[J].Creativity Research Journal,2004,16(1):113-117.
[8]PLUCKER J A,QIAN M,et al.Is what you see what you really get?Comparison of scoring techniques in the assessment of real-world divergent thinking[J].Creativity Research Journal,2014,26(2):135-143.
[9]LI M J,QIAN H,et al.Foundation model enhanced derivative-free cognitive diagnosis[J].Frontiers of Computer Science,2024,19(1):191318.
[10]DONG Z,CHEN J Y,et al.LLM-driven cognitive diagnosis with solo taxonomy:A model-agnostic framework[J].Frontiers of Digital Education,2025,2(2):1-17.
[11]WANG Y X,FENG J L,et al.A Style-Aware Polytomous Diagnostic Model for Individual Traits[C].In 28th European Conference on Artificial Intelligence.Bologna:IOS Press BV,2025.2698-2705.
[12]WU J,ZHENG C J.MetaCD:A Meta Learning Framework for Cognitive Diagnosis based on Continual Learning[J].ar Xiv preprint ar Xiv:2512.22904,2025.
[13]MA H,WANG C,et al.Large language models are zero-shot cross-domain diagnosticians in cognitive diagnosis[J].Frontiers of Digital Education,2025,2(2):1-14.
[14]PLUCKER J A, QIAN M H, et al. Is what you see what you really get?Comparison of scoring techniques in the assessment of real-world divergent thinking[J].Creativity Research Journal,2014,26(2):135-143.
[15]YANG T C,ZHANG Q F,et al.Automatic assessment of divergent thinking in Chinese language with Trans Ds:Ai transformer-based language model approach[J].Behavior Research Methods,2024,56(6):5798-5819.
[16]MEDNICK S.The associative basis of the creative process[J].Psychological Review,1962,69(3):220-232.
[17]COLLINS A M,LOFTUS E F.A spreading-activation theory of semantic processing[J].Psychological Review,1975,82(6):407-428.
[18]RUAN K,WANG X,et al.Evaluating LLMs’divergent thinking capabilities for scientific idea generation with minimal Context[J/OL].Nature Communication,2026.https://doi.org/10.1038/s41467-026-70245-1.
[19]ACAR S,BERTHIAUME K,et al.Applying automated originality scoring to the verbal form of Torrance tests of creative thinking[J].Gifted Child Quarterly,2023,67(1):3-17.
[20]BEATY R E,KENETT Y N,et al.Semantic memory and creativity:The costs and benefits of semantic memory structure in generating original ideas[J].Thinking&Reasoning,2023,29(2):305-339.
[21]LUCHINI S A,MALIAKKAL N T,et al.Automated scoring of creative problem solving with large language models:A comparison of originality and quality ratings[J/OL].Psychology of Aesthetics,Creativity,and the Arts,2026.https://doi.org/10.1037/aca0000736.
[22]FENG K H, DING K Y, et al. Sa Mer:A scenario-aware multi-dimensional evaluator for large language models[C].Vancouver:ICLR Press,2025.
[23]WANG H X,XIONG W,et al.Interpretable preferences via multi-objective reward modeling and mixture-of-experts[C].Bangkok:ACL Press,2024.10582-10592.
[24]KIM S,SHIN J,et al.Prometheus:inducing fine-grained evaluation capability in language models[C].Kigali:ICLR Press,2023.
[25]PATTERSON J D, MERSEAL H M, et al. Multilingual semantic distance:Automatic verbal creativity assessment in many languages[J].Psychology of Aesthetics,Creativity,and the Arts,2023,17(4):495-510.
[26]ZHU Q N,WANG M,et al.Current trends and future prospects of large-scale foundation model in K-12 education[J].Frontiers of Digital Education,2025,2(2):1-23.
[27]GAO W,LIU Q,et al.Agent4Edu:Generating learner response data by generative agents for intelligent education systems[C].Philadelphia:AAAI Press,2025.
[28]LIU S Y, FENG J T, et al. COMET:“cone of experience” enhanced large multimodal model for mathematical problem generation[J].Science China Information Sciences,2024,67(12):220108.
[29]郭家俊.如何在课堂上评价学生的创造性问题解决能力[J].上海教育,2023,(3):46-47.
[30]GOLDSCHMIDT G.Linkographic evidence for concurrent divergent and convergent thinking in creative design[J].Creativity Research Journal,2016,28(2):115-122.
[31]GUILFORD J P.Cognitive styles:What are they?[J].Educational and Psychological Measurement,1980,40(3):715-735.
[32]Cropley A.In praise of convergent thinking[J].Creativity Research Journal,2006,18(3):391-404.
基本信息:
中图分类号:G434
引用信息:
[1]张议文,钱鸿,王小雯,等.基于大语言模型的跨学科创造性问题解决能力自动测评方法研究[J].中国电化教育,2026,No.472(05):85-94.
基金信息:
中央高校基本科研业务费项目华东师范大学文理交叉跨学科培育项目(项目编号:2026ECNU-WLJC009); 浙江全省智能教育技术与应用重点实验室开放研究基金(项目编号:2025ZNJYKF014)资助
2026-05-10
2026-05-10