{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,7,27]],"date-time":"2026-07-27T06:01:40Z","timestamp":1785132100488,"version":"3.55.0"},"reference-count":39,"publisher":"Frontiers Media SA","license":[{"start":{"date-parts":[[2026,7,27]],"date-time":"2026-07-27T00:00:00Z","timestamp":1785110400000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["frontiersin.org"],"crossmark-restriction":true},"short-container-title":["Front. Psychol."],"abstract":"<jats:sec>\n                    <jats:title>Introduction<\/jats:title>\n                    <jats:p>Large language models (LLMs) are increasingly used in higher education, yet empirical evidence for their effectiveness in art education remains scarce. This study aimed to evaluate whether LLM assistance could improve undergraduate art history question-answering performance and explanatory support.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Methods<\/jats:title>\n                    <jats:p>This study developed the Art History Theory Question Set (AHTQS), comprising 104 single-choice items with Bloom-level annotations, and benchmarked three LLMs (ChatGPT-4o, DeepSeek-V3, and Qwen2.5-Plus). DeepSeek-V3 showed the highest accuracy (96.2%) and lowest observed run-to-run variability and was selected for a randomized crossover pilot study with six undergraduates. The primary outcome was the change in examination accuracy from independent to LLM-assisted answering. A Likert-scale evaluation involving nine students and three instructors was also conducted to assess the clarity and coherence of LLM-generated explanations.<\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Results<\/jats:title>\n                    <jats:p>\n                      A one-sided Wilcoxon signed-rank test showed significant improvement with LLM support [\n                      <jats:italic>W<\/jats:italic>\n                      (6) = 21.0,\n                      <jats:italic>p<\/jats:italic>\n                      = 0.0156,\n                      <jats:italic>r<\/jats:italic>\n                      = 0.879], with median accuracy increasing from 43.3% to 93.3% (median gain = 42.3%). Five of six students showed higher accuracy under the LLM-assisted condition, and no clear evidence of a sequence or carryover effect was detected (Mann-Whitney U = 7.0,\n                      <jats:italic>p<\/jats:italic>\n                      = 0.3758). Domain-level analyses indicated significant gains in all four categories (\n                      <jats:italic>p<\/jats:italic>\n                      &amp;lt; 0.05). Error-frequency analysis further showed marked reductions in high-frequency mistakes. The Likert-scale evaluation indicated high perceived clarity and coherence of LLM explanations, with favorable but more cautious instructor ratings.\n                    <\/jats:p>\n                  <\/jats:sec>\n                  <jats:sec>\n                    <jats:title>Discussion<\/jats:title>\n                    <jats:p>These pilot findings suggest that supervised LLM assistance may support art history question-answering and explanatory feedback. Future studies should validate these findings in larger cohorts, assess delayed learning retention, and examine open-ended, image-based, and higher-order art history tasks before curriculum-level implementation.<\/jats:p>\n                  <\/jats:sec>","DOI":"10.3389\/fpsyg.2026.1875025","type":"journal-article","created":{"date-parts":[[2026,7,27]],"date-time":"2026-07-27T05:40:08Z","timestamp":1785130808000},"update-policy":"https:\/\/doi.org\/10.3389\/crossmark-policy","source":"Crossref","is-referenced-by-count":0,"title":["Effect of large language model assistance on undergraduate art history question-answering performance: a randomized crossover pilot study"],"prefix":"10.3389","volume":"17","author":[{"given":"Yunting","family":"Zhang","sequence":"first","affiliation":[{"name":"School of Fine Arts, Jining University","place":["Jining, China"]},{"name":"Department of Visual Information Design, College of Creative Culture and Art, Namseoul University","place":["Cheonan, Republic of Korea"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Fan","family":"Zhang","sequence":"additional","affiliation":[{"name":"GE Healthcare Systems Trade Development (Shanghai) Co., Ltd.","place":["Shanghai, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Zili","family":"Zhang","sequence":"additional","affiliation":[{"name":"School of Medicine, Shanghai Jiao Tong University","place":["Shanghai, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"1965","published-online":{"date-parts":[[2026,7,27]]},"reference":[{"key":"B1","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2303.08774","article-title":"GPT-4 technical report","author":"Achiam","year":"2023","journal-title":"arXiv [Preprint]"},{"key":"B2","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1145\/3449287","article-title":"To trust or to think: cognitive forcing functions can reduce overreliance on AI in AI-assisted decision-making","volume":"5","author":"Bu\u00e7inca","year":"2021","journal-title":"Proc. ACM Hum. Comput. Interact."},{"key":"B3","doi-asserted-by":"publisher","first-page":"5","DOI":"10.1386\/adch.11.1.5_1","article-title":"A question of values: why we need art and design in higher education","volume":"11","author":"Budge","year":"2012","journal-title":"Art Des. Commun. High. Educ."},{"key":"B4","doi-asserted-by":"publisher","first-page":"319","DOI":"10.2307\/249008","article-title":"Perceived usefulness, perceived ease of use, and user acceptance of information technology","volume":"13","author":"Davis","year":"1989","journal-title":"MIS Q."},{"key":"B5","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2309.11495","article-title":"Chain-of-verification reduces hallucination in large language models","author":"Dhuliawala","year":"2023","journal-title":"arXiv [Preprint]"},{"key":"B6","first-page":"126","article-title":"\u201cEvaluating the impact of prompt engineering on ChatGPT's performance in civil engineering education,\u201d","volume-title":"Proceedings of the 4th International Conference on Trends in Advanced Research","author":"Durgun","year":"2025"},{"key":"B7","first-page":"11","article-title":"The impact of hallucinated information in large language models on student learning outcomes: a critical examination of misinformation risks in AI-assisted education","volume":"9","author":"Elsayed","year":"2024","journal-title":"North. Rev. Algorithmic Res. Theor. Comput. Complex."},{"key":"B8","first-page":"47","article-title":"Bloom's taxonomy","volume":"41","author":"Forehand","year":"2010","journal-title":"Emerg. Perspect. Learn. Teach. Technol."},{"key":"B9","doi-asserted-by":"publisher","first-page":"598","DOI":"10.47495\/okufbed.1540360","article-title":"Evaluating the effectiveness of AI-powered conversational models in construction management education: a comparative investigation of student performances and ChatGPT","volume":"8","author":"Gen\u00e7","year":"","journal-title":"Osmaniye Korkut Ata Univ. Fen Bilimleri Enst. Derg."},{"key":"B10","doi-asserted-by":"publisher","DOI":"10.1080\/13467581.2025.2574556","article-title":"Is conversational AI ready for engineering licensure? A ChatGPT performance benchmark and prompting strategy evaluation","author":"Gen\u00e7","year":"","journal-title":"J. Asian Archit. Build. Eng"},{"key":"B11","doi-asserted-by":"publisher","first-page":"4","DOI":"10.14713\/ahpp.v1i1.2150","article-title":"Making the absent present: the imperative of teaching art history","volume":"1","author":"Harris","year":"2016","journal-title":"Art Hist. Pedagog. Pract."},{"key":"B12","doi-asserted-by":"publisher","first-page":"81","DOI":"10.3102\/003465430298487","article-title":"The power of feedback","volume":"77","author":"Hattie","year":"2007","journal-title":"Rev. Educ. Res."},{"key":"B13","doi-asserted-by":"publisher","DOI":"10.3389\/frai.2025.1609097","article-title":"An overview of model uncertainty and variability in LLM-based sentiment analysis. Challenges, mitigation strategies and the role of explainability","author":"Herrera-Poyatos","year":"2025","journal-title":"arXiv [Preprint]"},{"key":"B14","doi-asserted-by":"publisher","first-page":"15873","DOI":"10.1007\/s10639-023-11834-1","article-title":"Large language models in education: a focus on the complementary relationship between human teachers and ChatGPT","volume":"28","author":"Jeon","year":"2023","journal-title":"Educ. Inf. Technol."},{"key":"B15","doi-asserted-by":"publisher","first-page":"864","DOI":"10.1109\/RO-MAN60168.2024.10731312","article-title":"\u201cPhysicsAssistant: An LLM-powered interactive learning robot for physics lab investigations,\u201d","author":"Latif","year":"2024","journal-title":"2024 33rd IEEE International Conference on Robot and Human Interactive Communication (RO-MAN)"},{"key":"B16","doi-asserted-by":"publisher","first-page":"934","DOI":"10.1016\/j.scib.2025.01.034","article-title":"Large language models for diabetes training: a prospective study","volume":"70","author":"Li","year":"2025","journal-title":"Sci. Bull."},{"key":"B17","first-page":"1","article-title":"A technique for the measurement of attitudes","volume":"140","author":"Likert","year":"1932","journal-title":"Arch. Psychol."},{"key":"B18","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2412.19437","article-title":"DeepSeek-V3 technical report","author":"Liu","year":"2024","journal-title":"arXiv [Preprint]"},{"key":"B19","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2404.00971","article-title":"Exploring and evaluating hallucinations in LLM-powered code generation","author":"Liu","year":"2024","journal-title":"arXiv [Preprint]"},{"key":"B20","doi-asserted-by":"publisher","first-page":"520","DOI":"10.1007\/s11528-024-00954-1","article-title":"Higher education faculty perceptions of ChatGPT and the influencing factors: a sentiment analysis of X","volume":"68","author":"Mamo","year":"2024","journal-title":"TechTrends"},{"key":"B21","doi-asserted-by":"publisher","first-page":"1","DOI":"10.14742\/ajet.8825","article-title":"Digital transformation in engineering education: exploring the potential of AI-assisted learning","volume":"39","author":"Pham","year":"2023","journal-title":"Australas. J. Educ. Technol."},{"key":"B22","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2412.15115","article-title":"Qwen2.5 Technical Report","author":"Qwen","year":"2024","journal-title":"arXiv [Preprint]"},{"key":"B23","doi-asserted-by":"publisher","first-page":"2","DOI":"10.62608\/2831-3550.1030","article-title":"AI-powered storytelling and interactive timelines for arts and humanities education","volume":"3","author":"Raoulji","year":"2025","journal-title":"Int. J. Emerg. Disruptive Innov. Educ."},{"key":"B24","doi-asserted-by":"publisher","first-page":"37","DOI":"10.3390\/informatics11020037","article-title":"Analysing the impact of generative AI in arts education: a cross-disciplinary perspective of educators and students in higher education","volume":"11","author":"S\u00e1ez-Velasco","year":"2024","journal-title":"Informatics"},{"key":"B25","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2412.07923","article-title":"Asking again and again: exploring LLM robustness to repeated questions","author":"Shaier","year":"2024","journal-title":"arXiv [Preprint]"},{"key":"B26","doi-asserted-by":"publisher","first-page":"6","DOI":"10.62608\/2831-3550.1027","article-title":"Enhancing art history education with AI: accessibility, engagement, and skill development","volume":"3","author":"Soverns-Reed","year":"2025","journal-title":"Int. J. Emerg. Disruptive Innov. Educ."},{"key":"B27","doi-asserted-by":"publisher","first-page":"34188","DOI":"10.52202\/079017-1077","article-title":"LLM-check: investigating detection of hallucinations in large language models","volume":"37","author":"Sriramanan","year":"2024","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"B28","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1145\/3712065","article-title":"ELF: educational LLM framework of improving and evaluating AI generated content for classroom teaching","volume":"17","author":"Tan","year":"2025","journal-title":"ACM J. Data Inf. Qual."},{"key":"B29","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2401.01313","article-title":"A comprehensive survey of hallucination mitigation techniques in large language models","author":"Tonmoy","year":"2024","journal-title":"arXiv [Preprint]"},{"key":"B30","doi-asserted-by":"publisher","first-page":"13","DOI":"10.2307\/3193705","article-title":"Interdisciplinary art education reconsidered","volume":"51","author":"Ulbricht","year":"1998","journal-title":"Art Educ."},{"key":"B31","doi-asserted-by":"publisher","first-page":"611","DOI":"10.3390\/math8040611","article-title":"Benefits and limitations of the artificial with respect to the traditional learning of mathematics","volume":"8","author":"Voskoglou","year":"2020","journal-title":"Mathematics"},{"key":"B32","volume-title":"Review of Evidence: Arts Education Through the Lens of ESSA","author":"Wan","year":"2018"},{"key":"B33","doi-asserted-by":"publisher","first-page":"100326","DOI":"10.1016\/j.caeai.2024.100326","article-title":"Generative AI in higher education: seeing ChatGPT through universities' policies, resources, and guidelines","volume":"7","author":"Wang","year":"2024","journal-title":"Comput. Educ. Artif. Intell."},{"key":"B34","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2203.11171","article-title":"Self-consistency improves chain of thought reasoning in language models","author":"Wang","year":"2022","journal-title":"arXiv [Preprint]"},{"key":"B35","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2406.19314","article-title":"LiveBench: a challenging, contamination-limited LLM benchmark","author":"White","year":"2024","journal-title":"arXiv [Preprint]"},{"key":"B36","doi-asserted-by":"publisher","first-page":"100106","DOI":"10.1016\/j.apjo.2024.100106","article-title":"Accuracy of large language models in answering ophthalmology board-style questions: a meta-analysis","volume":"13","author":"Wu","year":"2024","journal-title":"Asia Pac. J. Ophthalmol."},{"key":"B37","doi-asserted-by":"publisher","first-page":"11809","DOI":"10.52202\/075280-0517","article-title":"Tree of thoughts: deliberate problem solving with large language models","volume":"36","author":"Yao","year":"2023","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"B38","doi-asserted-by":"publisher","first-page":"61","DOI":"10.1038\/s41586-024-07930-y","article-title":"Larger and more instructable language models become less reliable","volume":"634","author":"Zhou","year":"2024","journal-title":"Nature"},{"key":"B39","doi-asserted-by":"publisher","first-page":"898","DOI":"10.1038\/s41597-026-07261-9","article-title":"A dataset for evaluating large language models on Chinese National Medical Licensing Examinations","volume":"13","author":"Zong","year":"2026","journal-title":"Sci. Data"}],"container-title":["Frontiers in Psychology"],"original-title":[],"link":[{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fpsyg.2026.1875025\/full","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,7,27]],"date-time":"2026-07-27T05:40:13Z","timestamp":1785130813000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.frontiersin.org\/articles\/10.3389\/fpsyg.2026.1875025\/full"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,7,27]]},"references-count":39,"alternative-id":["10.3389\/fpsyg.2026.1875025"],"URL":"https:\/\/doi.org\/10.3389\/fpsyg.2026.1875025","relation":{},"ISSN":["1664-1078"],"issn-type":[{"value":"1664-1078","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,7,27]]},"article-number":"1875025"}}