医学教育管理 ›› 2026, Vol. 12 ›› Issue (5): 687-694.doi: 10.3969/j.issn.2096-045X.2026.05.017

• 调查研究 • 上一篇    下一篇

医学生感知学习风格差异及其对医学英语AI智能体个性化适配的启示

  

  1. 首都医科大学医学人文学院与全科医学学院,北京 100069
  • 收稿日期:2026-04-27 修回日期:2026-06-01 出版日期:2026-11-13 发布日期:2026-11-13

Differences in perceptual learning styles of medical students and their implications for personalized adaptation of medical English AI agent

  1. School of Medical Humanities and General Practice, Capital Medical University, Beijing 100069, China
  • Received:2026-04-27 Revised:2026-06-01 Online:2026-11-13 Published:2026-11-13

摘要: 目的 探究医学生感知学习风格特征及其对混合式教学评估的影响,为医学英语AI智能体个性化适配与优化提供实证依据。方法 以首都医科大学2024级本科二年级 126 名学生为对象,采用感知学习风格偏好(PLSPQ)量表与自编混合式教学评估问卷及半结构式访谈收集教学评价数据,有效样本122份。采用单因素方差分析比较不同学习风格学生在教师角色、自主学习能力、线上资源及混合式教学整体评价4个维度上的得分差异,并运用Games-Howell法进行事后两两比较。结果 触觉型(31.1%,38/122)、个人型(20.5%,25/122)为医学生占比最高的主要学习风格;教师角色评价最高(4.06±0.64),线上资源评价最低(3.40±0.82);不同学习风格在教师角色(F=6.846,P<0.001)、线上资源维度(F=2.957,P=0.015)具有统计学差异,其中听觉型对教师依赖最高(4.82±0.32),个人型学生依赖度最低(3.62±0.67),二者均分差 1.201 分(P<0.001)。结论 本研究证实医学生感知学习风格存在群体分化,不同学习风格对教师角色、线上资源评价存在明显差异。应依托医学生学习风格搭建动态学情数据库并嵌入医学英语AI智能体,适配AI个性化学习路径,推行“教师导学+智能伴学”的双轨教学模式,重构人机协同的医学英语教学生态,实现精准因材施教。

Abstract:

Objective To explore the patterns of medical students' perceptual learning styles and their influences on blended teaching evaluation, so as to provide empirical evidence for the personalized adaptation and optimization of the medical English AI agent. Methods  A total of 126 second-year undergraduates of Grade 2024 from Capital Medical University were investigated using the Perceptual Learning Style Preference Questionnaire (PLSPQ), a self-designed questionnaire on blended teaching evaluation and semi-structured interviews, with 122 valid samples finally obtained. One‑way analysis of variance was adopted to compare score differences across four dimensions including teacher role, autonomous‑learning ability, online resources and overall blended‑teaching evaluation among students with different learning styles, and Games‑Howell method was used for post‑hoc pairwise comparisons. Results  Tactile (31.1%, 38/122) and individual (20.5%, 25/122) learning styles accounted for the largest proportions. The score of teacher‑role dimension was the highest (4.06±0.64), whereas online‑resource dimension scored the lowest (3.40±0.82). Statistically significant inter‑group differences were found in the teacher‑role (F=6.846, P<0.001) and online‑resource dimensions (F=2.957, P=0.015). Auditory learners reported the highest score of teacher dependence (4.82±0.32), while individual learners had the lowest score (3.62±0.67), with a mean difference of 1.201 (P<0.001) between the two groups. Conclusion  This study confirms the group differentiation of perceptual learning styles among medical students, and obvious differences exist in their evaluation on teacher role and online resources. A dynamic learning profile database based on learning styles should be constructed and embedded into the medical English AI agent to match personalized AI learning paths. The dual-track mode of "teacher guidance+AI companion learning" should be implemented to reconstruct a new human-machine-collaborative medical English teaching ecology and achieve precise and targeted teaching.

Key words:

"> Perceptual Learning Styles| AI Agent| Medical Students| Medical English| Blended Teaching|Human-Machine Collaboration| Personalized Learning