Scholarlink

Background <p>Self-directed learning (SDL) is a core competency in medical education. The emergence of generative artificial intelligence (AI) has reshaped technology-enhanced learning, yet it remains unclear how AI-related contextual factors interact with personality traits to influence SDL.</p> Methods <p>A cross-sectional survey was conducted among 1,208 medical undergraduates from universities in Shanghai and other regions of China. Data were analyzed using general linear modeling, hierarchical regression, conditional process analysis, and latent profile analysis (LPA). Sensitivity analyses and common method variance diagnostics were performed.</p> Results <p>In the GLM, conscientiousness (β = 2.76, <i>p</i> &lt; 0.001, η²<i>p</i> = 0.16), self-esteem (β = 0.70, <i>p</i> &lt; 0.001), openness (β = 0.70, <i>p</i> &lt; 0.001), and self-control (β = 0.13, <i>p</i> = 0.036) all positively predicted SDLB. Conditional process analysis revealed that the indirect effect of openness on SDLB via AI cognition was significantly moderated by subjective norms (moderated mediation index = − 0.009, 95% CI [− 0.015, − 0.003]); this effect weakened as subjective norms intensified. LPA identified four learner subgroups: Active Adopters, Cognitively Advanced but Constrained, Passive Followers, and Disengaged. Within-profile regressions showed heterogeneous personality effects.</p> Conclusions <p>Personality traits, AI cognition, and subjective norms jointly shape SDL among Chinese medical undergraduates. The positive indirect effect of openness on SDL via AI cognition is attenuated by stronger subjective norms, highlighting that AI-mediated learning environments act as contexts of conditional indirect effects. The identification of distinct learner profiles emphasizes the need for differentiated educational strategies rather than a uniform approach to AI integration.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

The role of AI-mediated learning contexts in the personality–self-directed learning relationship among Chinese medical undergraduates: a cross-sectional study

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年5月13日

查看全文
  • Chenrui Li,
  • Fan Hu,
  • Yongxia Qiao,
  • Meihui Zhang,
  • Quyige Gao,
  • Fengyi Wang,
  • Naiya Du,
  • Yunyi Xia,
  • Meizhu Pan,
  • Yong Cai,
  • Ying Wang

摘要

Background

Self-directed learning (SDL) is a core competency in medical education. The emergence of generative artificial intelligence (AI) has reshaped technology-enhanced learning, yet it remains unclear how AI-related contextual factors interact with personality traits to influence SDL.

Methods

A cross-sectional survey was conducted among 1,208 medical undergraduates from universities in Shanghai and other regions of China. Data were analyzed using general linear modeling, hierarchical regression, conditional process analysis, and latent profile analysis (LPA). Sensitivity analyses and common method variance diagnostics were performed.

Results

In the GLM, conscientiousness (β = 2.76, p < 0.001, η²p = 0.16), self-esteem (β = 0.70, p < 0.001), openness (β = 0.70, p < 0.001), and self-control (β = 0.13, p = 0.036) all positively predicted SDLB. Conditional process analysis revealed that the indirect effect of openness on SDLB via AI cognition was significantly moderated by subjective norms (moderated mediation index = − 0.009, 95% CI [− 0.015, − 0.003]); this effect weakened as subjective norms intensified. LPA identified four learner subgroups: Active Adopters, Cognitively Advanced but Constrained, Passive Followers, and Disengaged. Within-profile regressions showed heterogeneous personality effects.

Conclusions

Personality traits, AI cognition, and subjective norms jointly shape SDL among Chinese medical undergraduates. The positive indirect effect of openness on SDL via AI cognition is attenuated by stronger subjective norms, highlighting that AI-mediated learning environments act as contexts of conditional indirect effects. The identification of distinct learner profiles emphasizes the need for differentiated educational strategies rather than a uniform approach to AI integration.