Scholarlink

Background <p>Communication competence is a core component of clinical competence among undergraduate nursing students. However, conventional communication teaching is often constrained by limited contextual authenticity, insufficient opportunities for repeated practice, and inadequate feedback support. Large language model–based virtual patients (LLM-VPs) offer a new approach to nursing communication training, but limited evidence exists on how undergraduate nursing students, based on their prior communication-learning experiences, anticipate the educational value, functional requirements, and boundaries of such systems.</p> Objective <p>This study aimed to explore undergraduate nursing students’ anticipated perceptions, functional needs, preliminary acceptance, and views on potential risks and boundaries regarding the use of LLM-VPs for communication training before they had used a dedicated LLM-VP system. The findings were intended to provide early evidence for the integration of this technology into nursing education.</p> Methods <p>This study employed a descriptive qualitative design. A total of 15 undergraduate nursing students were recruited through convenience sampling. Data were collected through semi-structured interviews after written informed consent had been obtained from each participant. Data collection and analysis were conducted concurrently until theoretical saturation was reached. The collected data were coded and analyzed using thematic analysis, with the assistance of NVivo 14 software.</p> Results <p>Four main themes were identified: emerging educational gaps and competence needs; expected system form and design requirements; mechanisms underlying trust and sustained use; and perceived risks and boundaries of applicability. Overall, nursing students expressed cautiously optimistic expectations toward LLM-VPs and articulated diverse functional expectations for aligning these systems with nursing-specific learning needs.</p> Conclusions <p>From the anticipatory perspective of undergraduate nursing students, LLM-VPs have practical appeal and potential utility in nursing communication education. However, their educational value remains conditional and requires further validation in real-use studies. Students expected LLM-VPs to function not as replacements for real clinical encounters but as supplementary training tools that support repeated practice, immediate feedback, and preclinical rehearsal. Future implementation should be grounded in nursing-specific adaptation, deep curricular integration, rigorous knowledge verification, ethical governance, and empirical evaluation of real-world educational effects to support standardized, cautious, and sustainable use in nursing education.</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

Undergraduate nursing students’ perceptions, needs, and expectations regarding large language model–based virtual patients: a qualitative study

  • 期刊论文

  • 出版模式：

    开放获取

  • 发表日期：

    2026年6月29日

查看全文
  • Jianwen Zeng,
  • Xule Zhu,
  • Xin Liu,
  • Sixie Li,
  • Shiying Shen,
  • Qiong Zhang,
  • Shihua Cao

摘要

Background

Communication competence is a core component of clinical competence among undergraduate nursing students. However, conventional communication teaching is often constrained by limited contextual authenticity, insufficient opportunities for repeated practice, and inadequate feedback support. Large language model–based virtual patients (LLM-VPs) offer a new approach to nursing communication training, but limited evidence exists on how undergraduate nursing students, based on their prior communication-learning experiences, anticipate the educational value, functional requirements, and boundaries of such systems.

Objective

This study aimed to explore undergraduate nursing students’ anticipated perceptions, functional needs, preliminary acceptance, and views on potential risks and boundaries regarding the use of LLM-VPs for communication training before they had used a dedicated LLM-VP system. The findings were intended to provide early evidence for the integration of this technology into nursing education.

Methods

This study employed a descriptive qualitative design. A total of 15 undergraduate nursing students were recruited through convenience sampling. Data were collected through semi-structured interviews after written informed consent had been obtained from each participant. Data collection and analysis were conducted concurrently until theoretical saturation was reached. The collected data were coded and analyzed using thematic analysis, with the assistance of NVivo 14 software.

Results

Four main themes were identified: emerging educational gaps and competence needs; expected system form and design requirements; mechanisms underlying trust and sustained use; and perceived risks and boundaries of applicability. Overall, nursing students expressed cautiously optimistic expectations toward LLM-VPs and articulated diverse functional expectations for aligning these systems with nursing-specific learning needs.

Conclusions

From the anticipatory perspective of undergraduate nursing students, LLM-VPs have practical appeal and potential utility in nursing communication education. However, their educational value remains conditional and requires further validation in real-use studies. Students expected LLM-VPs to function not as replacements for real clinical encounters but as supplementary training tools that support repeated practice, immediate feedback, and preclinical rehearsal. Future implementation should be grounded in nursing-specific adaptation, deep curricular integration, rigorous knowledge verification, ethical governance, and empirical evaluation of real-world educational effects to support standardized, cautious, and sustainable use in nursing education.