Scholarlink

<p>In the age of generative artificial intelligence (GenAI), gaining a better understanding of students' dependency on GenAI tools in academic settings has become a pressing concern. This study investigates the psychological and contextual mechanisms underlying GenAI dependency in Chinese higher education. Drawing on Affordance Theory and Task–Technology Fit Theory, the study develops a comprehensive framework linking GenAI attributes (practicality, reliability, and accessibility), GenAI trust, and GenAI dependency while evaluating the moderating role of organisational support. A mixed-methods approach was adopted, combining Partial Least Squares Structural Equation Modelling (PLS-SEM) with a 2×2 two-factor experiment, involving 426 Chinese university students, and a 2×2 two-factor experiment with a sample of 160 Chinese university students. The SEM results show that GenAI attributes (practicality, reliability, and accessibility) significantly predict GenAI trust, and concurrently significantly affect GenAI dependency. Furthermore, organisational support strengthened the relationship between trust and dependency. The experimental results confirm significant effects of both task type (creative vs. mechanical) and execution mode (AI-assisted vs. Manual) on GenAI dependency, with no significant interaction effect. These findings offer theoretical insights into how perceptions of technology and environmental support shape behavioural dependency on GenAI. In practical terms, the results provide the foundation for concrete recommendations in higher education institutions, aimed at instructors and student support systems seeking to integrate GenAI responsibly into teaching and learning.</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

From trust to dependency: how generative AI attributes and organisational support shape student use of generative AI

  • 期刊论文

  • 发表日期：

    2026年3月28日

查看全文
  • Haishun Wang,
  • Jianwei Tian

摘要

In the age of generative artificial intelligence (GenAI), gaining a better understanding of students' dependency on GenAI tools in academic settings has become a pressing concern. This study investigates the psychological and contextual mechanisms underlying GenAI dependency in Chinese higher education. Drawing on Affordance Theory and Task–Technology Fit Theory, the study develops a comprehensive framework linking GenAI attributes (practicality, reliability, and accessibility), GenAI trust, and GenAI dependency while evaluating the moderating role of organisational support. A mixed-methods approach was adopted, combining Partial Least Squares Structural Equation Modelling (PLS-SEM) with a 2×2 two-factor experiment, involving 426 Chinese university students, and a 2×2 two-factor experiment with a sample of 160 Chinese university students. The SEM results show that GenAI attributes (practicality, reliability, and accessibility) significantly predict GenAI trust, and concurrently significantly affect GenAI dependency. Furthermore, organisational support strengthened the relationship between trust and dependency. The experimental results confirm significant effects of both task type (creative vs. mechanical) and execution mode (AI-assisted vs. Manual) on GenAI dependency, with no significant interaction effect. These findings offer theoretical insights into how perceptions of technology and environmental support shape behavioural dependency on GenAI. In practical terms, the results provide the foundation for concrete recommendations in higher education institutions, aimed at instructors and student support systems seeking to integrate GenAI responsibly into teaching and learning.