Related Experiment Videos
Artificial intelligence dependency in college students: a critical narrative review
Xuehua He1, Shan Li1, Rongping Cha1
1Nursing Department, Sir Run Run Shaw Hospital, Zhejiang University School of Medicine, Hangzhou, Zhejiang, China.
Frontiers in Psychology
|August 14, 2026
Summary
Generative AI (GenAI) use reshapes student learning, raising concerns about AI dependency. This review highlights risks like cognitive delegation and reduced self-regulation, urging focus on AI literacy and supportive interventions.
Area of Science:
- Educational Psychology
- Human-Computer Interaction
- Artificial Intelligence Ethics
Background:
- Generative artificial intelligence (GenAI), especially large language models (LLMs), is rapidly adopted by university students.
- This adoption impacts learning, problem-solving, and socio-emotional development, presenting both benefits and emerging concerns.
- AI dependency is conceptualized as an educational-psychological construct, not a clinical disorder, focusing on cognitive and emotional reliance.
Purpose of the Study:
- To critically review the existing literature on AI dependency among college students over the past decade.
- To synthesize evidence on conceptualizations, theoretical frameworks, prevalence, measurement, determinants, and interventions related to AI dependency.
- To identify research gaps and propose future research directions.
Main Methods:
- A critical narrative review synthesizing evidence from the past decade.
- Focus on conceptualizations, theoretical frameworks, prevalence, measurement tools, determinants, and intervention strategies.
- Analysis of emerging trends like the "ask-AI-first" pattern and its implications.
Main Results:
- An "ask-AI-first" usage pattern is common, but frequency does not equate to maladaptive dependency.
- Concerns center on cognitive delegation, emotional reliance, reduced self-regulation, academic integrity risks, and distress when AI is unavailable.
- Assessment tools show fragmentation, though some emerging measures have promising psychometric properties.
- Multi-level determinants include individual vulnerabilities, academic pressure, norms, and technology affordances.
Conclusions:
- Recommendations include promoting AI augmentation over automation and enhancing AI literacy.
- Assessment redesign is crucial to maintain cognitive engagement and academic integrity.
- Supportive interventions for at-risk students and further longitudinal, cross-cultural research are needed to understand AI dependency trajectories and effective interventions.
