Paper Number
ICIS2025-2375
Paper Type
Complete
Abstract
As a transformative technology for human creativity, Generative AI exerts a double-edged effect on learners, acting either as a crutch or a booster. By integrating classical psychometrics with questioning-based learning theory, this study proposes a novel computational framework for evaluating divergent thinking in GenAI-assisted learning. To implement this framework, we leverage advancements in mathematical natural language metrics within an ideation context, developing a composite text-based approach to quantify divergent thinking across three key dimensions: fluency, flexibility, and originality. A field experiment reveals that questioning-enhanced GenAI generally boosts students’ divergent thinking across all dimensions. However, heterogeneity analyses detect its adverse effect when combined with high task difficulty or initial interactions. Notably, it particularly benefits international students facing barriers to domestic lecturer resources, while this benefit also increases through long-term use. These findings advance current literature and offer a practical pathway for tailoring GenAI to foster divergent thinking in educational contexts.
Recommended Citation
Chen, Runnan; LI, Andy Tao; and Ma, Tianyi, "Fostering Divergent Thinking via Generative AI: The Role of Questioning-Based Learning" (2025). ICIS 2025 Proceedings. 16.
https://aisel.aisnet.org/icis2025/learn_curricula/learn_curricula/16
Fostering Divergent Thinking via Generative AI: The Role of Questioning-Based Learning
As a transformative technology for human creativity, Generative AI exerts a double-edged effect on learners, acting either as a crutch or a booster. By integrating classical psychometrics with questioning-based learning theory, this study proposes a novel computational framework for evaluating divergent thinking in GenAI-assisted learning. To implement this framework, we leverage advancements in mathematical natural language metrics within an ideation context, developing a composite text-based approach to quantify divergent thinking across three key dimensions: fluency, flexibility, and originality. A field experiment reveals that questioning-enhanced GenAI generally boosts students’ divergent thinking across all dimensions. However, heterogeneity analyses detect its adverse effect when combined with high task difficulty or initial interactions. Notably, it particularly benefits international students facing barriers to domestic lecturer resources, while this benefit also increases through long-term use. These findings advance current literature and offer a practical pathway for tailoring GenAI to foster divergent thinking in educational contexts.
When commenting on articles, please be friendly, welcoming, respectful and abide by the AIS eLibrary Discussion Thread Code of Conduct posted here.
Comments
24-Learning