Year: 2026
Type: article
Abstract: This paper employs a quantitative multimethod design to extend the UTAUT framework for generative AI (GenAI) adoption among art students. Study 1 tested the integrated model via structural equation modeling using cross-sectional survey data from 450 Chinese art students and developed a trust construct grounded in instrumental technology trust, supplemented by art-value-aligned items. Trust in GenAI showed the strongest direct association with behavioral intention ( β = 0.425). Social influence was the dominant correlate of trust ( β = 0.262), with 42.1% of its total effect mediated indirectly through trust. Metacognitive control showed a positive direct association with behavioral intention ( β = 0.146), alongside an indirect association through trust. The structural model based on survey data explained 59.7% of the variance in behavioral intention. Study 2 used a between-subjects experiment with 80 art students to test whether situational metacognitive prompting causally affected creative output. Participants receiving metacognitive prompts (Group A) obtained higher expert ratings than the control group (Group B) on semantic content ( d = 0.67), visual style ( d = 1.30), and spatial composition ( d = 0.63); within-group feature-distance analyses also indicated greater output diversity in Group A across all three dimensions. These findings suggest that trust and metacognitive control are key pathways in GenAI adoption for artistic creation, and that low-cost situational metacognitive prompting may enhance both the quality and diversity of AI-assisted artworks.
Source: Frontiers in Psychology
Author: Renhao Zhu
Institution: Hengdian College of Film & Television
Language: English
Subfield: Information Systems
Field: Computer Science
Domain: Physical Sciences
SDG: Quality education
Open Access status: gold
Estimated APC paid: $3,801
OpenAlex ID: W7224428899