본문 바로가기
  • Home

Meaning Reconstruction in Generative AI Use Experience: The Roles of Pre-use Cognitive Bias and User Responsibility Attribution

  • Industry Promotion Research
  • Abbr : IPR
  • 2026, 11(3), pp.493~503
  • DOI : 10.21186/IPR.2026.11.3.493
  • Publisher : Industrial Promotion Institute
  • Research Area : Interdisciplinary Studies > Interdisciplinary Research
  • Received : June 23, 2026
  • Accepted : July 8, 2026
  • Published : July 31, 2026

Hwang Jeong Woo 1 Kim Seung In 1

1홍익대학교

Accredited

ABSTRACT

This study examines how pre-use cognitive bias relates to meaning reconstruction through perceived technological imperfection and responsibility attribution in generative AI use. As users encounter hallucinations, unstable outputs, and expectation–performance gaps, AI experience needs to be understood beyond acceptance or continued use. A cross-sectional survey was conducted with 400 adult generative AI users. The data were analyzed using confirmatory factor analysis, HTMT, correlation analysis, regression analysis, and PROCESS Macro Model 6. H1 was not supported because pre-use cognitive bias and perceived technological imperfection were related in a direction opposite to the hypothesis. H2 and H3 were supported: perceived technological imperfection was related to responsibility attribution, and responsibility attribution to meaning reconstruction. The serial indirect effect in H4 was not significant, whereas the PCB→RA→MR path was significant. These findings suggest that meaning reconstruction is more closely associated with users’ interpretation of error responsibility than with technological imperfection alone. The study offers UX implications by emphasizingerror awareness, responsibility boundaries, verification support, and user intervention.

Citation status

* References for papers published after 2025 are currently being built.