@article{ART003120942},
author={No-Ah Han and Yoo-jin Hwang and Zoon-Ky Lee},
title={Study on the Impact of XAI Explanation Levels on Cognitive Load and User Satisfaction : Focusing on Risk Levels in Financial AI Systems},
journal={Journal of The Korea Society of Computer and Information},
issn={1598-849X},
year={2024},
volume={29},
number={9},
pages={49-59},
doi={10.9708/jksci.2024.29.09.049}
TY - JOUR
AU - No-Ah Han
AU - Yoo-jin Hwang
AU - Zoon-Ky Lee
TI - Study on the Impact of XAI Explanation Levels on Cognitive Load and User Satisfaction : Focusing on Risk Levels in Financial AI Systems
JO - Journal of The Korea Society of Computer and Information
PY - 2024
VL - 29
IS - 9
PB - The Korean Society Of Computer And Information
SP - 49
EP - 59
SN - 1598-849X
AB - In this paper, we examine the impact of XAI explanations on user satisfaction and cognitive load according to the risk levels defined in the EU AI Act. XAI aims to make the internal processes of complex AI models understandable to humans and is widely used in both academia and industry. The importance and value of XAI are continuously rising; however, there has been little research determining the necessary level of explanation according to AI system risk levels. To address this gap, we designed an experiment with 120 participants, divided into 8 groups, each exposed to one of four levels of explainability(XAI) within low-risk and high-risk financial AI systems. A quantitative approach was used to measure cognitive load, user satisfaction, mental effort, and the clarity of the material design across the different AI system interfaces. The results indicate that the amount of information in explanations significantly affects cognitive load and user satisfaction, depending on the risk level. However, the impact of the level of explanation on user satisfaction was mediated by the material design, which determined how easily the information was understood. This research provides practical, regulatory, and academic contributions by offering guidelines for determining the necessary level of explanation based on AI system risk levels.
KW - XAI(Explainable AI);Right to Explanation;Cognitive Load;System Risk Levels [요
DO - 10.9708/jksci.2024.29.09.049
ER -
No-Ah Han, Yoo-jin Hwang and Zoon-Ky Lee. (2024). Study on the Impact of XAI Explanation Levels on Cognitive Load and User Satisfaction : Focusing on Risk Levels in Financial AI Systems. Journal of The Korea Society of Computer and Information, 29(9), 49-59.
No-Ah Han, Yoo-jin Hwang and Zoon-Ky Lee. 2024, "Study on the Impact of XAI Explanation Levels on Cognitive Load and User Satisfaction : Focusing on Risk Levels in Financial AI Systems", Journal of The Korea Society of Computer and Information, vol.29, no.9 pp.49-59. Available from: doi:10.9708/jksci.2024.29.09.049
No-Ah Han, Yoo-jin Hwang, Zoon-Ky Lee "Study on the Impact of XAI Explanation Levels on Cognitive Load and User Satisfaction : Focusing on Risk Levels in Financial AI Systems" Journal of The Korea Society of Computer and Information 29.9 pp.49-59 (2024) : 49.
No-Ah Han, Yoo-jin Hwang, Zoon-Ky Lee. Study on the Impact of XAI Explanation Levels on Cognitive Load and User Satisfaction : Focusing on Risk Levels in Financial AI Systems. 2024; 29(9), 49-59. Available from: doi:10.9708/jksci.2024.29.09.049
No-Ah Han, Yoo-jin Hwang and Zoon-Ky Lee. "Study on the Impact of XAI Explanation Levels on Cognitive Load and User Satisfaction : Focusing on Risk Levels in Financial AI Systems" Journal of The Korea Society of Computer and Information 29, no.9 (2024) : 49-59.doi: 10.9708/jksci.2024.29.09.049
No-Ah Han; Yoo-jin Hwang; Zoon-Ky Lee. Study on the Impact of XAI Explanation Levels on Cognitive Load and User Satisfaction : Focusing on Risk Levels in Financial AI Systems. Journal of The Korea Society of Computer and Information, 29(9), 49-59. doi: 10.9708/jksci.2024.29.09.049
No-Ah Han; Yoo-jin Hwang; Zoon-Ky Lee. Study on the Impact of XAI Explanation Levels on Cognitive Load and User Satisfaction : Focusing on Risk Levels in Financial AI Systems. Journal of The Korea Society of Computer and Information. 2024; 29(9) 49-59. doi: 10.9708/jksci.2024.29.09.049
No-Ah Han, Yoo-jin Hwang, Zoon-Ky Lee. Study on the Impact of XAI Explanation Levels on Cognitive Load and User Satisfaction : Focusing on Risk Levels in Financial AI Systems. 2024; 29(9), 49-59. Available from: doi:10.9708/jksci.2024.29.09.049
No-Ah Han, Yoo-jin Hwang and Zoon-Ky Lee. "Study on the Impact of XAI Explanation Levels on Cognitive Load and User Satisfaction : Focusing on Risk Levels in Financial AI Systems" Journal of The Korea Society of Computer and Information 29, no.9 (2024) : 49-59.doi: 10.9708/jksci.2024.29.09.049
