journal article
Download Only Collection
Influence of Personal Innovativeness and Different Sequences of Data Presentation on Evaluations of Explainable Artificial Intelligence
Jang, Wonseok; Chang, Youjin; Kim, Bomin; Lee, Young Ji; Kim, Seung-Chan
doi: 10.1080/10447318.2023.2209995pmid: N/A
Abstract By integrating the construal level theory (CLT) and explainable artificial intelligence (XAI) framework, this study identified personal innovativeness (PI) as a key user trait and different sequences of data presentation as key interfaces of an XAI system that determines how users evaluate and accept recommendations received from XAI. Based on CLT, the results of Study 1 identified the degree of psychological distance that users form with AI technology as a key underlying mechanism that explains the positive effects of users’ PI on their evaluation of AI recommendation systems. From the perspective of XAI, the results of Study 2 further demonstrated that users’ evaluations of XAI recommendation systems are determined by their level of PI and how the XAI interface presents information to them (i.e., whether the explanation comes first and is then followed by a decision or vice versa). The results of this study provide several meaningful theoretical and practical implications for human-AI interactions.