How Transparency Measures Can Attenuate Initial Failures of Intelligent Decision Support Systems

Matt, C.; Schlusche, C. (2022). How Transparency Measures Can Attenuate Initial Failures of Intelligent Decision Support Systems. In: Thirtieth European Conference on Information Systems (ECIS 2022). Timișoara, Romania. 18.-24. Juni 2022.

Text Matt_Schlusche__2022__-_Transparency_Measures_Intelligent_DSS.pdf - Published Version Restricted to registered users only Available under License BORIS Standard License. Download (440kB)

Owing to high functional complexity, trust plays a critical role for the adoption of intelligent decision support systems (DSS). Especially failures in initial usage phases might endanger trust since users are yet to assess the system’s capabilities over time. Since such initial failures are unavoidable, it is crucial to understand how providers can inform users about system capabilities to rebuild user trust. Using an online experiment, we evaluate the effects of recurring explanations and initial tutorials as transparency measures on trust. We find that recurring explanations are superior to initial tutorials in establishing trust in intelligent DSS. However, recurring explanations are only as effective as tutorials or the combination of both tutorials and recurring explanations in rebuilding trust after initial failures occurred. Our results provide empirical insights for the design of transparency mechanisms for intelligent DSS, especially those with high underlying algorithmic complexity or potentially high damage.

Interest & Impact

Downloads

6 since deposited on 24 May 2022
1 in the past 12 months

Citations

5 Citations in Web of Science ®
6 Citations in Scopus

Search

in Google Scholar™

Services

Actions (login required)

Edit item

Actions (login required)

Edit item