Can knowledge based systems be designed to counteract deskilling effects?

Vicky Arnold, Philip A. Collier, Stewart A. Leech, Jacob M. Rose, Steve G. Sutton

Research output: Contribution to journalArticleResearchpeer-review

11 Citations (Scopus)

Abstract

Recent calls in the information systems research community argue that we know intelligent systems deskill users, and future research should focus on how to design systems that do not deskill, rather than continue to examine whether the phenomenon occurs. This should be a wakeup call for public accounting firms focused on implementing restrictive audit support systems, which leads to de-skilling of novice accounting professionals. Our research focuses on redesigning knowledge-based systems to facilitate expertise development and counteract the de-skilling effects that result from use of such systems. Specifically, we manipulate the design of the system interface by providing information cues in a screen format consistent with expert knowledge representations and manipulate automatic provision versus voluntary use of explanations for users during task completion. Results show that after using the knowledge-based system to complete a series of reenacted client engagements over a three-day period, both the interface design manipulation and automatic provision of explanations had a positive effect on novice accounting professionals’ development of expert-like knowledge structures. The results of the study have important implications for the development of knowledge-based systems intended to support accounting professionals’ (and other knowledge workers’) expertise development processes.

Original languageEnglish
Article number100638
Number of pages20
JournalInternational Journal of Accounting Information Systems
Volume50
DOIs
Publication statusPublished - Sept 2023
Externally publishedYes

Keywords

  • Artificial Intelligence
  • Deskilling
  • Explanation Systems
  • Knowledge Structures
  • Knowledge-Based Systems

Cite this