Publication Date

2021

Document Type

Dissertation

Committee Members

Deborah Steele-Johnson, Ph.D. (Advisor); Corey Miller, Ph.D. (Committee Member); Nathan Bowling, Ph.D. (Committee Member); Ion Juvina, Ph.D. (Committee Member)

Degree Name

Doctor of Philosophy (PhD)

Abstract

Users change their behavior when interacting with automated systems based upon their trust levels. Users faced with an unknown system will adjust their trust levels as they learn more about that system. Past automation trust research has implicitly assumed that users are passive recipients of information when interacting with new systems. Feedback-seeking behavior, a pattern of behavior involving actively eliciting information about one’s performance, is a well-researched concept within interpersonal research. Applying this interpersonal research to the domain of automation, I examined cases in which individuals sought feedback regarding the reliability of an unfamiliar automated system by asking for answers the user already possessed. I found evidence that feedback-seeking behavior exists within interactions with automation and called these behaviors intentional tests of the automated system. Users conducted more intentional tests on the system when faced with increased uncertainty (H1) and when encountering relatively early (H2) or easy (H3) trials. During these tests, users spent relatively little time assessing the system responses (H4). The effect of these intentional tests upon trust was significant yet relatively short-lasting (H5). This research shows another example of a case in which researchers may generalize the results of interpersonal research to the domain of automation. Engineers may also use these results to begin addressing a long-standing problem in automation trust: the inability for interventions to interact with long-term user behavior. These results demonstrated that intentional tests exist, can be a useful tool, may be able to be identified automatically, and have at least some unintuitive properties that merit further study.

Page Count

111

Department or Program

Department of Psychology

Year Degree Awarded

2021

Download


Share

COinS