The Effects of System Technology and Probability Type on Trust, Compliance, and Reliance

Abstract

The purpose of this research was to examine the effects of system technology (binary vs. likelihood) and probability type (false-alarm prone vs. miss-prone) on trust, compliance and reliance using a false alarm forgiving task. Onehundred university students participated in this study. Participants completed three simulated flight missions composed of two primary flight tasks and a secondary engine-monitoring task. During the first mission, participants performed the engine-monitoring task without the aid of an automated alarm system. During the second and third sessions, participants had the opportunity to perform the engine-monitoring task with the aid of an alarm system. Results partially supported our hypotheses. Consistent with prior research, system probability type had a statistically main effect on trust. As expected, system technology and probability type had a statistically significant interaction effect on compliance. However, contrary to our prediction, results showed an unexpected interaction effect between system technology and probability type on reliance. The results of this study assist in defining the association among these three constructs and will aid in the future development of application principles for human-automation interaction.

Effects of False-Alarm vs. Miss-Prone Automation and Likelihood Alarm Technology on Trust, Reliance, and Compliance in a Miss-Prone Task

“…Compliance refers to an operator responding in accordance with an alarm signal. Prior research shows that LAT tends to reduce compliance in FP systems (Bustamante, 2008;Stanton et al, 2009). Reliance and compliance do not appear to be completely independent of each other .…”

“…Depending on the costs and benefits of a miss or a falsealarm in a given task, the task itself can be considered FP or MP. As an example of a FP task, Stanton, Ragsdale, and Bustamante (2009) evaluated the effects of LAT vs. BAT and FP vs. MP systems on trust, reliance, and compliance.…”

“…High levels of trust might be related to automation-induced complacency, which reduces operator performance (Bailey & Scerbo, 2007). Stanton et al (2009) found trust to be higher for FP automation than for MP.…”

Trust in Automation

“…Although some research has supported this hypothesis (Johnson, 2004), other research suggests that false alarms and misses have similar effects (Madhavan et al, 2006;Rovira & Parasuraman, 2010). Additionally, at least two studies have shown that participants trusted false-alarm-prone automation more than missprone automation (Davenport & Bustamante, 2010;Stanton, Ragsdale, & Bustamante, 2009). Overall, the specific impact that false alarms and misses have on trust likely depends on the negative consequences associated with each error type in a specific context.…”

Impact of Indoor Location Information Reliability on Users’ Trust of an Indoor Positioning System

“…If a user knows the principles of positioning calculation, their tolerance for positioning error could be higher than someone without this knowledge (Bell, et al, 2011). In addition, users having complete knowledge of the local environment tend to ignore the system's actual performance, which results in an inappropriate accuracy judgment (Stanton, Ragsdale, & Bustamante, 2009). Similarly, users tend to be critical of system performance when they are highly confident and have abundant knowledge of their surroundings (Kantowitz, Hanowski, & Kantowitz, 1997).…”