Effects of attribute and goal framing on automation reliance and compliance

Effects of Information Source, Pedigree, and Reliability on Operator Interaction With Decision Support Systems

“…Based on findings on human perception of automated and human advisers (e.g., Dzindolet, Pierce, Beck, & Dawe, 2002;Dzindolet et al, 2001;Lerch et al, 1997) and the effects of framing on automation trust (e.g., Dzindolet et al, 2002;Lacson, Wiegmann, & Madhavan, 2003), we hypothesized trust in automation to be higher than in humans when portrayed as novices, as automation is perceived as more rational than humans (e.g., Dijkstra, 1999). When both are portrayed as experts, we expected human advisers to be trusted more than automated aids, as portraying a human as an expert will lead to the association of higher dispositional credibility (i.e., degree of trustworthiness based on personal traits) with the human (see Lerch et al, 1997).…”

The Impact of Performance Incentives during Training on Transfer of Learning

“…Incentives, such as giving or taking away points for a hit, miss, correct rejection and false alarm are expected to affect decisions in detection behavior. In this study, we will extend the work by Lacson et al (2005), by examining performance as a result of framing effects when goals are related to incentives (rewarding points versus deducting points) for performing the task correctly or incorrectly. Specifically we are interested in how incentives (or rewards) versus punishment versus no incentives might affect detection behavior and specifically transfer of training.…”

“…Goal framing is structuring information in the form of positive or negative consequences for a behavior or strategy, such as the cost of false alarms versus the benefits of a hit (Levin et al, 1998). In a study on goal framing and automation reliance and compliance, Lacson, Wiegmann and Madhavan (2005) investigated how positive and negative goal frames affected the participants' trust towards an automated aid in a simulated process control task. They found that negatively framed instructions (minimize misses and false alarms) increased reliance (i.e., agreeing when the aid says "no" or remains silent) whereas positively framed instructions (maximize hits and correct rejections) decreased reliance (i.e., disagreeing with the aid when it said "no").…”

“…Second, the design was not completely crossed in that the study did not compare the performance of participants with framed incentives with those that received neutral (equal) frames or no frames at all. Lastly, the task by Lacson et al, (2005) was opaque in that the task was impossible for participants to perform on their own without the assistance of an automated aid, which, we believe, was confounded with the goal frames themselves.…”

Framing and Context Effects in Visual Search Training

“…The framing effect consistently causes risk seeking when one is presented with a negative frame and a risk aversion when one is presented with a positive frame. Applications of framing effect research has been found in aviation multi-tasking (Nygren, 1997), military threat judgment (Perrin, Bennett, Walrath, & Grossman, 2001), and automation use in signal detection (Dzindolet, Pierce, Beck, & Dawe, 2002;Lacson, Wiegmann, & Madhavan, 2005).…”