Agent Transparency, Situation Awareness, Mental Workload, and Operator Performance: A Systematic Literature Review

Abstract

Objective In this review, we investigate the relationship between agent transparency, Situation Awareness, mental workload, and operator performance for safety critical domains. Background The advancement of highly sophisticated automation across safety critical domains poses a challenge for effective human oversight. Automation transparency is a design principle that could support humans by making the automation’s inner workings observable (i.e., “seeing-into”). However, experimental support for this has not been systematically documented to date. Method Based on the PRISMA method, a broad and systematic search of the literature was performed focusing on identifying empirical research investigating the effect of transparency on central Human Factors variables. Results Our final sample consisted of 17 experimental studies that investigated transparency in a controlled setting. The studies typically employed three human-automation interaction types: responding to agent-generated proposals, supervisory control of agents, and monitoring only. There is an overall trend in the data pointing towards a beneficial effect of transparency. However, the data reveals variations in Situation Awareness, mental workload, and operator performance for specific tasks, agent-types, and level of integration of transparency information in primary task displays. Conclusion Our data suggests a promising effect of automation transparency on Situation Awareness and operator performance, without the cost of added mental workload, for instances where humans respond to agent-generated proposals and where humans have a supervisory role. Application Strategies to improve human performance when interacting with intelligent agents should focus on allowing humans to see into its information processing stages, considering the integration of information in existing Human Machine Interface solutions.

Exploring navigator roles and tasks in transitioning towards supervisory control of autonomous collision avoidance systems

“…Proper information provision and scenario training have been proposed to mitigate these effects [26]. Also, transparency of the system may be important to allow for understanding of the automation's information processing, decision making and future actions [26][27][28]. As such, further work will focus on detailing the navigator's information needs…”

Navigators’ views of a collision avoidance decision support system for maritime navigation

“…While the operator is in full control of the decision selection (Level 3) and action implementation (Level 4), communication to the user about how, what and why information is being presented, also known as system opacity, is important (Lee, 2008; Helldin, 2014; Westin et al, 2016). In a recent review paper exploring empirical evidence of automation transparency, for ‘agent-generated proposals’, where agent is interchangeable with automation, there appears to be a positive effect of automation transparency on operator performance, SA, and mental workload (van de Merwe et al, 2022). Increased transparency should provide navigators with an increased understanding and predictability of the system to better evaluate whether to use this information instead of trying to evaluate why it was presented (Endsley, 2017; van de Merwe et al, 2022).…”

Automation transparency and the design of intelligent aircraft engine interfaces

“…There is debate within the scientific community as to whether automation transparency can provide performance benefits or improve trust in such systems (van de Merwe et al, 2022). It is not the case that for all tasks an increased level of automation transparency leads to better performance or trust.…”

“…Automation transparency is a concept in user interface design that is of particular significance when artificial intelligence is used to inform and guide user behaviour. A transparent interface ‡ www.boeing.com/resources/boeingdotcom/compa ny/about_bca/pdf/statsum.pdf would offer insight into system reasoning making an intelligent system more understandable to the operator (Rajabiyazdi & Jamieson, 2020;Skraaning & Jamieson, 2021;van de Merwe et al, 2022). Interface choices need to be made since along this continuum there exists the risk of providing either too little or too much information to support the specific task (Endsley, 2017).…”