Toward a model for effective human-automation interaction: The mediated agency

Abstract

In our increasingly technological world, automation largely improved some aspects of our life. Nonetheless, automation can also have negative consequences. Indeed, operators seem often helpless to takeover an automated system in case of failure. This “out-of-the-loop” problem occurs when operator is unable to understand the intentions and to predict the outcome of actions of the system, causing a decrease of control. The following article illustrates how the psychological approach of agency can help (1) to better understand this OOTL performance problem and (2) to propose design principles to improve human machine interaction in case of system automation.