Supporting Coordination in Widely Distributed Cognitive Systems: The Role of Conflict Type, Time Pressure, Display Design, and Trust

摘要

Coordination has been defined as the management of dependencies between the goals, tasks, and resources of various agents. Recently, effective coordination between human and machine agents has become increasingly important due to increasing levels of system autonomy and authority. The coordination strategy most often preferred by human operators due to a perceived high level of control over machine actions is called management by consent. Under this approach, the machine is not allowed to act unless and until the operator has given explicit consent to proposed goals and actions. Since preferences do not necessarily translate into superior system performance, and to contribute to a better understanding of successful human automation coordination, this study investigated the effects of conflict type, time pressure, operator trust, and display design on an operator's ability to provide informed consent. The context for this research was (a simulation of) the highly automated cockpit of a modern aircraft. 30 professional B-757 pilots flew a set of 8 descent scenarios while responding to a series of air traffic control clearances. Each scenario presented pilots with a different conflict that would arise either from the goals specified in the clearance or from the implementation of the clearance by the automation.