Informed consent in distributed cognitive systems: the role of conflict type, time pressure, and trust

摘要

The introduction of automation technology to a variety of highly complex dynamic domains such as aviation has led to unexpected difficulties with human-machine coordination (Sarter, Woods, and Billings, 1997). To date, attempts to address thesedifficulties have focused primarily on modifications of operator training or on improvements of feedback design. Another way of supporting coordination - the context-sensitive use and implementation of different automation management strategies - hasreceived less attention. Results of a recent survey (Olson and Sarter, 1998) show that pilots overwhelmingly prefer a management-by-consent approach due to a perceived higher level of control over machine actions since, under this approach, the automation can not take action unless and until it receives explicit pilot consent (Billings, 1997). Because operator preferences do not necessarily translate into superior performance, a subsequent simulator study was conducted to examine the impact of factors such as time pressure, display design, and conflict type, and trust on pilots' ability to give informed consent, i.e., consent based on a full understanding of the current and future implications of proposed machine actions (Billings, 1997).