Abstract
Adaptive automation (AA) refers to human-computer systems in which the “division of labor” and/or the interface between human and machine agents is not fixed but dynamic. Flexible computer aiding of the human operator, dynamic task allocation between the operator and computer systems, or context-dependent information management systems constitute variants of AA. Despite several successful simulations of AA, there are concerns regarding the efficacy of adaptive systems. Most work to date, with a few exceptions, has been conducted on relatively low-fidelity simulations of real-world systems. There is also a need for translating the theoretical promise of AA into practical reality. This panel examined these issues and discussed whether AA systems can be fielded effectively. In particular, the panelists considered how theory-based research on AA can be translated into effective design recommendations for the development of AA systems in the real world.
Get full access to this article
View all access options for this article.