Adaptive Automation and Human Performance. 3. Effects of Practice on the Benefits and Costs of Automation Shifts

Abstract : Adaptive automation, or adaptive function allocation, is thought to maximize the benefits associated with cockpit automation while maintaining pilot involvement,enhancing situation awareness, and regulating workload. These claims have not been tested empirically. The present study examined the effects of short-cycle adaptive automation and practice on performance-of flight-related functions in a multi-task environment. Twenty four non pilot subjects were tested on a PC-based flight- simulation task that included three primary flight functions -- tracking, monitoring, and fuel management. Each function could be automated or performed manually. The results provide preliminary evidence that dynamic automation shifts over short cycles, of the type likely in adaptive systems, benefit performance of flight- related tasks, with no evidence of costs to performance following the return to manual control. Benefits are realized despite the added workload of supervisory control of automated functions. However, training procedures other than simple practice may be necessary to maximize and maintain the performance benefits associated with adaptive automation.

[1]  Tsang Ps,et al.  Cognitive demands in automation. , 1989 .

[2]  W. T. Singleton,et al.  Man-machine systems , 1974 .

[3]  Stanley M. Halpin,et al.  Adaptive User Interfaces for Planning and Decision Aids in C3I Systems , 1986, IEEE Transactions on Systems, Man, and Cybernetics.

[4]  Christopher D. Wickens,et al.  The Impact of Automation on Error Detection: Some Results from a Visual Discrimination Task , 1987 .

[5]  Michael A. Vidulich,et al.  The benefits and costs of automation in advanced helicopters - An empirical study , 1989 .

[6]  D. Navon Resources—a theoretical soup stone? , 1984 .

[7]  Raja Parasuraman,et al.  Effects of shifts in the level of automation on operator performance , 1991 .

[8]  Daniel G Bobrow,et al.  On data-limited and resource-limited processes , 1975, Cognitive Psychology.

[9]  Jacqueline R. Idaszak,et al.  Active Participation in Highly Automated Systems: Turning the Wrong Stuff Into the Right Stuff , 1989 .

[10]  Christopher D. Wickens,et al.  Failure Detection in Dynamic Systems , 1981 .

[11]  Daniel Gopher,et al.  Adaptive training of perceptual-motor skills: issues, results, and future directions , 1978 .

[12]  Marion P. Kibbe Information Transfer from Intelligent EW Displays , 1988 .

[13]  John E. Deaton,et al.  Theory and Design of Adaptive Automation in Aviation Systems , 1992 .

[14]  B. Rouse William,et al.  Adaptive Aiding for Human/Computer Control , 1988 .

[15]  Raja Parasuraman,et al.  Human-Computer Monitoring , 1987 .

[16]  Daniel Gopher,et al.  On the Economy of the Human Processing System: A Model of Multiple Capacity. , 1977 .

[17]  Mark H. Chignell,et al.  Mental workload dynamics in adaptive interface design , 1988, IEEE Trans. Syst. Man Cybern..

[18]  David C. Nagel,et al.  Pilots of the future: human or computer? , 1985, CACM.