Measures of Information Complexity and the Implications for Automation Design.

摘要

Information complexity associated with automation aids is a bottleneck that limits their use. While automation systems are designed to bring new functions to users and increase their capacities, automation also creates new tasks associated with acquiring and integrating information from displays. For example, a complex display increases information load to human operators and reduces usability. Thus, the efficiency of an automation system largely depends on the complexity of displayed information. To evaluate the costs and benefits of an automation aid, it is important to understand how much information is shown on the display, how users look at multiple information sources to build and maintain situation awareness, and whether the information is displayed in a compatible way so it can be integrated and understood easily without the user having to make internal conversions or calculations. In this paper, we present a set of measures to assess information complexity. The metrics count information complexity as the combination of three basic factors: numeric size, variety, and relation; each factor is evaluated by the functions at three stages of brain information processing: perception, cognition, and action. Ideally, these measures provide an objective method to evaluate automation systems for acquisition and design prototypes.