Devalopment of a quantitative measure of display complexity based on decomposition method

Abstract

In an advanced main control room (MCR), computer-based information displays are provided to support operators’ information searching behavior. However, increasing amounts of information contained on the display can lead to an increased level of complexity, resulting in a degradation of human performance. Therefore, it is necessary to evaluate the design complexity of information displays to reduce the inherent design problem. A purpose of this study is to suggest a method to measure the design complexity of computer-based information displays. This study starts by introducing Gestalt theory to explain the concept of complexity. Based on this concept, a decomposition method originally developed for reducing problems related to system design is adopted to measure the complexity of computer-based information displays. The measuring process is mainly divided into three parts. First, a display is represented as a property-component incident matrix and the similarity between the properties and between components is analyzed. After analyzing the similarity, matrix partitioning is implemented depending on the results of a similarity analysis. The complexity is then measured based on the information theoretic definition of complexity with a property-component diagonal matrix. With this process, the calculated results show that the suggested method is feasible for measuring the complexity of computer-based displays. Also, the results of experiments showed that the reaction time and results as measured on a Likert scale are related to the complexity. By using suggested method to measure display complexity, we could get the value of complexity with cognitive aspects because complexity is not only dependent on the number of components but also the properties of each component. Moreover, we could see that the displays with high level of similarity could result in high level of complexity than low level of similarity.