Man-Machine Communication of the Structure of Engineering Design Problem Information

Abstract

Design theory should identify candidate designers, specify design behavior to be measured, and discriminate between behavior of designers and nondesigners. A study evaluated different formats of computerized structural display and data entry to measure design behavior. A program using graph theory as an internal data structure in the computer and as a display to the operator was developed. Graphics display was used to construct a learning task with directed graphs and graphic data entry. The subjects were given varying portions of the spatial cues ordinarily available in fixed display. Results showed that spatial cues offered through interactive graphics were a significant advantage in the learning of directed graphs. When spatial cues were available, no difference between different learning formats was found. When spatial cues were denied during learning, delay of the posttest was detrimental to performance. Regularly ordered formats of displays of directed graphics led to confusion for some individuals. In general, the use of interactive graphics for learning structural digraphs was better than the use of nonspatial media. (CH)