Interactive Effects of Combined Visual Variables on Search Performance in WebGIS Interfaces

Visual search for targets in webGIS interfaces plays a vital role in human-computer interaction. In this study, the Analytical Hierarchy Process method was used to screen visual variables commonly applied in webGIS, and a total of six visual variables were selected: hue, transparency, shape, size, stroke, and shadow. A behavior experiment was conducted to investigate the interactive effects of combinations of these visual variables on search performance. A total of 18 participants were invited to perform an icon counting task in the webGIS interface. Participants' reaction times and task accuracy were measured as a function of the following independent variables: 15 combinations of visual variables, five levels of target number, and two types of background images. The results showed that combining visual variables reduced search performance and that variables with higher visual saliency were dominant for task accuracy. The four combinations of hue and shape, hue and size, stroke and transparency, and stroke and size performed best in time-pressured tasks, while the shadow style significantly reduced performance when combined with any of the other five variables. Task accuracy improved significantly when the dominant visual variable carried critical information. The findings of this study provide valuable guidelines for designers and developers to develop more efficient and usable webGIS interfaces.