This study investigated the effect of hue on performance under different types of display density. In the past, researchers have primarily conceptualized density in terms of the total number of filled characters on the screen (i.e., overall density). Recently, however, Tullis (1983) suggested this definition was too restrictive. In addition to overall density, he suggested that two other types of density be considered: local density, or the number of characters immediately surrounding the target; and grouping, or the number of perceptual groups formed by the display elements. The present research explored the effect of hue using each of the types of density in three separate experiments.

Subjects were presented several profiles of job applicants on a computer screen and asked to select the applicant with the highest skill score. Density manipulations specific to each experiment were in terms of the number of profiles (Experiment 1- overall density x hue), intra-profile spacing (Experiment 2- local density x hue), and inter-profile spacing (Experiment 3- grouping x hue). In all three experiments, the presence of hue was either relevant or irrelevant to the task. Results of a series of 2 x 2 x 3 ANOVAs showed that the addition of hue had a significant effect on visual search performance. The direction and magnitude of this effect depended on the relevance of hue and the specific manipulation of display density. Findings were explained in terms of the Gestalt principles of organization and several recommendations were made for screen design.