Previous research has found the use of graphical Decision Support Systems (DSS) to be more effective than tabular DSS in some decision situations, but not in others [8, 9]. This paper presents the results of two laboratory experiments testing the hypothesis that the features provided by graphical DSS may best "fit" some tasks, while those of tabular DSS best fit other tasks. The first experiment, which examined decision outcomes, found that a tabular DSS better fit a less complex task, while a graphical DSS better fit a more complex task with high information load in which decision makers needed to understand relationships among data elements. The second experiment, which examined decision processes, found that decision makers using graphical DSS tended to use less information in making their decisions than those using tabular DSS.