Do Graph Readers Prefer the Graph Type Most Suited to a Given Task? Insights from Eye Tracking

Research on graph comprehension suggests that point differences are easier to read in bar graphs, while trends are easier to read in line graphs. But are graph readers able to detect and use the most suited graph type for a given task? In this study, we applied a dual repre-sentation paradigm and eye tracking methodology to determine graph readers’ preferential processing of bar and line graphs while solving both point difference and trend tasks. Data were analyzed using linear mixed-effects models. Results show that participants shifted their graph preference depending on the task type and refined their preference over the course of the graph task. Implications for future research are discussed.

Problem-Independent Approach to Multiprocessor Dependent Task Scheduling

This paper concerns Directed Acyclic Graph task scheduling on parallel executors. The problem is solved using two new implementations of Tabu Search and genetic algorithm presented in the paper. A new approach to solution coding is also introduced and implemented in both metaheuristics algorithms. Results given by the algorithms are compared to those generated by greedy LPT and SS-FF algorithms; and HAR algorithm. The analysis of the obtained results of multistage simulation experiments confirms the conclusion that the proposed and implemented algorithms are characterized by very good performance and characteristics.

Graph-Task Dependencies in Three-Dimensional Data: Influence of Three-Dimensionality and Color

This study compared conventional 2D graphs with 3D graphs and color based graphs for presenting 3-dimensional data. These data were in the format representing the effects on Y, of 4 levels of X and 4 levels of Z. Z was represented by a parameter (line type) in the 2D display, by space (depth) in the 3D display, and Y was represented by color in the color display. Subjects answered questions about the displayed data that varied in the degree to which they required focused attention on a single data point, to integration across the entire data space. The results indicated that the 3D display supported slowest and least accurate performance for the focused attention questions, a cost that dissipated when the questions became more integrative. Performance with the color display suffered badly in both speed and accuracy with the most integrative questions. The 2D display performed consistently well in both speed and accuracy. The results are consistent with prior data and with emerging theoretical perspectives on graph-task dependencies.