Evaluating Multi-User Selection for Exploring Graph Topology on Wall-Displays.

Area of Science:

• Human-Computer Interaction
• Information Visualization
• Collaborative Systems

Background:

• Wall-displays enable multi-user analysis of large datasets, crucial for complex domains like biology and social network analysis.
• Effective collaborative techniques are needed to leverage wall-display capabilities for tasks involving large graph exploration.

Purpose of the Study:

• To compare the impact of two selection techniques (basic localized and propagation) on collaborative graph exploration using wall-displays.
• To assess how selection techniques influence task performance, specifically shortest path identification, in pairs.

Main Methods:

• A controlled experiment was conducted with pairs analyzing graphs on a touch-enabled wall-display.
• Two selection techniques were implemented: basic localized selection and propagation selection.
• Performance was evaluated based on accuracy and speed for a shortest path identification task.

Main Results:

• Pairs tended to divide workspace spatially, even when tasks did not require it.
• Basic localized selection resulted in parallel work and lower accuracy.
• Propagation selection, with its larger visual footprint, fostered closer coordination and improved speed and accuracy for complex graphs.

Conclusions:

• The choice of selection technique significantly impacts collaborative graph exploration on wall-displays.
• Propagation selection enhances coordination and performance for complex graph analysis tasks.
• Understanding pair interaction strategies, like spatial division, is key to designing effective collaborative visualization tools.