Effect of color properties in multiple time series graph comprehension

Area of Science:

• Data Visualization
• Human-Computer Interaction
• Cognitive Psychology

Background:

• Time series graphs are crucial for data representation in business and research.
• Effective use of color properties in time series graphs for enhanced comprehension is underexplored.
• Understanding how color attributes like hue and lightness affect user performance is vital.

Purpose of the Study:

• To investigate the impact of hue and lightness on the comprehension of 4-time series data.
• To evaluate the effects on response time (RT) and accuracy.
• To compare monochrome and multi-hue color palettes for data visualization.

Main Methods:

• Developed monochrome (red, green, blue) and multi-hue (red, blue, green, purple) palettes.
• Conducted two experiments with 40 participants performing maximum and discrimination tasks.
• Measured response time and accuracy based on different color hues and lightness levels.

Main Results:

• Green monochrome palettes showed superiority in response time and accuracy for the discrimination task.
• Lightness significantly affected performance in the multi-hue experiment.
• Optimal performance (lowest RT, highest accuracy) occurred at 60% lightness; 20% lightness resulted in poorer performance.

Conclusions:

• Green is a highly effective hue for monochrome time series visualizations, particularly in discrimination tasks.
• Lightness is a critical factor in multi-hue time series graphs, influencing both speed and accuracy of data interpretation.
• Optimizing lightness levels (e.g., 60-80%) can significantly improve user performance with time series data.