Visual features as carriers of abstract quantitative information.

Area of Science:

• Visual perception
• Quantitative cognition
• Human-computer interaction

Background:

• Basic visual features are fundamental to how humans perceive and interpret information.
• Understanding how abstract quantitative data is encoded visually is crucial for effective data visualization and communication.

Purpose of the Study:

• To investigate the capacity of basic visual features (luminance, color, orientation, size) to convey abstract quantitative information, specifically Pearson correlation.
• To determine if the perception of correlation using different visual features relies on a common abstract representation.

Main Methods:

• Four experiments presented graphical representations where data dimensions were encoded by element position and a visual feature.
• Observers estimated and discriminated Pearson correlation in these representations.
• Performance metrics included just noticeable difference and estimated correlation as a function of feature noise.

Main Results:

• All tested visual features (luminance, color, orientation, size) effectively supported correlation perception.
• Performance was similar across features, with just noticeable difference being linear and estimated correlation logarithmic with distance from perfect correlation.
• Performance variations were primarily due to feature noise levels, aligning with channel capacity estimates.

Conclusions:

• Abstract quantitative information, like correlation, can be reliably conveyed by low-level visual features.
• This perception likely involves an abstract parameter space where visual dimensions are normalized, independent of specific sensory properties.
• Findings have implications for designing more effective and intuitive data visualizations.