Shifting Expectations for Encoding Rules Mitigates Misinterpretation of Connected Scatterplots.

Area of Science:

• Data Visualization
• Human-Computer Interaction
• Cognitive Psychology

Background:

• Connected scatterplots link time-series data points sequentially.
• Misinterpretation of time direction is common due to unconventional encoding (TIME IS A LINE vs. RIGHT IS LATER).
• Visualization comprehension is influenced by user expectations of encoding rules.

Purpose of the Study:

• Investigate user expectations in connected scatterplot interpretation.
• Develop and test design interventions to improve time direction comprehension.
• Reduce errors in understanding time-series data visualizations.

Main Methods:

• Conducted three preregistered experiments with 1429 participants.
• Implemented visual treatments to suppress incorrect chart-type expectations.
• Introduced directional cues (arrows, trace-line effect, animation) to emphasize correct expectations.

Main Results:

• Visual treatments and directional cues strengthened the TIME IS A LINE expectation.
• Explicit directional cues, especially when redundant, were most effective in reducing misinterpretations.
• Interventions successfully reduced errors in understanding realistic connected scatterplots.

Conclusions:

• Design interventions can significantly improve the interpretability of connected scatterplots.
• Understanding and managing user expectations is crucial for effective data visualization design.
• Findings offer practical guidelines for connected scatterplots and theoretical insights for novel visualizations.