Understanding Visual Cues in Visualizations Accompanied by Audio Narrations
Authors
It is often assumed that visual cues, which highlight specific parts of a visualization to guide the audience's attention, facilitate visualization storytelling and presentation. This assumption has not been systematically studied. We present an in-lab experiment and a Mechanical Turk study to examine the effects of integral and separable visual cues on the recall and comprehension of visualizations that are accompanied by audio narration. Eye-tracking data in the in-lab experiment confirm that cues helped the viewers focus on relevant parts of the visualization faster. We found that in general, visual cues did not have a significant effect on learning outcomes, but for specific cue techniques (e.g. glow) or specific chart types (e.g heatmap), cues significantly improved comprehension. Based on these results, we discuss how presenters might select visual cues depending on the role of the cues and the visualization type.
Research Questions / Practical Problems
Question signals indexed for this paper.
- 60%
Ranked-List Visualization: A Graphical Perception Study
CHI '19· Data Storytelling +1
- 60%
Neighborhood Perception in Bar Charts
CHI '19· Uncertainty Visualization +1
- 60%
Pushing the (Visual) Narrative: The Effects of Prior Knowledge Elicitation in Provocative Topics
CHI '20· Data Storytelling +1
- 60%
Putting scientific results in perspective: Improving the communication of standardized effect sizes
CHI '22· Data Storytelling +1
- 60%
Round Numbers Can Sharpen Cognition
CHI '22· Data Storytelling +1
- 60%
Towards Design Guidelines for Effective Health-Related Data Videos: An Empirical Investigation of Affect, Personality, and Video Content
CHI '22· Data Storytelling +1
- 60%
Negative Emotions, Positive Outcomes? Exploring the Communication of Negativity in Serious Data Stories
CHI '22· Data Storytelling +1
- 60%
PAIRcolator: Pair Collaboration for Sensemaking and Reflection on Personal Data
CHI '25· Data Storytelling +1
Based on Jaccard similarity of research subtopics & professions (≥60%)