[Paper Review] Improving Engagement of Animated Visualization with Visual Foreshadowing

Wenchao Li, Yun Wang+2 more|arXiv (Cornell University)|Sep 8, 2020

Video Analysis and Summarization28 references4 citations

TL;DR

This paper introduces visual foreshadowing—a narrative technique adapted from film and literature—to improve viewer engagement in animated visualizations of time-series data. By embedding subtle visual cues in advance of significant data changes, the approach enhances anticipation, attention, and cognitive involvement, as validated by a user study showing significantly higher engagement scores with foreshadowing compared to baseline animations.

ABSTRACT

Animated visualization is becoming increasingly popular as a compelling way to illustrate changes in time series data. However, maintaining the viewer's focus throughout the entire animation is difficult because of its time-consuming nature. Viewers are likely to become bored and distracted during the ever-changing animated visualization. Informed by the role of foreshadowing that builds the expectation in film and literature, we introduce visual foreshadowing to improve the engagement of animated visualizations. In specific, we propose designs of visual foreshadowing that engage the audience while watching the animation. To demonstrate our approach, we built a proof-of-concept animated visualization authoring tool that incorporates visual foreshadowing techniques with various styles. Our user study indicates the effectiveness of our foreshadowing techniques on improving engagement for animated visualization.

Motivation & Objective

• To address the challenge of low viewer engagement in long-form animated visualizations due to attention fatigue and lack of focus.
• To adapt narrative foreshadowing from film and literature as a design strategy for animated data visualizations.
• To develop and evaluate visual foreshadowing techniques that guide attention and build anticipation for upcoming data changes in animated bar charts.
• To create a proof-of-concept authoring tool enabling easy integration of visual foreshadowing effects into animated visualizations.
• To empirically assess the effectiveness of visual foreshadowing in improving user engagement, attention, and cognitive involvement.

Proposed method

• Formal definition of visual foreshadowing as a design technique that uses visual cues in advance of data changes to build viewer anticipation.
• Design of four visual foreshadowing techniques: Pre-scene, De-emphasis, Highlight, and Timeline, categorized into explicit and implicit types.
• Implementation of a proof-of-concept authoring tool that supports creation, previewing, and editing of visual foreshadowing effects on temporal ranking data.
• Application of visual foreshadowing to animated bar charts, particularly bar chart races, to illustrate changes in rankings over time.
• Use of timing and duration controls to regulate when and how long foreshadowing cues appear before key data events.
• Conduct of a user study comparing animated visualizations with and without visual foreshadowing across engagement, attention, and cognitive involvement metrics.

Experimental results

Research questions

• RQ1 How can visual foreshadowing be effectively applied to animated visualizations to improve viewer engagement?
• RQ2 What types of visual cues (explicit vs. implicit) are most effective in guiding attention and building anticipation in animated data stories?
• RQ3 How does the timing and duration of visual foreshadowing affect viewer perception and engagement?
• RQ4 What is the impact of visual foreshadowing on focused attention, enjoyment, and cognitive involvement in animated visualizations?
• RQ5 Can visual foreshadowing be generalized across different animated visualization types beyond bar charts?

Key findings

• Animated visualizations with visual foreshadowing achieved significantly higher engagement scores, with a mean rating of 6.54 (SD 0.54) compared to 4.13 (SD 0.61) without foreshadowing.
• Viewers reported greater focused attention when foreshadowing was used, with a mean score of 6.17 (SD 0.72) versus 3.71 (SD 1.03) in the baseline condition.
• Cognitive involvement was substantially higher in the foreshadowing condition, with a mean rating of 6.04 (SD 0.58), compared to 4.50 (SD 1.04) without it.
• Participants preferred implicit foreshadowing (e.g., De-emphasis) over explicit cues (e.g., Pre-scene), as it created stronger anticipation and a more satisfying payoff at the end of the animation.
• Users suggested that excessive use of foreshadowing could be overwhelming, and that cues should be placed with sufficient lead time before the event to be effective.
• The study indicates that visual foreshadowing effectively enhances narrative flow and viewer retention in animated visualizations of temporal data.