Picture perfect science communication: How public audiences respond to informational labels in cinematic-style 3D data visualization.

Visualizing research data can be an important science communication tool. In recent decades, 3D data visualization has emerged as a key tool for engaging public audiences. Such visualizations are often embedded in scientific documentaries screened on giant domes in planetariums or delivered through video streaming services such as Amazon Prime. 3D data visualization has been shown to be an effective way to communicate complex scientific concepts to the public. With its ability to convey information in a scientifically accurate and visually engaging way, cinematic-style 3D data visualization has the potential to benefit millions of viewers by making scientific information more understandable and interesting. Maximizing the effectiveness of 3D data visualization can benefit millions of viewers. To support a wider shift in this professional field towards more evidence-based practice in 3D data visualization to enhance science communication impact, we have conducted a survey experiment comparing audience responses to two versions of 3D data visualizations from a scientific documentary film on the theme of 'solar superstorms' (n = 577). This study was conducted using a single (with two levels: labeled and unlabeled), between-subjects, factorial design. It reveals key strengths and weaknesses of communicating science using 3D data visualization. It also shows the limited power of strategically deployed informational labels to affect audience perceptions of the documentary film and its content. The major difference identified between experimental and control groups was that the quality ratings of the documentary film clip were significantly higher for the 'labeled' version. Other outcomes showed no statistically significant differences. The limited effects of informational labels point to the idea that other aspects, such as the story structure, voiceover narration and audio-visual content, are more important determinants of outcomes. This study concludes with a discussion of how this new research evidence informs our understanding of 'what works and why' with cinematic-style 3D data visualizations for the public.