5 lessons for labelling 3D Data visualization for public engagement


The importance of accurate labels and annotations is drilled into researchers early in their careers, but the role they play in cinematic data visualizations aimed at broader audiences is rarely addressed. Eric A. Jensen, Kalina Borkiewicz, Jill P. Naiman, Stuart Levy and Jeff Carpenter outline five lessons on how labels can be usefully deployed on cinematic data visualizations and how this can inform evidence-based science communication practices.

Data visualization is a broad field that includes everything from a flat drawing of a pie chart to a 3D flight through the cosmos in virtual reality. It is obvious to (most) people that a bar chart or a scatterplot needs axes and labels to be useful. In this way, labels are vital, but they can also require domain-specific knowledge for them be meaningful for a viewer.

However, when a data visualization is, say, a cinematic 3D rendering of the sun, designed for a non-expert audience, are annotations necessary? Or are they a distraction in an otherwise-immersive virtual environment? You may be surprised, but there has been little research on this topic.

It is increasingly recognised that science communication practice should employ an evidence-based approach to deliver impact and not just make “default” decisions based on what the scientist or data visualizer considers to be the best choices. However, we lack concrete examples of what evidence-based science communication looks like and how this translates into a set of “best practices”, especially for cinematic visualizations. To this end, it is important to clarify what kind of evidence can be used to inform practice and how it can be used. Here we present findings from our multidisciplinary team in relation to implementing evidence-based science communication in the visualization of scientific data for public audiences.

The Advanced Visualization Lab (AVL) at the University of Illinois at Urbana-Champaign is a multidisciplinary team that has created several widely distributed documentaries and planetarium shows. The AVL team has typically relied on their decades of experience to make decisions on labels and annotations in their cinematic-style visualizations. Recently, the AVL has shifted its focus towards evidence-based practice. To aid this process we developed an online survey study to test the intelligibility of informational labels designed by the AVL team for the cinematic-style 3D data visualizations featured in the film Solar Superstorms.

This film explores the Sun’s inner workings and features visualizations computed on the giant supercomputing initiative Blue Waters. The primary focus of the audience research was to add labels to specific visualizations within this documentary to identify problematic labels and gain insights into how to improve them. A secondary focus was to demonstrate on a practical level how the evidence-based science communication process works. You can find more detail in our open-access article, published in Sustainability. It aims to begin to address the practicalities and effectiveness of different kinds of labels.

What we found

Visualizations used in this study spanned a range of environments in space – from the Sun all the way to the large-scale formation of stars. Originally, these visualizations were shown without labels, as is normal in the field of cinematic visualization. The study involved an online survey in which participants rated the overall quality of the data visualizations in the film. Impressively, the visualizations received an average rating of 84.95 out of 100, indicating a positive overall impression. Notably, the participants appreciated how the informational labels interacted with other elements of the film, such as the voiceover narration, contributing to a comprehensive viewing experience.

One of the labels that garnered significant praise was “Earth to scale,” which effectively provided viewers with a sense of perspective by illustrating the size of the Sun relative to our planet.

Visualization showing the sun with a label showing the relative size of the Earth.

The use of simplified colour tables, such as “cool gas” and “hot gas,” also received positive feedback as they made the information more approachable for non-expert audiences.

Visualization representing cool gas (blue) and hot gas (orange) in space..

However, the study also revealed areas for improvement. Some informational labels, particularly those indicating time, raised concerns among participants regarding their precision. Valuable feedback from the survey participants suggested potential solutions, such as syncing the timescale label with the unfolding of the scene to provide a more precise and intelligible representation of temporal dynamics.

Visualization showing swirling plasma on the sun with label showing calendar time represented in the image.

In response to audience feedback, the AVL team revised the labels. For example, they replaced a static outline of Australia, which served as a foreground element layered on top of the image, with an outline of the Earth embedded in the visualization. This dynamic label, scaling with the animation, effectively conveyed scale and provided important depth cues, ultimately enhancing the clarity and intelligibility of the visualization.


Visualization with outline of Australia to scale.

After:Visualization with outline of the Earth to scale.

Lessons Learned for Better Labelling

We drew five valuable lessons from this work that point towards more effective and evidence-based science communication practices:

  1. Strategic Placement of Annotations: Carefully position annotations within the visualization to provide contextual information without overwhelming the immersive experience, using depth cues where applicable. Consider using annotations that are unobtrusive yet easily accessible to viewers, guiding their understanding without distracting from the overall visualization.
  2. Utilise Familiar Reference Points: Integrate relatable objects or concepts, such as the Earth or common landmarks, to establish a sense of scale and perspective within the visualization. Familiar reference points help audiences grasp the magnitude and significance of depicted phenomena.
  3. Synchronise Timescale Labels: Ensure that temporal information aligns seamlessly with the unfolding of the visualization, allowing viewers to comprehend the dynamic nature of the depicted processes accurately. Syncing timescale labels with the animation provides a more precise representation of temporal dynamics.
  4. Adapt Labels to Audience Familiarity: Tailor labelling choices to suit the knowledge level and interests of the target audience, simplifying where necessary. By using language and concepts familiar to the viewers, practitioners can bridge the gap between complex scientific data and the understanding of non-expert audiences.
  5. Incorporate Audience Feedback: Engage with viewers through surveys, user testing, and feedback sessions to gain insights into the effectiveness of labelling techniques. Utilise audience feedback to refine and optimise annotations, ensuring they resonate with the intended audience and contribute to a more impactful science communication experience.

By employing these techniques, practitioners can unleash the full potential of labelling in 3D data visualizations, fostering greater comprehension, engagement, and appreciation in a way that is informed by evidence rather than guesswork.


The content generated on this blog is for information purposes only. This Article gives the views and opinions of the authors and does not reflect the views and opinions of the Impact of Social Science blog (the blog), nor of the London School of Economics and Political Science. Please review our comments policy if you have any concerns on posting a comment below.

Image Credit: All images reproduced with permission of the authors.

Print Friendly, PDF & Email


Leave a Reply

Your email address will not be published. Required fields are marked *