The use of imagery is commonplace and critical to our understanding of complex subjects, but we rarely stop to consider the thinking behind it. What makes a good or bad infographic? How does the choice of a chart or colours impact the accuracy of the communication? In this article, we explore the principles of visualisation and the use of a new generation of technology that’s transforming our ability to understand and interact with data.

Revealing science in a way that words cannot

Images capture our attention, inform us and engage our imagination, revealing new perspectives.¹ They can open up subjects that are unobservable by the naked eye – from molecules and viruses to the remote corners of our universe.² They can help not only to explain complicated concepts to the general public but also to educate busy scientists and medical professionals.³ They can reveal science in ways that words cannot.¹

The visualisation of science can be divided into two categories:

• Visuals created to entice audiences to engage more fully with related content, for example images used to introduce scientific articles or feature in advertising or promotional campaigns.
• Visuals designed primarily to convey information, commonly referred to as infographics. An example in pharma marketing is the use of charts and diagrams in detail aids, although infographics are also widely employed in many other scientific contexts.

The eyes have it

More of our neurons are dedicated to vision than to the other four senses combined and according to research carried out by MIT neuroscientists, the brain can identify images seen for as little as 13 milliseconds. Consequently, anyone who wants to be successful in reaching us and communicating with us is wise to consider the use of visual imagery.

It is therefore no surprise that we live in a culture in which we are constantly bombarded by images, increasingly those we create ourselves. Images compete for our attention but not all succeed – the majority are screened out by our selective perception. To cut through this screening process, images need to be relevant or unexpected and, if they want to influence us, ideally both. This applies as much to science-related images as to any other. When selected with imagination and creativity, they have the power to capture our attention and interest and lead us to seek further information. This is the primary function of the first category of scientific visualisation: images of this kind make us start reading the article, become involved in the video or approach the exhibition stand. They, together with the words that accompany them, are the first link in the story that follows, a story in which the visualisation of information is a fundamental and intriguing part of bringing science to life.

The growing importance of visualising information and doing it well

It has been estimated that trillions of rows of data are now generated every day.⁴ If we want to make good use of this data, we must first make sense of it and then be able to communicate it to others, which is why the best data scientists not only know how to manipulate data but how to turn it into compelling stories. An important part of this process is the creation of infographics, which range from the figurative to the abstract and embrace representative illustrations, illustrated diagrams and data visualisation.⁵

Data can be said to change how we draw conclusions about the world, whilst data visualisation helps us better understand the data.⁶ Visual elements such as charts, tables, diagrams, graphs, trees, scatterplots, maps, waveforms, simulation and volume provide an accessible way to spot trends, patterns and outliers in data. Data visualisation has therefore been described as a form of visual art that captures our attention and keeps our eyes on the message – storytelling with a purpose.⁷

That said, there are good and bad infographics; there is an art to creating a visual representation of complex information that is both intelligible and visually attractive, drawing in the viewer to the subject area and adding clarity rather than confusion. Unfortunately, science is littered with poor data visualisations that may mislead not only readers but even the scientists who make them. The problem stems in part from scientists receiving little visualisation training, which means that they may take less care with visualising data than generating it or writing about it. It is also a complex field which requires better understanding of the various strengths, weaknesses and biases that human perception brings to the viewing of data.

There is a lot of research already available revealing the types of charts that are the most appropriate, effective and easiest to decipher. For example, if it’s important to show that one particular disease is far more lethal than others, a graphic using the size of circles to represent the numbers of deaths will work well. However, to emphasize much smaller differences in the numbers of deaths among the less-lethal diseases, a bar chart will be far more effective.

Problems can also arise with the use of colour. Although colour can be very useful for highlighting different aspects of data, its application is not entirely straightforward owing to the peculiarities of the human visual system. Our perception of a colour can be influenced by other nearby colours. In some cases the effect is quite dramatic, leading to all sorts of optical illusions. Such issues can affect interpretation of commonly used devices such as the rainbow scale or heat maps.

It has been claimed that these kinds of problems persist because scientists aren’t always aware of them or convinced about making the extra effort, or because they simply tend to follow convention. Scientific journals could help by always following best practice. However, the most effective and practical answer is likely to be the development of tools that incorporate better design principles, enabling scientists to apply them automatically when visualising data.⁸

The future of data visualisation

Visualisation of data was once thought of as a means of better presenting data but now scientists recognise its usefulness in data exploration.⁹ Computers have made a significant difference to the process of data visualisation and, as the data become more and more complex, artificial intelligence is playing an increasing role. The future of the field is about making data visualisation more dynamic, allowing scientists to be more creative as they interact with the data. To enable this, much more use will be made of augmented and virtual reality. Augmented reality (AR) allows engagement with the real word but with digital content added; one example is smart glasses, first brought to light by Google Glass. Virtual reality (VR) allows the user to be completely immersed in a digitally generated environment and is now commonly used in surgery. When AR and VR are used together, the result is referred to as mixed reality (MR).

MR has been described as ground-breaking in data visualisation because it offers a completely different way to represent data and facilitate human and computer interaction. It creates the ability to break away from the size limitations of conventional computer screens and to display data in the same 3D space in which users are physically situated. One example of a field in which it will have significant application is biopharma, where it is likely to help speed up drug discovery.¹⁰

Conclusion

Visualisation plays a much bigger role in bringing science to life than most people realise. It can fire our imagination, as well as educate and inform. Increasingly, it is also doing much more than helping scientists communicate clearly with other scientists or making science more accessible to non-scientists. It is now helping scientists bring science to life for themselves by giving them new insights into the data they produce. This includes using MR to create an immersive experience that allows them to physically adjust the data as they explore new possibilities for creating a better future. It truly is where science and art meet to bring science to life.

1. Antille N. The art of bringing science to life. Science Node. 2020. Available at https://sciencenode.org/feature/The%20art%20of%20bringing%20science%20to%20life.php. Accessed April 2021.
2. Hodges ERS, editor. The Guild handbook of science illustration. 2nd edition. Hoboken, NJ: John Wiley & Sons, Inc.; 2003.
3. Perilli K. Scientific illustration: what is it? The Franklin Institute. 2019. Available at https://www.fi.edu/blog/scientific-illustration-what-is-it. Accessed April 2021.
4. Li Q. Overview of data visualization. In: Embodying data: Chinese aesthetics, interactive visualization and gaming technologies. Germany: Springer Singapore; 2020: Chapter 1.
5. Christiansen J. Visualising Science: Illustration and beyond. Scientific American. 2018. Available at https://blogs.scientificamerican.com/sa-visual/visualizing-science-illustration-and-beyond/. Accessed April 2021.
6. Bohl F. The future of data visualisation. Towards Data Science. 2018. Available at https://towardsdatascience.com/the-future-of-data-visualization-2f976b90b93d. Accessed April 2021.
7. Tableau. Data visualisation beginners guide: a definition, examples and learning resources. Available at
8. Mason B. Why scientists need to be better at data visualisation. Knowable Magazine. 2019. Available at https://knowablemagazine.org/article/mind/2019/science-data-visualization. Accessed April 2021.
9. Owens J. Data visualization innovations in life sciences and drug discovery. Technology Networks. 2018. Available at https://www.technologynetworks.com/informatics/articles/data-visualization-innovations-in-life-sciences-and-drug-discovery-296360 . Accessed April 2021.
10. Limaye N. Data visualization in biopharma: leveraging AI, VR and MR to support drug discovery. Technology Networks. 2019. Available at https://www.technologynetworks.com/biopharma/articles/data-visualization-in-biopharma-leveraging-ai-vr-and-mr-to-support-drug-discovery-320108. Accessed April 2021.