Jan 30, 2023
Accessibility in scientific visualizations
When creating scientific figures, accessibility is an important consideration to ensure that the information is easily understood by all viewers, including those with visual impairments or other disabilities.
General advice for creating accessible visualizations: Here, we start with some hands-on practical advice to ensure accessibility of scientific visualization that we encourage to implement in the ReTune consortium:
- Text labels: Text labels should be large enough and placed close to the data in a clear, high-contrast font. This measure is crucial to make results accessible to people with any visual disability.
- Color: As further discussed below, color can be an important factor in making scientific figures accessible. It is important to use colors that have sufficient contrast and are easily distinguishable. Colormaps should avoid green-red contrasts to account for people with different color perception or color vision deficiency, here Viridis is a common and appreciated choice. Color palettes should avoid red, especially in combination with green, contrasting blue, orange and yellow can be used. For people who cannot perceive color, it can be helpful to check and provide alternative versions of the figures, such as grayscale or black and white.
- Simplicity: Scientific figures should be simple and easy to understand. Avoid using overly complex or cluttered figures, as they can be difficult for viewers to process.
- Data representation: Data should be represented in a clear and consistent manner. Avoid using unnecessary 3D representations or other complex visualizations that can be difficult for some viewers to understand.
- Alt-text: Alt-text is a text description that can be added to an image, and it is especially helpful for viewers who use screen readers. Alt-text should provide a concise description of the figure and its main findings.
- Interactivity: Interactive figures can provide a more engaging and accessible way to allow integrate both complexity and simplicity when presenting scientific data. It can allow viewers to explore the information in more detail and to better understand the findings.
Colormaps & Color vision deficiencies: Colormaps are an essential tool for visualizing scientific data, often used for two-dimensional heatmaps, neuroimaging and brain connectivity analyses. However, not all colormaps are created equal when it comes to accessibility. The way in which colors are represented can make a significant difference for individuals with color vision deficiencies, or for those who are visually impaired. Color vision deficiency, also known as color blindness is a condition that affects the way in which an individual perceives color. It is important to note that color blindness is a variation in the way the eyes perceive color and not a disease. The severity of this altered color perception can vary, with some individuals experiencing only mild differences, while others may experience stronger differences. The most common form of color blindness is red-green color blindness, which is caused by a genetic mutation on the X chromosome, affecting people in Europe with a single X chromosome more frequently (~8%) than people with two X chromosomes (~0.5%). In addition to genetic causes, there are also acquired forms of color blindness, which can be caused by diseases or injuries to the eyes, or by certain medications. These other forms of color blindness, such as blue-yellow color blindness can affect people of all genders and sex. It is also worth noting that color blindness can also occur as a side effect of aging, with elderly people more likely to experience difficulty distinguishing colors. When creating scientific colormaps, it is important to consider the needs of those who may have difficulty perceiving color. One way to do this is to ensure that the colormap has sufficient contrast between different colors, so that they can be distinguished easily. Additionally, it is important to use colors that are easily distinguishable, such as avoiding using shades of red and green together, which can be difficult for individuals with red-green color blindness to distinguish. Another important consideration is to ensure that the colormap is perceptually uniform, meaning that the perceived change in color corresponds to the actual change in the data being represented. This ensures that the colormap accurately represents the data, and that it can be understood by all viewers regardless of their color vision. In addition to these considerations, it can also be important to provide alternatives to the colormap, such as grayscale or black and white versions, for those who may have difficulty perceiving color.
There are many tools available that can help simulate color vision deficiencies for quality control of your work, such as Color Oracle | Color Oracle or Coblis — Color Blindness Simulator – Colblindor (color-blindness.com)
Accessibility of scientific figures is a crucial aspect of data visualization and communication. By considering factors such as text labels, color, simplicity, data representation, alt-text and interactivity, we can create figures that are easily understood and accessible to all viewers. To make figures accessible for people with color vision deficiency, color palettes and colormaps should be designed accordingly. By taking into account the needs of those who may have difficulty perceiving color, and providing alternative versions, we can ensure that the data is accurately represented and easily understood by all viewers.
Further tools and reading on data visualization and accessibility:
The misuse of colour in science communication | Nature Communications
Colour me better: fixing figures for colour blindness (nature.com)
Colour blindness: journals should enable image redisplay (nature.com)
Color blind friendly palettes for data visualizations with categories (biologists.com)
Points of view: Color blindness | Nature Methods
