About 240 data scientists have proven themselves well enough in their work to become certified instructors for ‘TDivers’, the R programming language for manipulating and visualizing data. JooYoung Seo is unique among them – the first certified blind teacher.
Most net users present data in the form of charts and graphs. Seo, an information and learning scientist at the University of Illinois at Urbana–Champaign who lost his sight at age ten due to glaucoma, uses touch, voice and speech. He is one of a small but growing group of researchers working to make science accessible to people with limited vision. “The overall challenge is that content is designed visually,” says Seo. But visualization is only one of the modes of representation. We can represent information in many modal ways.
There are no good estimates of how many scientists with low vision are working today, but a 2020 study1 In the year In 2018, fewer than 100 of the 52,124 researchers identified as having a visual impairment received funding from the US National Institutes of Health (NIH). “It’s a fraction of a fraction,” said Bonnylin Swenor, director of the Center for Disability Research at Johns Hopkins University in Baltimore, Maryland. This is far less than the number of scientists with visual impairments, Swenor added, because affordability barriers prevent many scientists from applying for research grants. And it’s a smaller fraction of the total number of visually impaired individuals. In 2017, approximately 7 million people (2.17 percent of the population) in the United States were living with ‘uncorrectable’ visual impairment or blindness.2. “If the goal is to equal the spread in the U.S. — which I’m arguing it should be — we’re not close,” she said.
Eliminate problems
That’s partly because, often, the most basic research tasks—finding articles, submitting papers, attending conferences, and reviewing manuscripts—require time-consuming, customized solutions. “Blind scientists may have a way that works for them,” said Daniel Hajas, innovation manager at the Global Disability Innovation Center at University College London, which works to speed assistive technologies to market. But it’s not comprehensive and it took them a long time to get there.
She remembers a mentor telling her early in her career that she had to work four times as hard as her sighted colleagues to succeed in science. “This shouldn’t be anyone’s way,” she says. But awareness of the importance of access to science is growing, and new tools are on the way.
Paint better: Editing images for color blindness
People with very low display capabilities generally use a screen reader, which converts digital text into synthesized speech or a touch (braille) display, to read large print.
Screen readers navigate digital documents using tags that identify elements such as headers, figures, tables, and footnotes. But the tools often fail on PDF files, which have long been the default digital format for journal articles and other research materials, because they generally do not have such labels. The two-column format widely used in magazines can be confusing for screen readers who generally read from left to right on a page. It is possible to make PDFs accessible, but it is challenging to do and many publishers are not yet there. “Improving PDF accessibility requires changes in culture, systems and processes that can be challenging for publishers,” says Judd Robinson, global head of Springer Nature Digital in London. “But we are committed to doing this.” (Nature(The news team is editorially independent from the publisher, Spring Nature.)
An An online tool called SciA11yDeveloped by the Allen Institute for Artificial Intelligence (AI2) in Seattle, Washington, it uses machine learning to extract the content and structure of PDFs (and other file formats, such as LaTeX) and recreate them in HTML to create tables. Content that contains links to tagged section topics that are navigable with screen readers. AI2 has built in functionality for two-way navigation between online citations and their related references, says Jonathan Bragg, who leads the project.
Detailed survey3, six visually impaired scientists repeatedly described how they were unable to access or read PDFs. One respondent said that they had problems two-thirds of the time and used at least six different ways to read papers. “It was eye-opening to see the different tools that people use when reading and the struggles that those papers face when they’re not formatted properly,” Bragg said.
Speaking in code: How to program with voice
Currently, SciA11y — ‘a11y’ is Internet shorthand for accessibility — is an online demo: researchers can upload PDFs for reworking in HTML. But the team is still working on key functions, Bragg says. For example, the software still makes mistakes, sometimes it fails to extract titles and leaves them as body text in HTML. It also struggles with tables and images, which is the current focus of development.
In this respect, SciA11y is no exception: for visually impaired people, tables and figures present more difficult challenges to solve than text itself. Precious little scientific texts — in PDF, HTML, or other formats — contain numerical descriptions, called alt text, that allow people who are blind or have low vision to understand the images. Moreover, it is not always clear what kind of explanation will be most useful. “There are different schools of thought,” said Amy Bower, a blind physical oceanographer at the Woods Hole Oceanographic Institution in Massachusetts. “Should you just describe what’s there or add meaning?”
Some researchers, including Bower, solve this problem by working with visual translators. Hajas is collaborating with researchers at the Massachusetts Institute of Technology in Cambridge and others to design it. Oli, a screen reader interface This allows users to move up and down different levels of description – from a single sentence explaining the basic premise of an image, to axes and legends with more detailed features, along with the actual data values it’s made from. Olly currently supports basic chart types like bar charts and scatters, and Hajas says the team is working on new features like heat maps.
SEO is developing a data visualization tool, and it is drawing more than words. The device, called the Multimodal Access and Interactive Data Representation System (MAIDR), encodes data as both sounds – sonification – and Braille, providing tactical analysis with the help of a renewable Braille display. “I can hear trends in the data and feel patterns,” says Seo.
Accessing information through touch can be especially powerful, says Mona Minkara, a computational chemist and bioengineer at Northeastern University in Boston, Massachusetts, who began losing her eyesight as a child. Minkara collaborated on the 2022 study.4 3D printed graphics called lithophanes that represent data. Manufactured from plastic thin enough to let light shine through, lithophanes can hide a wide variety of chemical and life science information—for example, a scanning electron micrograph of a butterfly wing, electrophoresis gel bands, or the ultraviolet spectrum of a protein. The technology allows her and her sighted lab colleagues to work on the same data at the same time, she says.
NatureTech Center
Sensing a change in protein function in your fingers, like a difference in thickness, for example, “increases cognitive function,” she says. “It becomes embedded in your imagination and becomes more a part of you, so you understand it on a deeper level.”
Sonivision gives scientists with limited vision another way to ‘see’ data, for example in astronomy. “We’re used to thinking that astronomy is a visual science, but most of the information we get is just numbers,” says Anita Zanella, an astronomer at Italy’s National Institute of Astrophysics in Rome. “We translate them into images because that’s how we make sense of them,” but ears are often better at detecting faint signals. By translating numerical values into sounds with specific parameters—for example, the brightness of a star can be converted to sound—researchers can detect important changes.5.
Researchers in genomics And Geology Although common principles and standards that allow scientists to compare data across auditory formats are still under development, they are exploring sonification, Zanella says. Still simple online tools, e.g Highcharts Sonification StudioDeveloped by the Georgia Institute of Technology in Atlanta, it allows researchers in any field to upload data and explore ways to represent it with sound.
According to several researchers interviewed for this article, in addition to new tools, the way researchers and developers think about accessibility has evolved significantly over the past five years. In the United States, at least some federal funding bodies have prioritized accessibility. On December 30, NIH published recommendations (see go.nature.com/3gtmtk2) to break down affordances and create fair and inclusive research practices. (Swenor co-chaired the NIH’s Subgroup on Disabilities, which developed the proposal.)
But for vision-impaired researchers to truly succeed, tools must advance beyond fragmented efforts, says Hajas. “Until we make the whole ecosystem accessible — how to find the articles, how to read the articles, how to find the pictures, and everything else — there’s going to be a lack that prevents people from being good scientists,” he said.
We offer you some site tools and assistance to get the best result in daily life by taking advantage of simple experiences