Trace Center Directs New Digital Accessibility Research During Covid-19 Pandemic

Hayleigh Moore -

The Trace Research & Development Center at the UMD iSchool leads new initiatives to address accessibility barriers heightened due to the COVID-19 pandemic.

A smartphone laid diagonally on a dimly lit wooden table with a light shining down on it from the top right corner and "COVID-19" displayed on its screen.
Photo by Martin Sanchez on Unsplash

The COVID-19 pandemic has thrown into stark relief the critical importance of digital accessibility and the particularly high stakes for people with disabilities. With stay-at-home orders, extended periods of quarantine and many working remotely to avoid infection, any lack of access to computers, websites, apps, and other aspects of our digital world amounts to disconnection from the tools we all rely on for daily living — from grocery delivery to applying for unemployment benefits and from education to banking. For people with disabilities or older adults, both the risk of infection and its impact are higher, making remote access even more important. 

When the pandemic began last year, the focus of the Trace Research and Development Center at the University of Maryland College of Information Studies (UMD iSchool) on the technical, legal, and institutional sides of digital accessibility lent itself to exploring how to address these issues and develop innovations and techniques in accessibility in a time of global crisis. 

Dr. Jonathan Lazar, a professor in the UMD iSchool and a principal investigator at Trace, is an internationally known expert on digital accessibility for people with disabilities. In a project already underway when the pandemic struck, Lazar was conducting a series of interviews with university directors of digital accessibility. These interview sessions gave Lazar a front-row seat  to witness how these institutions adjusted their management of digital accessibility as they navigated the abrupt shift to fully remote environments. 

In a Google TechTalk, Managing Digital Accessibility at Universities During the COVID-19 Pandemic, Professor Lazar explained his findings, including the the problem of universities quickly procuring new technologies and software, bypassing long-established procurement controls which included accessibility checks conducted by each university’s team of digital accessibility experts. Without these checks, the software and technologies acquired by faculty, staff, and students to use during the pandemic for tasks such as online testing, chatting virtually with university admissions, and viewing and tracking critical COVID-19 updates, were inaccessible to many people with disabilities. A more detailed report on these findings is available at: https://link.springer.com/article/10.1007/s10209-021-00792-5

The shift to new ways of teaching and learning implemented in response to the pandemic forced many students to cope with a new learning environment affected by their technological capacity, housing conditions, or loss of social connections. Dr. Hernisa Kacorri, Assistant Professor at the UMD iSchool and a principal investigator at the Trace Center, invited the students and teaching assistants in her class to co-author the paper, Reflections on Remote Learning and Teaching of Inclusive Design in HCI. They discuss how faculty, teaching assistants, and students involved with Inclusive Design in HCI, a graduate course taught at the iSchool in Fall 2020, coped with online learning. The research revealed that while the virtual environment fostered a space to develop and practice more inclusive interactions, such as describing oneself before continuing to speak and adding alt text to sketches and other visual artifacts, timing was critical.  Faculty reported spending 30% more time than developing the class from scratch. Students found class to be too long due to “Zoom fatigue,” but also too short to fit interactions.

Another key barrier highlighted by the pandemic was the number of people who have great difficulty or are unable to use a home computer. Just when they became so essential to us all for both communication and just about everything else, it became apparent that many who needed them were not able to use computers. The Morphic project, led by the Trace Center investigators Drs. Gregg Vanderheiden and J. Bern Jordan, was developed to provide easier access to built-in and 3rd party accessibility tools, and also allows the creation of ultra-simple ‘1-click’ user interfaces for those who otherwise are unable to use a computer. The software, called Morphic, has recently been released by the non-profit Raising the Floor. 

Work being conducted by Dr. Amanda Lazar, Assistant Professor at the UMD iSchool and a principal investigator at the Trace Center, has explored new methods for testing designs with people with dementia, a topic even more critical during a crisis like a global pandemic. There have long been additional barriers to in-person research, such as difficulty recruiting participants and accounting for transportation needs and accessible physical facilities. PhD students Rachel Wood (supervised by Dr. Jonathan Lazar) and Emma Dixon (supervised by Dr. Amanda Lazar) led a team of PhD students that worked throughout 2020 to study how to effectively run remote usability testing sessions involving people with dementia. Their work has been accepted for publication in ACM Transactions on Accessible Computing. 

In addition, in a recent extended abstract describing work led by PhD student Hanuma Teja Maddali (supervised by Dr. Amanda Lazar), Supporting Remote Participation when Designing with People with Dementia, a new system was proposed to compensate for these obstacles and improve prior research methods. The proposal “builds on top of commercially available video-conferencing systems to leverage video supported non-verbal communications such as reading face and body language.” 

Similarly, PhD students Kyungjun Lee and Jonggi Hong together with their advisor Dr. Hernisa Kacorri, explored ways that they can overcome obstacles when moving studies from the lab to blind participants’ homes, where experimenters don’t have access to blind users’ interactions as they can’t sit close to them or guide them as it is typically done in the lab. They demonstrated how integrating Zoom calls with smart glasses worn by blind participants can help facilitate remote interactions necessary for evaluation studies and how camera field of view and sound source can affect what is being captured. 

Another exploration with smart glasses came in the form of a collaboration between Kacorri and Lee, and researchers from Carnegie Mellon University, New York University, and IBM Research to understand the potential and limitations in current wearable cameras for making social distancing more accessible for blind people. The spatial behavior of passersby can be critical to blind individuals to initiate interactions and preserve personal space, so the team built GIAccess which leverages computer vision built into wearables cameras to help blind people access proxemics, such as presence, distance, position, and head pose, of pedestrians. Feedback from participants informed the team that the wearable smart glasses were easy to use since they didn’t have to worry about camera aiming and led to several design implications for future assistive cameras for a number of use cases.  The team’s paper on this work, Accessing Passerby Proxemic Signals through a Head-Worn Camera, has been accepted at ASSETS 2021 and will be presented in October.

In the lead-up to the November 2020 election, as the pandemic raged on, concerns grew about access to the vote for people with disabilities. Though most states offered no-excuse absentee/mail-in voting during the pandemic, the absentee/mail-in voting option was not accessible to many  people with certain disabilities (e.g., voters who are blind or low-vision and use screen-reading software were required to use inaccessible PDF applications and ballots). These voters were told that they could either use the accessible voting machine in-person in their local polling place, or have a relative or friend fill out their absentee ballot.   

As Dr. Jonathan Lazar noted, “If you don’t provide people with print-related disabilities a private ballot in the safety of their own homes during the pandemic, but you are giving that access to people without disabilities, again, you are not providing an equal experience, and in fact are asking people to take unneeded risks with their health.” 

Dr. Lazar worked as an expert witness in two legal cases related to absentee ballot accessibility. In addition, during the 2021 legislative session of the Maryland General Assembly, he gave testimony four times to legislative committees  considering bills related to digital accessibility.  

Furthermore, based on his ongoing research work with Adobe to continue to improve production workflows to create accessible PDF documents, he spearheaded a collaboration with Adobe to provide free support to states making absentee balloting in the PDF format accessible for voters with print-related disabilities. As a result, Adobe offered free software to election officials, free consulting on solving technical problems, as well as training to state officials. 

Finally, work led by iSchool professor and Trace affiliate faculty member, Dr. Niklas Elmqvist, and involving Dr. Jonathan Lazar and PhD students Pramod Chundury and Biswaksen Patnaik, focused on providing more accessible equivalents for data visualizations such as those used to help the public understand the spread of COVID-19.

To keep up with the latest Trace Center activity, follow them on Twitter @trace_center or visit trace.umd.edu

Trace Research & Development Center iSchool News Accessibility and Inclusive Design Computational Linguistics, Machine Learning, and Information Retrieval Data Privacy and Sociotechnical Cybersecurity Human-Computer Interaction Information Justice, Human Rights, and Technology Ethics