New Adobe research examines user-centric designs for diverse digital reading needs

Today, Adobe researchers and engineers are presenting two papers at ACM CHI’2023 in Hamburg, Germany targeting flexible reading formats and tools for readers with dyslexia and without.

“I’m not really a reader” or “I don’t like to read” are some of the sentiments we’ve heard frequently from family, friends, and research study participants. Some of these readers have been formally diagnosed with dyslexia or attention deficit disorder (ADD), or attention deficit hyperactivity disorder (ADHD), while some have self-reported dyslexia or reading challenges. Others may just consider themselves slower or less proficient readers. What’s in common to all these readers? They could all benefit from alternative reading formats and custom interfaces, rather than reading document text in the inflexible, inaccessible, one-size-fits-all format it usually comes in. In fact, whether young or old, struggling or proficient, dyslexic or not, our research shows that all readers can benefit from a more tailored reading experience.

Last year we reported on our research showing that different fonts work best for different readers, increasing reading speeds without impacting comprehension. The challenge lies in finding the best reading format for every reader. In our most recent research, we involve the reader directly in the customization process. While they might not be able to pick the fastest format to read in, readers will surely gravitate towards the formats that appeal to them more, to ideally make the reading experience more comfortable and enjoyable.

Our research is built on user-centered and inclusive design principles, meaning that we’ve included users with different reading abilities and with different demographics at the outset when designing our reading experiences and tools. At this year’s main international conference on Human-Computer Interaction, ACM CHI’2023, we presented design recommendations from two readability studies.

Reading themes improve the reading experience for readers with different needs

Digital reading affords us with the flexibility to move away from one-size-fits-all to formats that are better suited for different readers’ needs. In “THERIF: A Pipeline for Generating Themes for Readability with Iterative Feedback”, we used multiple iterations of crowdsourcing, machine learning, and design sessions to arrive at 3 reading themes — Compact, Open, and Relaxed that each offer a different combination of font and spacing (inter-character, word, and line).

Lead author and Adobe research engineer Tianyuan Cai was pleasantly surprised to find that the format with the largest spacing ended up with Poppins, a sans serif font: “it’s great that this fell directly out of our data, because both large spacing and sans serif fonts are associated with better reading experiences for readers with dyslexia, so our themes naturally bundle features that work well together.”

Findings:

• We designed and validated our Compact, Open, and Relaxed themes by running studies with over 800 participants in total, roughly half of which had a high likelihood of having dyslexia (recruited based on a dyslexia questionnaire we administered).
• The three reading themes have progressively increased character, word, and line spacing. The font also varies between the themes, with serif (Georgia and Merriweather) fonts selected for Compact and Open, and a sans serif font (Poppins) selected for the Relaxed theme.
• Participant demographics were related to format preferences. Participants over 55 preferred the themes with the larger spacings, and a significant proportion of participants with dyslexia preferred the Relaxed theme. A larger proportion of readers without dyslexia preferred the Open theme.

Digital reading rulers can help readers with dyslexia and without

Physical reading rulers (sometimes referred to as dyslexia rulers) are a common intervention to help readers with dyslexia keep their place on a line of text using transparent colored plastic or even just the top of a regular ruler or piece of metal. Aleena Niklaus, first author of the paper “Digital Reading Rulers: Evaluating Inclusively Designed Rulers for Readers With and Without Dyslexia”, was surprised that digital versions of reading rulers were not more widespread, being limited to paid accessibility software and not widely known browser extensions. Moreover, while many digital reading ruler designs are possible, no guidelines were available about which designs are most effective for readers with dyslexia, and whether readers without dyslexia could also benefit. Aleena’s investigations into reading rulers started as a winning entry in an internal Hackathon, before spinning out the work into larger focus groups and crowdsourced user studies of different ruler designs, while evaluating user preference, reading speed and comprehension.

Findings:

• Over the course of 2 focus groups with a total of 10 participants with and without dyslexia, we converged on 4 reading ruler designs: Grey Bar, Shade, Underline, and Lightbox.
• In reading tests with 177 participants (51 percent with dyslexia), Grey Bar, Shade, and Underline rulers led to speed increases of 10-20 WPM on average when compared to using no ruler.
• Participants with dyslexia benefitted the most from reading rulers. They experienced the largest speed gains with Grey Bar, on average, but tended to prefer Lightbox as a design.
• Although readers without dyslexia did not read faster when using a reading ruler, many preferred having a reading ruler to read with.

Because dyslexia is so prevalent in the general population, we actively recruit participants with varying degrees of dyslexia for our studies, to build the reading tools that can help reduce the gaps between struggling and proficient readers.

Looking ahead

Based on these research projects, we found that digital reading rulers can significantly improve, not disrupt, reading experiences, not just for readers formally diagnosed with dyslexia, but for all readers. As a result, we believe that it’s important to make custom reading tools readily available to all readers, not bury them in accessibility menus.

In our future work, we look to continue to customize reading experiences to readers with different needs, including finding formats that work best for different situations, reading materials, and tasks.