Dyslexic Adults Understood Parts of a Story Less Well on an E-Reader

When we talk about accommodations on the show, e-books come up a lot. They are one of the standard tools for dyslexic students, along with having text read aloud, and for good reason. You can make the font bigger, change the spacing, look up a word, or have the device read to you.

So we were surprised by this 2019 study from a group of French researchers. They took a plain e-reader with no extra tools and asked whether dyslexic adults understand a story differently when they read it on a Kindle instead of on paper. For the dyslexic readers, the answer was yes, in a specific way. For everyone else, it made no difference.

The paper sits behind a paywall in Annals of Dyslexia, and we are grateful to the authors for sending us a copy so we could write this up.

What they did

The researchers recruited 60 university students at Aix-Marseille University and the University of Strasbourg. Thirty had dyslexia, all diagnosed in primary school, with an average of about five years of remedial reading help behind them. The other thirty were typical readers. The two groups were matched on age, year of study, vocabulary and nonverbal reasoning, and each group had the same mix of women and men and the same split between humanities and science majors.

Testing confirmed that the dyslexic students still had the classic signs as adults. They read nonsense words more slowly and less accurately, had a harder time pulling sounds out of words, and could hold fewer syllables in short-term memory.

Each student came in twice, at least a week apart. One visit they read a short story on a Kindle Paperwhite. The other visit they read a different short story in a small paperback the researchers made to match the Kindle's screen exactly: same page size, same font, same 14-point type, same words on each page. They even took the page numbers out of both versions, so nobody could use them to remember where things happened.

The stories were around 2,700 and 2,950 words, about 13 or 14 pages each. Which story went with which device, and which came first, was balanced across the students. None of them had used an e-reader before.

Students read silently in a comfortable chair, front to back, with no flipping ahead or back. Then they answered questions out loud, with no time limit. The questions came in four types:

• Literal: facts stated directly in the story, like who did what.
• Inferential: things you could work out but that were never said outright.
• "Where in the text": for 18 events, was it in the first, middle or last third of the story?
• "When in the story": 20 events printed on cards, to be put in order.

The last two are the unusual ones. They are measuring whether you have a sense of the shape of the story, where things sit and how they connect, rather than whether you remember a particular fact.

Reading speed: slower, but not because of the Kindle

The dyslexic students took longer to read, about 20 minutes compared to about 15 for the typical readers. That works out to roughly 140 words a minute versus 187. The device made no difference to speed for either group. So the comprehension differences below are not because anyone spent more or less time with the story on the Kindle than on paper.

In print, the dyslexic students held their own

This is the part of the paper we keep coming back to.

When they read on paper, the dyslexic students did about as well as the typical readers on the literal questions. Their average was a little lower, but the gap was not statistically significant. On the inferential questions there was no gap at all, on either device.

On the "where in the text" task, the dyslexic students who read on paper did better than the typical readers, and that difference was significant. On putting the plot back in order, the two groups were about even.

The authors point out that the conditions here were different from a lot of earlier research that found dyslexic adults struggling with comprehension. The texts were long enough to give real context. The questions were spoken, so students did not have to decode the questions on top of the story. The questions were open-ended rather than multiple choice, which the authors say can overload working memory. And there was no clock. Under those conditions, slow decoding did not stop these students from understanding the story.

On the Kindle, the picture flipped

On the e-reader, the dyslexic students scored lower than the typical readers on two kinds of questions.

Literal questions. On the Kindle, the typical readers clearly outscored the dyslexic readers, a large and statistically solid gap. One detail matters here, though. The dyslexic students' own literal scores were somewhat lower on the Kindle than on paper, but that drop on its own was not statistically significant. What changed is that a borderline gap in print became a clear one on the e-reader.

Keeping track of the story. This is where the effect was strongest. On "where in the text," the dyslexic students outscored the typical readers on paper and scored below them on the Kindle. Their own scores dropped sharply between the two devices. On putting the plot in order, they were again much better on paper than on the Kindle, and on the Kindle they scored below the typical readers.

The typical readers, meanwhile, scored about the same on paper and on the Kindle on every single measure.

Why would a Kindle matter?

The authors' explanation is about something they call paratextual cues, the information you get from a book that is not in the words themselves.

When you read a paper book, you can feel your progress. The pages you have read pile up in your left hand and the pages you have left get thinner in your right. You turn each page with a physical movement. Without thinking about it, you build up a sense that the dog showed up early, the argument happened somewhere in the middle, and the ending was close when the left side got heavy. In a story told in order, knowing where something was in the book is a big help in knowing when it happened and what it caused.

An e-reader feels the same on page one as on page fourteen. Every page turn is the same swipe in the same spot.

The idea is that dyslexic readers, who have spent years working around slow and effortful decoding, lean harder on cues like these to hold the whole story together. Take the cues away and that part of comprehension suffers. Typical readers, who are not working as hard at the word level, may not need them as much, at least for a story this short.

That last caveat comes from the authors too. An earlier study by some of the same researchers found that even typical adults were worse at reconstructing the plot of a much longer story, around 10,800 words, when they read it on a Kindle. The stories in this study were about a quarter of that length. It is possible typical readers would show the same effect with a full novel, and the dyslexic readers simply hit the limit sooner.

The authors are clear that this explanation is a hypothesis. The study shows that the device mattered for the dyslexic readers. It did not measure page weight or hand movement directly, so it cannot prove that is the reason.

What to keep in mind

This is a careful study, but it is a small one, and there are real limits on what it can tell us.

• Sixty students, all at university, all in France. These are dyslexic readers who have done well enough to get into higher education. Kids still learning to read, or adults who did not get that far, might respond differently.
• Short stories, not books. About 2,800 words is a long text for a lab study but short compared to anything a student reads for class. The effect could be bigger or smaller with a full novel or a textbook.
• Nobody had used a Kindle before. Some of what the researchers saw could be the unfamiliarity of a new device. Students who read on one every day might adapt.
• A bare e-reader. No text-to-speech, no adjustable spacing, no highlighting. It is a test of a screen versus paper, not of everything digital reading can offer.
• It runs against another well-known result. A 2013 study by Schneps and colleagues, which the authors cite, found that reading on a small screen with very short lines helped some dyslexic students, improving both speed and comprehension. The two studies tested quite different setups, so they do not cancel each other out, but they are a reminder that "e-reader" covers a lot of ground.
• Some results are borderline. The literal comprehension finding rests on a gap that was just short of significant on paper and clearly significant on the Kindle. The strongest results by far are the ones about placing events and reconstructing the plot.
• It is one study, from 2019. A single small experiment is a reason to pay attention, not a settled answer. If you know of a follow-up with dyslexic readers, please send it our way.

The bottom line

Dyslexic university students who read a story on paper understood it as well as their classmates, and were better at remembering where things happened. When they read the same kind of story on a Kindle, that advantage went away and they scored lower than their classmates. Readers without dyslexia were unaffected either way.

The likely reason is that a paper book gives you a physical sense of where you are in a story, and dyslexic readers seem to rely on that more.

None of this means you should take away an e-reader that is working for your kid. It does mean that for long reading, where the shape of the story matters, a paper copy is worth keeping in the mix. And it adds to a growing pile of evidence that dyslexic readers, given time and a fair way to show what they know, understand a lot more than their reading speed suggests.