Drafted with AI assistance, edited by Yahya Hafez.
Loading post…
Monthly notes on comprehension, accessibility, and AI tutoring. No spam, no upsells, unsubscribe anytime.
Dyslexia fonts have a compelling pitch. Weight the bottom of each letter so it cannot flip upside down. Give b, d, p, and q different shapes so they cannot be mirrored. Open up the letterforms so they stop blurring into each other. OpenDyslexic is free, built into e-readers and browser extensions, and recommended in countless classroom guides. Dyslexie, its commercial cousin, is sold to schools across Europe. Both are backed by sincere testimonials. Neither is backed by the controlled studies, and the reason why tells you what actually helps a dyslexic reader on the page.
OpenDyslexic was released in 2011 by Abelardo Gonzalez, and Dyslexie was designed in 2008 by the Dutch designer Christian Boer, who has dyslexia himself. Both are built on the same theory: that dyslexic readers see letters rotate, flip, and swap, so letters with heavier bases and distinct silhouettes should stay put.
That theory is the problem. As three decades of brain research show, dyslexia is not a visual disorder. Its core is phonological, a difficulty mapping printed letters onto the sounds of spoken language. Letter reversals are common in all beginning readers and are not what distinguishes dyslexic ones. A font engineered to stop letters flipping is solving a problem most dyslexic readers do not have. Whether it helps anyway is still an empirical question, and since 2013 researchers have put it to the test repeatedly.
The first rigorous comparison came from Luz Rello and Ricardo Baeza-Yates in 2013. They recruited 48 people with clinically diagnosed dyslexia, aged 11 to 50, and had each read twelve comparable texts, one per font, while an eye tracker measured reading time and how long their eyes paused on each fixation (Rello & Baeza-Yates, 2013 (opens in new tab)). Font type did matter. Courier, Helvetica, Arial, and Verdana produced the shortest fixations. Italics hurt, with Arial Italic the worst font in the set on both measures. OpenDyslexic, the only font designed for these readers, was no easier to read than the ordinary sans serifs. A second finding is worth remembering for later: which font participants said they preferred had almost no relationship to which font they read fastest.
The adult result might not transfer to the children these fonts are mostly marketed for, so two later studies looked there. Jessica Wery and Jennifer Diliberto ran a single-subject alternating treatment design with elementary students diagnosed with dyslexia, comparing OpenDyslexic against Arial and Times New Roman on letter naming, word reading, and nonsense-word reading (Wery & Diliberto, 2017 (opens in new tab)). No student read faster or more accurately in OpenDyslexic, and the group as a whole did not either. Some children said the font looked new or different. None said they preferred reading in it.
The largest test targeted Dyslexie. Sanne Kuster and colleagues first had 170 Dutch children with dyslexia read in Dyslexie and in Arial, then had 102 children with dyslexia and 45 without read word lists in Dyslexie, Arial, and Times New Roman (Kuster et al., 2018 (opens in new tab)). In neither experiment did Dyslexie produce faster or more accurate reading, and the children preferred the standard fonts. The authors' conclusion was measured: the font neither benefits nor impedes reading in children with or without dyslexia. There is no harm in using it. There is also no reason to spend money on it instead of on phonics instruction that is known to work.
One study did find that Dyslexie beat Arial, and it is the most instructive of the lot. Eva Marinus and colleagues at Macquarie University had 39 low-progress readers read four texts in four font conditions, all matched on letter size (Marinus et al., 2016 (opens in new tab)). In plain Dyslexie the children read about 7 percent more words per minute than in plain Arial. Then the researchers widened Arial's letter spacing and word spacing to match Dyslexie's, and the advantage disappeared. The special letter shapes contributed nothing. Dyslexie's spacing settings, which are generous by default, did all the work.
That lines up with what the spacing research had already found. In 2012, Marco Zorzi and colleagues showed that extra-wide letter spacing roughly doubled reading accuracy and sped reading by more than 20 percent in 74 dyslexic children, with no effect on typical readers (Zorzi et al., 2012 (opens in new tab)). The likely mechanism is crowding: neighbouring letters interfere with recognition more strongly in dyslexic readers, and space is the antidote. A 2018 study in Bosnia reached the same verdict from the other direction. Dyslexie did no better than Times New Roman once the latter was given extra spacing, and extra spacing erased the penalty of deliberately awkward fonts like Times New Roman Italic and Curlz (Duranovic et al., 2018 (opens in new tab)).
There is one caveat. Jessica Galliussi and colleagues tested 64 children with dyslexia and 64 without on letterform, letter spacing, and word spacing separately (Galliussi et al., 2020 (opens in new tab)). Dyslexia-friendly letter shapes again did nothing. Wider letter spacing on its own, without a matching increase in word spacing, actually slowed reading, because once letters spread out the gaps between words stop standing out. Spacing works, but it has to be tuned as a pair.
Size matters too. In a 92-person eye-tracking study that varied size, spacing, colours, and column width, larger text and larger character spacing made both dyslexic and non-dyslexic participants read significantly faster (Rello & Baeza-Yates, 2017 (opens in new tab)). The earlier Rello work settled on 18 point as a sensible floor for web text.
The research points to a short, cheap checklist. None of it requires a special font.
The letter shapes never mattered. The space around them does, and so does the size. Any text you read on a screen can give you both, whichever font it happens to be set in.
