Dyslexie font - activism

John Savard

Apart from Dyslexie being commercial, there is another issue. People like myself, who don't know much about dyslexia, but who had heard of it from news articles, tend to associate it with trouble with reversed letters. So a font that differentiates between p and q, and b and d, seems like it would potentially be of some use. But more than one person struggling with dyslexia has written here that this is completly untrue, that it isn't even the case that some, but not all, dyslexics would be helped by that.

To get rid of that symmetry, one wouldn't need to use a special typeface; just use Goudy Friar with the right choice of alternates.

ibrahim kactioglu

According to new researches open dyslexic performs poorly.
You can read it here: https://cognitiontoday.com/2018/05/font-psychology-research-and-application/

Thomas Phinney

RaseOne said:

I think any serious attempt, done in good faith & based on solid theory is a step in the right direction.

Dyslexie and friends are NOT, in fact, based on “solid theory.” They are based on a folk theory about dyslexia that is not shared by the cognitive psychologists who have studied dyslexia from a scientific perspective. Most research contradicts that theory. As one article says, “The problem in dyslexia is a linguistic one, not a visual one” (https://www.medicalnewstoday.com/articles/186787.php, third sentence). 

On the other hand, there may be many learning/processing disabilities lumped together as “dyslexia.” This also suggests that at best, dyslexic fonts which are based on a belief that dyslexia involves seeing letters flipped and/or rotated, would be effective for just one subgroup out of a larger number.

But even were that true, if the font were good for some noticeable subgroup, it ought to provide a measurable benefit.

RaseOne said:

Some of the studies linked above don't really discount the efficacy of OpenDyslexic but more say it is limited or inconclusive. One in fact says it does do some good .... Another says that these type of modifications to glyphs can do some good

“Limited” in what way? You need to be more specific about WHICH studies you are claiming found these effects, and what exactly they said. Last time I checked, I found zero evidence supporting any statistically meaningful benefit from Dyslexie (or Open Dyslexic). But perhaps there has been more research. I would have thought any such positive results would have gotten a lot more discussion, but it is possible.

The study long cited by Boer (Leeuw, 2010) found no significant difference in reading speed between Arial and Dyslexie. The study mentioned "some positive and negative effects" on accuracy, but neither the positive nor negative effects reached statistical significance, nor were they large. The Master's thesis had a certain amount of spin unwarranted by the results, and has been spun further by Boer.

Note that it behooves us to be wary when so many studies have found no effect. If the cutoff for statistical significance is ≤ 5%, and people keep on running more experiments, testing a bunch of things each time, eventually they will get a result that was p ≤ 5% likely to happen by chance. In such circumstances, a result could may be more reflective of the number of things tested and number of experiments run, than the existence of an actual experimental effect. Of course, if the effect found is massively less likely to happen by chance, that would be helpful.

It would also help if studies were done comparing more fonts and using legible fonts. Honestly, if you compare Arial with something else for dyslexic reading... well, that is not a terribly high bar. I would very much like to see some humanist sans serif typefaces that have decent letterform differentiation. Lucida Sans? Open Sans? Lato? Fira Sans? Something like that. I would bet that either (1) no significant differences are found between any of the reasonable fonts tested, or (2) a reasonably-differentiated humanist sans serif outperforms both Arial and Dyslexie (or Open Dyslexic).

Oh, also super important to control for differences in size. Making a font bigger on the em square is a kind of cheating. Might as well just set the font bigger, then.

RaseOne

I can't really disagree with much of what you said there Thomas. After all if it were so simple to solve then it would probably already be obvious that certain types of fonts like a Garamond for example helped measurably but unless I misread one of the studies linked in this thread said sans-serif fonts tested better. That seems counterintuitive.

With that said... these are the first fonts made specifically to TRY to address this issue & I do think they are based on solid theory. A lot of things based on solid theory fail & there would be no way to know that in this case if no one ever built the fonts. I'm no more fond of unsupported claims than anyone else so we agree there. I just thought the almost universally very harsh criticism was a little heavy.

My son deals with an equally if not more misunderstood issue which also exists on a spectrum that overlaps other conditions & overlaps difficulties that people with no diagnosed condition at all face to some degree. Less than half of supposed treatments & countermeasures make any difference or any sense for that matter so my attitude towards these fonts is that they are a step in a potentially productive direction & people can make their own choices.

The concept of "sight reading" (words as units) probably supersedes the effectiveness but I'll stick with the idea that systemized orientation data is a sensible factor to add to the equation. If it helps a small subgroup or helps to differentiate one subgroup from another then that's a good thing.

Maybe such fonts would be better thought of as potential diagnostic tools or research tools rather than potential solutions.

Thomas Phinney

“I do think they are based on solid theory.”

Solid in what sense? The experts say otherwise. There is no evidence for it to date. So that’s why I am so harsh about it.

I would love for type design to be able to help people in new ways, and it would be super cool and fun if the things type design could do for dyslexia were different from those needed for legibility in general. So I would be *thrilled* to support this stuff if there was any evidence for it—or even if it had a solid theoretical basis. But I have seen none so far.

Nothing wrong with doing more testing, especially if it is done intelligently. But at this point I do not expect the sum total of testing to show any benefits for custom dyslexia fonts that try to “weight” the letters, when compared to equivalent “non-weighted” fonts.

And when those weighted fonts are based on Arial and Helvetica style neo-grotesque sans serifs, I won’t be surprised if they could be outperformed by humanist sans serifs that don’t even make any special “weighting” accommodations.

John Savard

Here you are! This page includes several fonts for dyslexics, including Gill Dyslexic. So if you want a humanist sans-serif as the basis... Myself, I find that if I want asymmetry for the "p"s and "q"s, I have to go to typefaces like Goudy Friar, and just going from Helvetica to Gill is not going far enough.

Here are the alternates I'd choose for the letters in Goudy Friar:

Essentially, the principle is to choose the alternate more like the conventional Roman in each case, rather than the one more authentically like the uncial, except where the choice of alternates is made to accent the contrast between b and d.