The case for an ugly font

There is a certain pleasure in a study that flatters your worst instincts. In 2011 a paper arrived that seemed to bless every cluttered, badly-typeset handout you'd ever squinted at: the harder something is to read, it suggested, the better it lodges in memory. Not harder in the sense of denser or more complicated — harder in the shallowest possible way. The same words, in an uglier font. Set your notes in a cramped, ungainly typeface and, the finding went, more of them would stay.

It's the kind of idea that spreads fast, because it's cheap, counterintuitive and a little bit rebellious. It also turned out to be a near-perfect case study in how a real effect can be real, and celebrated, and then mostly wrong — and in what separates a difficulty that helps you learn from one that just makes your eyes tired.

The delightful finding

The paper was Fortune favors the bold (and the italicized), by Connor Diemand-Yauman, Daniel Oppenheimer and Erikka Vaughan, published in Cognition. Their word for the effect was disfluency — the small, subjective sense of difficulty you feel when a mental operation doesn't go smoothly. A blurry word, a faint photocopy, an awkward font: each produces a flicker of friction, and the researchers wondered whether that friction might nudge the mind to slow down and process more deeply.

In the lab, it did. People given information to study in a hard-to-read font remembered more of it later than people given the same information in clean, comfortable type. That alone would have been a curiosity. What made the paper famous was the second study, which walked the idea out of the laboratory and into a real school. Across 222 students at a public high school in Chesterland, Ohio, teachers reformatted their actual classroom materials — worksheets, handouts, quizzes — into deliberately awkward fonts: Haettenschweiler, Monotype Corsiva, italicised Comic Sans. The comparison classes kept whatever the teacher normally used, usually Times New Roman or Arial. Then everyone sat their ordinary tests.

The students who'd studied from the ugly fonts did better. Not on a contrived memory task — on the regular assessments of their real courses, across subjects and grade levels. A cosmetic change, costing nothing, appeared to lift learning. The press wrote it up gleefully, and the theory behind it had a name ready to receive it.

A desirable difficulty

That name was desirable difficulties, coined by the memory researcher Robert Bjork. It is one of the most important and most misread ideas in the science of learning, and the disfluent-font result seemed to be its neatest illustration. The core claim is that conditions which make learning feel harder and slower in the moment often make it stronger and more durable in the long run — while conditions that make studying feel easy and fluent tend to produce learning that evaporates.

It is the through-line of almost everything we've written here. Testing yourself feels harder than re-reading, and teaches far more. Guessing a word before you're told the answer feels like a waste, and plants it deeper. Spacing your reviews out so each one is a small struggle beats massing them while the material is fresh. In every case the difficulty is the point: it is the sign that your mind is doing the work that lays down memory rather than gliding over the surface.

So a hard-to-read font fit the story beautifully. Here was a difficulty you could add with a dropdown menu. If struggle helps, and an ugly font is a struggle, then the ugly font should help. The logic was clean. The only problem was that it didn't hold.

Difficulty is not a single dial you can turn up for a bonus. It matters enormously which difficulty you add — and whether it makes you do more, or just makes the page worse.

The honest catch

The trouble with a delightful finding is that everyone wants to build on it, and building is how you find out whether the foundation is real. The disfluency effect got its grandest test in the form of an actual product. In 2018, a team at RMIT University in Melbourne released Sans Forgetica: a font engineered, with the help of psychologists, to sit in the sweet spot of difficulty — back-slanted, broken up with little gaps, just hard enough to read that your brain would supposedly have to lean in and, in leaning in, remember more. It was downloaded widely, fitted to browser extensions, and covered everywhere as the font that could make you smarter.

Then people checked. In 2020, Andrea Taylor, Kimberley Wade and colleagues published four experiments in the journal Memory under a blunt title: Disfluent difficulties are not desirable difficulties. Participants read material in Sans Forgetica or in plain Arial, and were tested on it. They reliably rated Sans Forgetica as harder to read — so the disfluency was real, the friction was doing its part. But the memory benefit simply wasn't there. Sans Forgetica produced recall no better than Arial, and in one experiment, on cue–target word pairs, it produced recall that was measurably worse. A font built expressly to help you remember was, if anything, getting in the way.

And this wasn't a lone contrarian result. Years earlier, in 2013, Xiao Yue, Alan Castel and Robert Bjork himself had published a careful study in Memory & Cognition with a telling title of its own: When disfluency is — and is not — a desirable difficulty. Blurring words, they found, changed how people felt about their learning: it lowered their confidence, made them judge that they'd remember less. But it did not reliably change how much they actually remembered. The friction registered as a feeling and stopped there. Bjork, the architect of desirable difficulties, was among the first to mark the boundary of his own idea.

Why the difference matters

So why does testing yourself work when an ugly font doesn't? Both feel harder. Both are, in the plainest sense, difficulties. The distinction is the whole lesson, and it's this: a difficulty helps only when it changes what your mind actually does with the material — not merely when it makes the input less pleasant to receive.

When you retrieve a word from memory instead of re-reading it, you are performing a genuinely different mental act: you are reconstructing, effortfully, a thing that wasn't in front of you. That act strengthens the very pathway you'll need later. When you guess a word's meaning before you're told, you generate a prediction, and the gap between your guess and the truth is what the answer fills. These difficulties are productive — they conscript you into deeper processing.

A hard-to-read font asks for none of that. It makes the letters harder to resolve, and once you've resolved them — a fraction of a second later — you're reading exactly the same words in exactly the same way. The difficulty lands at the door and never gets inside. It taxes your eyes without ever recruiting the machinery of memory. The Chesterland classrooms may well have captured something real; effects can be true in one setting and vanish in the next, and a novel font in a whole school can carry a shot of attention and novelty that fades the second everything is set that way. But as a reliable lever you can pull, disfluency has not survived contact with replication. The friction you can feel is not the friction that teaches.

The special case of a foreign word

For vocabulary, there's an extra reason to be wary of dressing words in difficult type — and it's specific to the strange task of learning a foreign one. When you study a word in your own language, the ugly font only obscures a form you already own; the meaning is safe underneath. But a new foreign word is its form. Its unfamiliar spelling, the exact sequence of letters, the shape you'll have to recognise on the page and reproduce yourself — that fragile visual form is precisely the thing you are trying to learn. Blur it, break it, back-slant it, and you risk muddying the one impression you most need to be clean.

This is where disfluency parts ways with a difficulty that genuinely helps handwriting. When we wrote about forming a word by hand, the benefit came not from the writing being hard to read but from the act of producing every stroke yourself — an act of generation, of doing more with the word. Effortful production helps. Effortful perception, on the current evidence, mostly doesn't. Keep the word itself crisp and legible; put your difficulty somewhere it will do some good.

What to actually do

The point of all this isn't that difficulty is a myth — it's that the useful kind is more specific than "make it annoying." A few ways to aim your effort where it pays:

• Add difficulty that makes you produce the word, not just read it. Cover the answer and reach for it from memory. That single move — retrieval — is the difficulty with the strongest evidence behind it, and it's the opposite of passively re-reading a hard-to-read card.
• Keep the word itself clean and legible. A new foreign word's spelling is the thing you're learning; don't obscure it. Save your effort for what you do with the word, not for how hard it is to make out.
• Let the gap between reviews be your friction. Spacing your encounters so each one is a small struggle is a genuine, well-replicated difficulty — and it costs nothing but patience.
• Guess before you check. Predict a word's meaning, then reveal it. The productive difficulty is in the reaching, and the answer that follows sticks better for it.
• Be suspicious of study hacks that only change how things look. A trick that makes the page different but leaves your mental steps unchanged is, at best, doing nothing. If a tip doesn't change what your mind has to do, it probably isn't changing what you learn.

Where vocabla fits

The disfluency story is, in the end, a story about where to spend a difficulty you can only spend once. You can waste it on the surface — a cramped font, a blurred line, a friction your eyes feel and your memory forgets — or you can put it exactly where the reaching happens: at the moment a word is asked of you and you have to produce it yourself.

That is the difficulty vocabla is built around. It doesn't dress your words up to look hard; it keeps each one crisp and readable, then hands it back to you at the point where recalling it takes real effort — inside a real sentence, just as it's starting to slip, so the struggle is the productive kind. The right difficulty, in the right place, is the whole of the desirable-difficulties idea once you strip away the fonts. Keep the word legible. Make the remembering hard.