Famous designers vs font readability research

I have been working as a designer for about 10 years. There are not many objective and verifiable laws in the design, and when I needed to find out something, I looked for information in professional books, cool guys' blogs, asked art directors from my friends, and molested people in the corridors.

And then I found out that scientists have already found out a hundred years ago about what designers still argue about.

We at Tinkoff compared the opinions of famous designers and scientists about which fonts are easier to read: antique (with serifs) or grotesques (without serifs).




Why am I doing this at all


Once I needed to write instructions for the jones on how to choose a font. I wrote that for large texts it is better to use serif fonts, because the eyes are less tired of such a font and serifs help to keep the line.
Then the editor came and commented: “Proofs?”

“But this is obvious,” I thought, and opened Wikipedia :
According to the generally accepted opinion [source not specified 1902 days], serifs direct eye movement along the lines when reading large arrays of printed text. They facilitate the connection of letters in a single line, facilitating the visual perception and readability of the text.

There were no proofs on Wikipedia, and I went to the bookcase.



In my opinion, a serif font is less tiring to read during regular reading of regular “paper” editions than grotesque, for two reasons. Firstly, serifs emphasize the end of strokes, becoming additional "meaningful identifiers". Secondly, serif letters are somewhat more complex in form, therefore, they are more different from each other than grotesque. And our reader’s eye needs a balance of individuality and unification more than a design eye, which rejoices in mirror-like refinement.

Yuri Gordon, “A Book about the Letters from Aa to Yaya,” Where Do You Come from and Why Need Serifs, p. 51

In general, the personal view of Yuri Gordon is a pretty weighty argument, but still. I searched in the articles of Jan Chichold:
A sans-serif font just seems the simplest. Its shape was specially simplified for children, and for adults it is more difficult to read than antiqua, because its serifs serve not only for decoration.

Jan Chichold, “The Shape of the Book,” On Typography, p. 21


Jan Chichold

Well, the old masters relied on a flair, but maybe someone will explain it in more detail with the proofs?
Sergey Surganov, art director of Medusa, designer of Notion:

(, , , ), , . , , , . , - , .

- «» « »

In short, we have three arguments:
“according to the generally accepted opinion”,
“in my opinion”,
“it is considered”
...

Books did not help, and I tried to figure out with the help of scientific articles whether there is any evidence that the serif font is easier, faster or nicer to read than sans-serif font. Of course, this is not a complete scientific review. But better than "according to conventional wisdom." Links to articles are at the end of each section.

Concepts: legibility vs readability


Legibility of a font depends on the accuracy of its elements, which usually means the ability to recognize individual letters or words.
Readability is directly related to the optimal layout and layout of the entire body of the text.

Serifs guide the line


Wikipedia and another million design articles say serifs direct eye movement along lines when reading large amounts of printed text.
Here you need to know that the eyes do not move smoothly. When we read or look for some object in the field of view, the eyes “jump” from point to point, making quick movements called saccades.


Video about saccades.

Moreover, the eyes perform saccades, even when a person tries to focus strictly on one point.
Jared Skryuz (Jaret Screws) conducted an experiment: asked 10 people to read several texts and monitored the movement of their eyes with the help of special equipment.
Here's what happened:



We measured the duration of fixing the gaze, the average number of words between them, the amplitude of saccades and the number of times a person lingered on the same word.



All differences were insignificant. Oh.

It seems that the hypothesis that the gaze moves more easily through the text typed in serif fonts has not been confirmed.


Serif and sans-serif fonts equally “hold” the line.

Although there is a catch. The test used the font 128 pt. This is ten times more than in a regular book, and perhaps the study is not very correctly applied to the normal reading process.

Quantitative Analysis of Font Type's Effect on Reading Comprehension Jaret Screws Clemson University Clemson, United States

Children, unlike adults, find it easier to read a sans-serif font


Even if this is not the case, alphabet manufacturers certainly believe in it. Try to find at least one alphabet with antiqua.


Children's ABCs from the first page of Google

Scientists took 80 children 10 years old and 80 children two years older, asked them to find the right word (screening test) in the text and measured how quickly the children managed to do it.


The control group was given the same font both times, and the serif font at first, and then without the

difference they turned out to be statistically insignificant, and the authors say that there seems to be no particular difference in legibility for serif fonts and sans-serif fonts.



Performance differences between Times and Helvetica in a reading task Rudi W. De Lange, Henry L. Esterhuizen and Derek Beatty

What about the visually impaired?


There are a lot of studies about visually impaired people and type.
A review of 18 studies was published in the Journal of Visual Impairment and Blindness, in which the total number of subjects was more than 1,500.
Researchers conclude: For people with low vision, sans-serif fonts, such as Arial, Helvetica, Verdana, or Adsans, are more readable than serif fonts.



The Legibility of Typefaces for Readers with Low Vision: A Research Review by Elizabeth Russell-Minda, Jeffrey W. Jutai, J. Graham Strong, Kent A. Campbell, Deborah Gold, Lisa Pretty, and Lesley Wilmot.

Fonts and Dyslexics


It is difficult for people with dyslexia to read. Maybe they can make their life easier with some fonts?


A page that displays text as seen by people with dyslexia on dyslexiarf.com

Scientists examined 97 people using technology that monitors eye movements, half of which had dyslexia.
They were allowed to read 12 similar texts in different fonts, measured the speed, number and duration of eye stops, understanding the text and using questionnaires to find out the preferences of users.



In the first place in terms of readability, both ordinary people and dyslexics came out with the Arial sans-serif font (it is interesting that Helvetica, which is similar to it until indistinguishable, is in fourth place). But in the second place unexpectedly was a monospaced squared Courier.


The preferences of people are generally unambiguous: the top 3 fonts for dyslexics and non-dyslexics are sans-serif.



The Effect of Font Type on Screen Readability by People with Dyslexia
LUZ RELLO and RICARDO BAEZA-YATES

Emotions and Fonts


Maybe it's not a matter of read speed and legibility, but a special mood created by serif fonts?
The Software Usability Research Lab at Wichita State University tried to figure out how people perceive fonts.
Participants filled out a questionnaire, which indicated what, in their opinion, the characteristics of different fonts.



Interestingly, serif fonts turned out to be the first in the ranking as formal, “adult”, practical, and stable. And sans-serif fonts were the first in the ranking ... nowhere. That is, they did not have any pronounced emotional trace.

Perception of Fonts: Perceived Personality Traits and Uses By A. Dawn Shaikh, Barbara S. Chaparro, & Doug Fox

For another study, they took two satirical excerpts from the New York Times: one on government issues, the other on education policy. They were printed in the Times New Roman and Arial fonts of the same size and randomly shown to 102 university students who rated them using predetermined adjectives.
Satirical articles published by Times New Roman were perceived as more ridiculous and meaner than those published by Arial.



Emotional and Persuasive perception of fonts Samuel Juni, Julie S. Gross

Most likely, this is the same effect of neutrality of sans-serif fonts as in the first study.

Does this mean that all sans-serif fonts are faceless? Generally not, but a serif neutral font is less neutral than a sans serif neutral font.

Readability of fonts on electronic screens


Most of the texts we read from electronic screens. Electronic devices and even different programs on the same device handle fonts differently.


Different types of smoothing

Scientists have tried to figure out how smoothing affects readability.
We compared four fonts printed on paper on the screen with anti-aliasing and without anti-aliasing.



It is interesting that the most readable was the screen font with anti-aliasing (and not printed), and this is a sans-serif font - Arial. The second after him is Verdana, also sans-serif.
But without smoothing, Georgia was the best read - serif font.



A STUDY OF THE READABILITY OF ON-SCREEN TEXT By Eric Michael Weisenmiller

findings


I wanted to check if serif fonts are better for large texts. It turned out that for an ordinary person there are almost no differences, and for those who have reading problems - the elderly, children, people with poor eyesight or dyslexia - a sans-serif font is preferable.

The article has reviews of just a few articles. I looked at a couple of dozen, but did not include in the article, so as not to make it endless. In fact, there are even more. But the result in all is about the same: either the differences turn out to be statistically insignificant, or the grotesques slightly outrun the antiqua.



If you are interested in this topic, you can do your own research: search on scholar.google.com for a query such as Font readability research. There you can find studies about the optimal line spacing, and about the perception of different font styles, and about differences in readability depending on size.

It is unlikely that this will help to become a fundamentally better designer, but will add weighty arguments.

All Articles