How We Read

hrant's picture

I'll be adding subsequent pages to make this
topic less confusing :-/ so feedback would be
particularly appreciated, like in terms of what
you think needs to be elaborated on the most:


hrant's picture

Thanks for the compliment.
I make my pilcrows with extra space on the right, so you don't have to put a space after, which can look too much anyway (especially on-screen) and can cause disasterous linebreaks too (again, especially on-screen).


hatz's picture

armenians are koooool...

i like your way of thinking... i have some questions though...

doesn't the area of the fovea depend on the distance the eye has from the text?..
and if so, then isn't the importance of the boumas (in the parafoveal area) comparatively small (because of having more letters in the fovea)?.. or not?..

another thing is that, to me, readability depends also on the reading habits with which somebody has grown up...
as i can speak for myself, growing up in germany with those clean letterforms and after twenty years getting in touch with my greek roots, greek letterforms appeared disturbing to my eyes...
although greek letterforms have more ascenders and descenders, german (well,.. latin) still appears more legible to me...

i don't know, if i'm now out of topic, but at least it was a comment...

anyway, i'm a fan of yours...


hrant's picture

> doesn't the area of the fovea depend on the distance the eye

Yes, but notice that our arms are a certain length... That's why in normal conditions (optimal type size, reading position, etc.) the fovea spans about three letters, and accounts for only about 1/3-rd of reading. When you have a small book, you hold it closer, and the type can/should be smaller. And when you have a large book, you lay it down somewhere, and the type can/should be larger.

> readability depends also on the reading habits

Yes, "familiarity" is the second important component of reading. But it's even less understood than our reading "hardware"... For example, is it gained in years, or minutes? The answer would be extremely relevant, but we just don't know yet.

> latin) still appears more legible to me...

*Legible*, yes, and *inherently* so. A writing system with more regularity (e.g. less extenders) appears more legible, because our consciousness is foveal, and everything is clear. A writing system that's less regular (e.g. more extenders) is actually mroe *readable*, because things are less ambiguous in the parafovea. This is also why sans serif forms appear more legible than serif forms, but a book in sans is no fun to read.

For example, Armenian is more readable than Latin, but less legible; the former is better for books, the latter better for posters. This is because of the difference in vertical information.

Look at the far-right of:


hatz's picture

thanx for specifying those things...

but i have to mention that while i'm reading a text, my eyes are moving... which means that the fovea isn't static... if i stop reading and try to focus on a word, again i have the impression that my eyes are in motion... maybe i need new glasses... the thing i want to say is that, to me, the foveal and the parafoveal areas are overlapping, because a text is meant to be read... and i think those informations we get while reading, reach unconsciously our brain which is still a terra incognita by itself...

> This is also why sans serif forms appear more legible than serif forms, but a book in sans is no fun to read. <

i guess it also depends on the purpose anybody's chosing between serifs and sans... if you want to read a book, surely serifs help you to keep a fluent flow in your reading... but if you want to read a manual or something more instructional, then i think sans are more suitable... so it's not only about the fun of reading, but also of understanding the content of what you're reading...

but again: thank you very much for your research... it's really helpful to have those things on my mind...

question: are you teaching anywhere?..

hrant's picture

> while i'm reading a text, my eyes are moving

It's an illusion. Reading proceeds in jumps and stops - I was shocked too when I first found out! This was discovered only about 100 years ago.

On the other hand, there is something called "micromovement": an extremely small vibration of the eyes (smaller than the smallest feature on a letterform) which seems to keep vision "awake" - when micromovements are suppressed, everything goes gray... Micromovement is below the scope of reading, but I guess it's possible that you can see "feel" it when you fixate for a long time?

> understanding the content

I agree that the style of a typeface helps create context for the content. But especially during immersive reading, that's limited. For any text more than a few paragraphs, sans fonts will cause fatigue. You can always break the rules of readability, but you have to know them first.

Teaching? I'm too busy learning!


hrant's picture

Jay, I have a friend in CIA BlackOps who'd love to dissec... uuuh, *analyze* your vision system.

Thanks for the kind words.


hrant's picture

I've also heard that minute amounts of the active ingredient in bubonic plague have been used to loosen the inside muscles just a tad... And you thought Lasik was scary.


hrant's picture

> everything to the right of my nose is peripheral vision.

I think it might not be *that* bad, since the eyes are pretty close*, relative to the span of vision. I suspect your left eye can see about half what your right eye can see anyway. The issue of depth is more significant, but maybe not for reading?

* Unless you're a rat. Did you know that rats have a blind area some distance in front of their noses? Very valuable information... if you're an insect. :-/

Have you found that you tire quickly (compared to somebody otherwise similar to you) when reading a long book?

> I'd guess that the fovea would be smaller side-to-side

I think the fovea would be about the same size, but the parafovea maybe 75% the size.

From what I remember, there have been perception tests done on people with monocular (or highly lateralized) vision, but I don't remember any offhand.

What I also wonder is whether there's a *cognitive* issue here, since each half of the brain does different types of tasks, and each eye shares half of its retina between them. There's a book called "The Alphabet and the Brain" which tries to show that reading rightward is inherently more efficient than leftward, irrespective of the writing system (which is related to this issue). I wasn't entirely convinced, but the foundation for the book (studies of lateralization of perception/cognition) is solid.


hrant's picture

> How did that rule come about?

More than a rule, it's a guideline. But actually, my statement was an over-simplification (probably my functionalist side having too much say - again...), and it depends on the kind of text.

There is the raw text, then there's the "atmosphere" that the font conveys, and they have to be balanced. For example, in a coffee-table book about Paul Klee, it's sad, but nobody's going to read all that fodder, so it might make more sense to use a sans no matter how much text there is. But if you want to reduce obstacles to the absorbtion of textual meaning, serifs help. The reason is simple: they help meld letters into words.

> The right eye sees the word 1st

Actually, the left halves of each retina see it first, and they both go to one side of the brain - I forget which... :-/ But there doesn't seem to be enough evidence to suggest that this creates an advantage for a certain direction. Do women read faster than men? Their brains are less lateralized.


hrant's picture

There is definitely the risk of "discrimination". The trick is to say "serifs help readability" instead of "serif fonts are more readable". A font like Syntax is certainly more readable than Lubalin Graph.

The reason serifs help in readability (although they actually hurt *legibility*, since they confuse the individual letter-structures*) isn't because they guide the eye (through horizontality or anything else), it's simply because they help individual glyphs blend into boumas.

* This difference in legibility and readability is central to a good understanding of type design, and it parellels the difference between contemplation and immersion, between display and text type, between consciousness and the subconscious, etc. There is certainly a lot of overlap between readability and legibility, but there is opposition too: you could say they're at a 90 degree angle.

Also, the use of type is indeed more important, not least because there are more bad type users than type designers! :-) No matter how readable a font is, bad setting can kill it. However, if one assumes good basic setting, the font's readability does become a factor. And anyway, anything else is outside the scope of the type *designer* to begin with...

Readability studies... what can one say? Most are flawed, and even the good ones can't be taken as gospel. They end with a concluding paragraph (because that's the way formal science works these days - you gotta earn your grant, baby), and those never really make enough sense. Ignore the "concluding" stuff, look *inside*, and don't expect too much. Why is this the case? Because reading is friggin' *complicated as hell*! :-) But honestly, if you spend enough time looking through a large number of individual studies, seeing if they're useful, exactly what they shed light on, etc. and you couple that to overall cognitive studies (like those of Herman Bouma), then you can start to see the model converging in your head, you start seeing the dark as a half-light. You start seeing that serifs help (and how they help), you start seeing that divergence helps (much more than the mainstream type designer will allow his aesthetic sensibilities to admit), and a bunch of other things.

But hey, don't take my word for it, really! :-) See what Javal, Ovink, Patterson & Tinker, Bouma, Taylor, Spencer and others have found, and check out some lo-fi designs by Mandel, Tracy, Carter and other true master craftsmen (as opposed to "artiste"s), and chances are you can gain your own insight and teach us all some new things! It's esoteric, arduous stuff, but to me it's really a blast.


anonymous's picture

I like it so far. In fact, your site is extremely informative as a whole.

By the way, I like putting space before *and* after my pilcrows. It seems strange when it crowds the following capital. Maybe that's just me.


anonymous's picture

Hrant --

I had a problem reading your page. It wasn't your page's fault, it was just the fact that I couldn't read it w/o thinking about the process of reading, and that's when the trouble started.

My right eye stopped to check out a particularly attractive bouma, but the left eye kept going. The left eye called the right lazy; Right called Left dominating. Now they are cross(ed) with each other.

Seriously, I enjoy reading your stuff, and have learned a lot from it. I'm also impressed w/ your left-to-right web-page design -- cool!

My only critisizms are so minor as to be not worth repeating.


anonymous's picture

It's been done.

I was born cross-eyed; back in the dark ages it was considered a good idea to go in w/ a scalpel & take a whack or 2 at the inner muscle that controls eye movement. That elongages the muscle, & corrects the problem.

The only issue is that it also creates a bunch of scar tissue, which creates its own problems.

Now-adays, they skip the surgery & do the same thing w/ stretchy exercises.


anonymous's picture

Wow. I hadn't heard that. Yeah, scary -- kinda like Botox for the eyes? Shiver.

Anyway, back on subject: any idea how being monocular affects the fovea? My left eye is very dominant, the right eye is basically along for the ride. The best way to describe it is that everything to the right of my nose is peripheral vision.

I'd guess that the fovea would be smaller side-to-side, necessitating more saccades?


anonymous's picture

> For any text more than a few paragraphs, sans fonts will cause fatigue. You can always break the rules of readability, but you have to know them first.

I know that is a widely followed rule although personally I have never had any problems reading sans serif text. How did that rule come about?

anonymous's picture

>Have you found that you tire quickly (compared to somebody otherwise similar to you) when reading a long book?

I don't think so, but how can one tell? I have noticed that I tire of John Grisham sooner than most. [grin] I spent most of my childhood reading, so it didn't affect me much. I do seem to read faster than most, but that could just be enthusiasm.

>which tries to show that reading rightward is inherently more efficient than leftward...

Wow. There's a concept. The right eye sees the word 1st, categorizes it, prepares the left to understand it? Is bouma a left-brain thing?


anonymous's picture

>>Have you found that you tire quickly (compared to somebody otherwise similar to you) when reading a long book?

>I don't think so, but how can one tell?

I suspect it takes extreme vision problems to make reading tiresome on the eyes. One of the most avid readers I know is legally blind and has to hold the book quite close to his face. Now, he's not the fastest reader - maybe because his fovea has to make more jumps per line. But I'm amazed he reads at all, when in fact he must be one of's best customers.


anonymous's picture

On readibility

I am especially curious about the readibility of text when it comes to choose a sans serif or a serif face for the bodytext. It seems to me that to some degree, it is a matter of the art director's taste, excluding the obvious examples where anyone can find it hard to read a text regardless of the serif condition of the typeface. Do you guys know (especially Hrant) if there's a scientific study on readibilty, one that can actually provide sources of information and methodology of the process?

I am somewhat opposed to the occasional discrimination toward sans serif for the running text of a publication, thinking that a study of the particular typeface involved in the design is more judicious than applying a general rule that might keep someone from conveying a message in a more direct way just because the general feeling is that sans serif will cause fatigue.

In a book by Rob Carter (Working with computer type, Rotovision SA, Switzerland ISBN: 0823064794) he says: "Due to the horizontal movement created by serif typefaces, it was believed for some time that serif letterforms were more readable than sans serifs; however, readability studies have proved that there's little difference between them. Tracking for example, has a lot more importance than that" (words might vary a little: translated form a non-english text). Unfortunately he doesn't elaborate on this point to fill us in on what his sources were.

Other than that I think that Hrant's research is interesting and I will be very curious to see more of it. Thanks for making it available to everyone, Hrant.


Syndicate content Syndicate content