Know any good legibility tests for typefaces?

[dberlowgone]

Nick: "So, in reading studies that focus on types with discrete letter forms, the outcome of letters-as-atoms would be expected."

Yes, and perhaps unexpected, when a discrete letter is studied. An example is the study of letters, including zeros, where a lone character, shorn via isolation, i.e. mis-atomized, is graded for legibility by the subjects. Their results, as I recall, showed that 9 of 10 subjects could not identify the zero of a popular font of the CT collection correctly. I recall the number of correct subject responses was high (1 in10), only because the subjects did not know they could guess that the zero could also be misread under such circumstances as a degree sign, a ring accent, or a dehydrated spittle mark, if it had the right stuff in it to begin with.

Eben: "None of us has an issue with this. Correct?"
I only have issues with the tests and the results Eben. Otherwise, I think James and his crew have done a remarkable job of designing & selling the concept.

"Throwing the baby of technology & science out with the bath water of industry is no solution."
Actually, it is the perfect solution, because it tests the resolve of those who wish to hoodwink you.

But now, the dreamer is coming down from the highlands, to have a 'blog war' on the topic.

Cheers!

[William Berkson]

Peter, Eben, I don't think you are understanding correctly the relationship between observation, scientific explanations of the observations, and scientific theories. I'm not optimistic that I can explain it to you briefly, so it will have to wait till we meet Rochester, if you are interested.

>“Throwing the baby of technology & science out with the bath water of industry is no solution.”
>Actually, it is the perfect solution, because it tests the resolve of those who wish to hoodwink you.

Well, my cousin has William Morris's old press, so I can hook you up with that. It is hand pulled, and that way you, like Morris, will avoid the evils of science: electric motor driven presses, and now the computer, and so on.

No doubt you'll do some lovely work, but we'll miss your new digital types :)

[ebensorkin]

Bill, I can wait until then. That's fine.

[enne_son]

Bill, I'm not sure to which post you’re referring when you say you don’t think I understand correctly the relationship between observation, scientific explanation of the observations, and scientific theories. I can wait til TypeCon, but for the sake of this thread I’d rather be given a little clue.

Perhaps you mean to say that by varying single parameters one at a time on the stimulus side in a bank of tests, and observing the result in terms of speed (for instance), I can do more than just tell which is faster. I can build up a picture of what affects perceptual processing in reading, and gradually compile a total picture of how they interact.

Of course I'd agree with that. It's implicite in the majority of my posts.

[William Berkson]

Peter, you wrote, "What is being resisted is an explanation of the form: 'x or y is actually objectively more readable because reading using x or y is faster or less tiring.'"

That's not an explanation, that's a definition, the meaning of readability. Highly readable *means* that people can read it with relatively good speed, comfort and comprehension. Less readable *means* people can read it with less comfort, speed and comprehension.

Speed and comprehension can be objectively measured, though comfort is more dicey as of yet. So what a scientific theory of readability should be looking for is theories that would explain why some texts are more readable than others. Part of the challenge of that development and testing is to objectively distinguish the more and less readable texts. So you are going to need objective measures that will reflect the fact that, eg. people prefer to read extended text in print rather than on screen. Of course there will be individual variation, and this needs to be taken into account in the measures--eg that over 45 are generally going to have more problems handling small type.

Then the theory is going to have to explain why those measures turn up the way they do. The theory may be about the reading process in general, and not specifically readability, but it will need to have logical consequences that predict the differences in the objective measures of readability.

In principle, so far as I can see there nothing scientifically objectionable in the testing I described. It would only need to be better specified to carry out the tests. The theory of reading would need to explain whatever the results are.

[enne_son]

Bill, perhaps what I should have said is “[w]hat is being resisted is an inferential judgement of the form: : ‘x or y is actually objectively more readible since reading using x or y is faster [more accurate], or less tiring.’”

Whether it is or not depends on the definition, and a definition is a social contract. What you propose is, it seems to me, a proto-scientific formalization which I think confuses what something is a function of with ways in which it manifests itself. Science needs formalizations to get its investigations going. Ole Lund in his Reading PhD dissertation on legibility research rasied issues of ‘construct validity’ and characterized the kind of formalization you propose as ‘easy operationalism.’

The definition I might propose defines readible text as text in which rapid automatic visual word-form resolution is not obstructed, but actively afforded. (This presupposes of course that something denominatable as ‘visual wordform resolution’ in fact exists.) I think that when type designers talk about readability, this way of tying it down for investigative cross-examination is more representative of what they're after and more cognitive-scientifically sound. That it is closer to the intentions of typographical practioners is reflected by the orientation to making optico-grammatically integral words rather than simply nice letters.

In any case science shouldn't be making the kinds of inferential judgements I described until it has a relatively accurate idea of what in actual functional terms is going on.

[Nick Shinn]

Highly readable *means* that people can read it with relatively good speed, comfort and comprehension.

That's not how I would describe readability.
I'd say it *means* people find the combination of content plus presentation interesting enough to start reading, and meaningful enough to continue. Limiting "meaningful" to the kind of comprehension that concerns deciphering words from letters and recognizing grammatical structures that look logical enough to not cause too many regressive saccades -- that's a pretty basic concept of meaning. Surely there's more to it than that?

[William Berkson]

Nick, the Oxford concise gives two definitions: "1. Able to read; legible. 2. interesting or pleasant to read."

The first definition clearly refers to the visual aspect of reading. The second has to do with the quality of the writing. Of course "there is more to it," as the reader's overall experience will involve both. But with a little cleverness we should be able to study the typographical dimension on its own.

Within typography a sophisticated but still pre-scientific definition of readability I quoted earlier: "the ease with which the eye can absorb the message and move along the line."

"Absorbing the message" clearly refers to comprehension, to understanding meaning. And 'ease' is some kind of comfort, or lack of resistance. Since we can decipher nearly illegible writing by slowing down, I think an idea that we can have some speed along with comprehenseion is implied or understood.

Yes, there's more to it, but the first, narrower typographic meaning is worthy of scientific study in my opinion.

Peter, I don't see that I am making much of any inference here about the underlying mechanisms of reading--of how we go from the printed page or screen to understanding. The only thing I pointed out is that two aspects have been operationalized: speed and comprehension. The third, comfort, has not been operationalized successfully yet. I suggested two possible measurable correlates of the subjective experience of discomfort or difficulty with the visual aspect of reading: reduction in comprehension with time and decline in proof-reading accuracy. One would have to test to see whether either of these is any good. I have proposed no short-cut or 'easy operationalization' here. It's difficult, but if achieved would be enlightening.

Then once you have such measures, you have the challenge of explaining their behavior in different circumstance. But at least you have something measurable, beyond the general self-report of comfort or discomfort. That I think would be a plus for further research.

As you know, I think that technical terms should be introduced only when necessary, and then with clear definitions. When you write your definition of readability as "readable text as text in which rapid automatic visual word-form resolution is not obstructed, but actively afforded," your technical terms smuggle in assumptions that may not be correct.

1. It may be that part of good readability, is having short words, such as
a, an, the, that have sufficient markers that the brain is able to skip and never read them. The meaning may be derived from a sub-set of words in many sentences.

2. Secondly, it might be that we can go directly from letters to meanings of whole phrases--the 'bouma' extends over multiple words. I don't think this is true, but your terminology assumes it rather than investigating it.

3. Finally when you nominalize and use the term 'affordance' you are creating something that seems to be a single variable that may turn out to have a number of independent processes going on, and the single term is not useful. I greatly respect your insights into the reading process, as you know, but I do think that introducing technical terms quickly is a serious obstacle to communication and may even muddy the waters conceptually .

You may object that 'readability' also implies a single variable, but that's not the case, because it is not being used as a technical term. As a general rule, terms that emerge out of peoples' experience represent something real, and that's why science would do well to respect them as a starting point, and only replace them with a very good reason.

[dberlowgone]

Bill:
1. It may be that part of good readability, is having short words, such as
a, an, the, that have sufficient markers that the brain is able to skip and never read them.

And, how 'bout dem Germans, they must have some hell of a time reading all those long words.
There are a variety of opinions, but none say we skip the and never readlittle ones. But you must think there is something between, skip and read-each-letter, then.

In any case, we've been through all this, it's like freshman readability orientation week — day one, (each and every day).

When is the actual Groundhog's Day?

Cheers!

[William Berkson]

>none say we skip the and never read little ones.

I can't cite the article--I bet Peter or Kevin can--but I believe that is exactly the currently most accepted view--we 'preview' in the parafovea, and often skip small words for the next saccade, never reading them, and fixate on later large ones.

>When is the actual Groundhog’s Day?

As you can see, above I tried to get out of this movie, but resistance was futile :)

>But you must think there is something between, skip and read-each-letter, then.

Yes, I started by arguing against Peter's rhetoric, and then but then I actually had to try to understand him. In the end, he converted me to the dark side--believing in the 'bouma'--and now I am stuck in this movie. Help!

[enne_son]

Bill, you are right in saying “terms that emerge out of peoples’ experience represent something real, and that’s why science would do well to respect them as a starting point.” For me the vital question is who is doing a better job: you in your seemingly innocent formalizations or me in my apparently unnecessary elaborations.

In response to Nick perhaps I'd say: maybe “efficient saccade planning” should be added to my definition. And I might need to add “and epiphenomenally enhanced” and change “actively” to “effectively.” so my definition reads “readible text is text in which rapid automatic visual word-form resolution and efficient saccade planning are not obstructed, but effectively afforded and epiphenomenally enhanced,” by which I mean helped out by more than strictly visual wordform resolutional virtues. Examples of directly visual word-form resolutional virtues are perfect spacing and distinctive cue values for criterial role architectural features. Examples of epiphenomenal virtues might be gestural atmospheric propriety, and clarity of presentational logic (at the very least).

Yes my definition makes assumptions. Don't yours?

Do you think visual word-form resolution is not real? Or should I have used “visual word-rocgnition instead?” Or is that less precise? Do you have a better sense of what word recognition is than you do of what visual word-form resolution is? Instead of “afforded,” should I have used “made possible?” Or is that kess succint?

Anyway, I apologize for the trouble I've caused. I am not a deliberately evil man.

[John Hudson]

And, how ’bout dem Germans, they must have some hell of a time reading all those long words.

And how 'bout dem Thais, they must have some hell of a time reading without separating the words at all.

[enne_son]

I hope readers of this thread will allow me another addition to this thread, just in case it is of use.

I could simplify my definition somewhat by saying “readible text is text in which holistic processing, efficient saccade planning and robust sense-following are not disrupted or obstructed by the particularities of type-design or the specifics of the typesetting, but effectively afforded and epiphenomenally enhanced.”

(Legible writing then, is writing in which the discrimination affordances necessary for accurate wholistic processing and sufficient for reliable analytic processing are not compromised by peculiarites in or degredations of letterform or the mis-management of letter-form-design parameters. The discrimination affordance will need to be perceptual, and of cue-providing role-architectural information)

Holistic processing in this complex has to be thought of in ‘bouma shape’ and ‘form resolutional’ terms, rather than ‘envelope structural’ and ‘lexical access‘ terms.

The reason to insist on more challenging definitions is because I believe this leads to the search for and the emergence of more discriminating tests of value. New tests to gauge for value might be in the realm of testing for the robustness of the word superiority effect, or in the realm of attentional effort, or in the realm of gauging sustainability of immersion with comprehension and delight. Perhaps Kevin‘s corrugator muscle and eye-fatigue research relates to the second one of these.

[William Berkson]

>Do you have a better sense of what word recognition is than you do of what visual word-form resolution is?

Yes, because 'word recognition' is English; that's my native tongue :)

What does the longer phrase 'visual word-form resolution' add? I don't know until you explain it, and that's why such phrases are a barrier to understanding. What I gather you think it adds is that it includes some of your theories about how we read. But the theories are not in the verbiage, which obscures rather than clarifies. If you first explain the your theory of word recognition, then when you write further I will know what you mean when you use the phrase 'word recognition'. And you don't have to pile anything on for me to understand each time you mention it. When you pile term on term, the whole thing becomes impenetrable.

For example, above you wrote:

“judgements of readability and legibility are possible because affordance varies complexly across many factors, and yes some typefaces compromise discrimination affordance of criterial role-architectural parts in certain conditions of use by reducing the formal grammer through a reduction of stroke contrast or the size and the distinctiveness of counters, for instance.”

I'm sorry, but that's just impenetrable without a lot of explanation, and when you have the explanation you can make the point or points more simply.

> I am not a deliberately evil man.

Ha! You have a pure heart, and good ideas on reading, I know. It's just that there's this demon called phenomenology that has possessed your keyboard. I'm trying to exorcise it :)

What is bad about this way of writing is that it obscures. At worst it gives a false impression of precision and depth that covers up a lack of ideas. At best--and that is in your case--it just makes a writer much hard to understand. Just as in graphic design, in writing prose lucidity is a fundamental virtue. It can be compromised on occasion for a reason, but the reason had better be a damn good one.

[enne_son]

Bill, I don't have rough and ready explanations for all the words I use. Some are of my own creation. Often they are cobbled together from terms I meet in things I read. They form because of things in my reading that resonate with my awareness of science and my knowledge of type and type-making. This resonance often has to do as much with the everyday ordinary meaning of these words, as it has to do with their assigned meaning in the specific contexts in which they appear

I don't choose them just to be obscure, but because I find them to be descriptive and precise. I guess I depend on the fact that the same resonance will occur in the minds of my readers. Perhaps I am foolharding in expecting that it might, and too undiscipline to not let them out until I can give a full account of what they mean.

Scattered throughout my posts on these subjects are various attempts at explanation of why and how I use my terms. I don't pretend what I write is easy to understand, and I try to be as lucid as I can without ressorting to formulations I'm uncomfortable with, or writing longer dissertations.

My only hope is some things will get through, and that the things that do won't be fool’s gold, but have legs.

[Nick Shinn]

the reason had better be a damn good one.

I'd say it is.
There's more in the cupboard than crystal goblets. Some of us like to tipple from ornate beer steins.

[William Berkson]

>There’s more in the cupboard than crystal goblets.

Yes, I should have qualified that to extended text and expository prose--of course clarity can still be compromised in these, but good reasons are more rare than in display type or fiction.

[ebensorkin]

When it comes to being curious about how to improve type I have keep being interested in that crystal and in Peter's fantastically garbed ideas. When it comes to use I will always be willing to choose a stein if it fits the tone I require. The instances where maximum clarity are crucial are pretty few & far between. These are not contradictory things to say at all.

[William Berkson]

>where maximum clarity are crucial are pretty few & far between

Eben, do you really mean to say that? Newspapers? Text books? Non-fiction? Lengthy fiction?

I would say rather that clarity and having different 'voices' with different aesthetics and moods are not in conflict. The goblet should be clear, but it can be beautifully shaped and decorated.

For a writer--an I speak as a writer first and a designer second--my interest is in having the visual clothing of my words be as inviting as possible to the reader and as easy as possible to understand. In other words, visual design that is both inviting and transparent.

[ebensorkin]

Where public safety and visually impaired folk and a variety of other related situations are concerned then yes I think that doing absolutely all you can is crucial. The greater the dangers the more thats the case. Highway signs fall into this category as well I think.

For general wayfinding on foot you have comparitively waaay more leeway. It's not a question then of safety so as much as wasted time and the sense that the makers do or don't care about you I suppose.

For texts: the longer the text and the longer you might reasonably need/want to keep reading the more important it will be. But Crucial? Maybe not. Too strong a word. If I am setting a novel I might legitimately choose something with a tad bit less readability in my eyes if the voice of the font was right. For magazines I would be willing to extend the lead further.

I am not saying the issue of Readability isn't important - far from it - I am saying there are degrees of importance.

The phrase maximum clarity are crucial is pretty strong wording isn't it?

By the way the word above is "garbed" as in robed or clothed. Not "garbled" as in unintelligible.

[Nick Shinn]

Bill, to return to the goblet/stein analogy (perhaps one should also include tea cups, coffee mugs and drinking straight from the bottle), and apply it to writing rather than type:

Isn't it reasonable to assume that complex ideas may best be expressed by complex grammar?
And precise meanings by long words?
And new ideas by new words?

Surely the alternative is dumbing down.

[ebensorkin]

Isn’t it reasonable

Quite right. And this is why specialist language comes about in all areas. We need new words so we make them. Or borrow them. Or we recycle them.

It is true that to understand Peter you have ask him what he mans by these things. But he is very nice about answering questions.

[William Berkson]

My view about good writing is similar to what Einstein said about a good scientific theory: it should be as simple as possible, but no simpler.

I think one should always be reluctant to introduce new terminology, though there are times when it is useful and desirable. Complex grammar is is not a requirement expressing for deep or complex ideas, and is best avoided as a disservice to the reader. Bertrand Russell, for example, wrote about very difficult and complex philosophical issues with admirably clear prose.

Lucidity is a wonderful achievement, whether in graphic design or in writing. It is by no means the only virtue in these fields, but it is one of the highest in both.

[ebensorkin]

I think you you should cut him some slack Bill. He is getting the ideas together. Not presenting them as fait accompli.

Besides, a lovely lucidly explained mistaken theory is no more accurate for it's being lucid.

[enne_son]

Eben, I'm okay with Bill’s remarks about the way I present my ideas. I try to formulate compact stand-alone and all-inclusive one-sentence definitions, but recognize they're often not good prose or easily navigatable. And than I don’t generally give readers what they need to unpack them. That’s not right, even though I think readers should be able to find their way into my ideas by exploring the ordinary-language and dictionary usage of my terms.

I make myself and my ideas vulnerable to readers throwing up their hands, thinking sooner or later readers will come around to my thickish brew.

In a personal not to Bill I wrote: “just so you know, 'visual word-form resolution' is my name for the visual or perceptual processing component in word recognition. Word recognition also has a cognitive processing component which has to do with connecting with meaning. Word recognition is also used in connection with contacting items in a 'mental lexicon' — basically a database of spellings. Contacting items in the mental lexicon is a grey area in theories of word recognition. It isn't really thought of in perceptual processing or cognitive processing terms. It's regarded as a post-perceptual, neural-computational thing. This use is very widespread. I feel the need to distantiate what I am talking about from it.”