Tuesday, October 27, 2009

Tip - writing for maximum reader comprehension

Today I read a post entitled "Serif vs. sans-serif legibility" on a blog which claims to be a "Blog about web design & development". (I have removed the link to the blog as a favour to the blog author after his comment below, since this post is critical of the material posted).

The post had been referred to in Hacker News - one of the feed combinators that I subscribe to in my Google Reader.
The post caught my eye because it was on a topic that interested me and on which there is quite a large and solid body of scientific research and knowledge, and I wondered if the post added anything to that body of knowledge. Disappointingly, it did not, because it was a superficial, elementary and unscientific look at the subject by someone who clearly seemed to be ignorant of that body of knowledge and of either human psychology or especially of human visual perception - the latter two factors being what I would have hoped might have been burned-in to the skull of anyone involved in "web design & development". Maybe modern educational standards just don't cut the mustard.


The post (and subsequent comments) was rather a mass of opinion and half-baked ideas from like-minded people. The author described himself as being located in Italy and a "... web designer, developer, teacher, web addicted, SEO specialist ...".
Though he was seemingly relatively ignorant on the subject matter, I suspect he is an intelligent author whose first language - judging from the sentence structure - is probably not English.

The post opened with:
"Probably one of everlasting discussion is legibility issue of serif and sans-serif fonts."
- which would seem to be (see below) an incorrect statement.

It finished with:
"What do you think?"

Well, if you were to go and read that post [sorry, link removed, per above], you would probably see why I would suggest that it doesn't really matter what we may "think", because it's been a matter of fact rather than a matter of opinion for many years now - but we might not all know that.

The fact is that there is - as I put it above - a large and solid body of scientific research and knowledge on this entire subject, starting with Operations Research in the UK during WW2 and then post-war research, (e.g., by the UK Automobile Association) much of which was repeated subsequently, and all of which has conclusively shown that serifed fonts improved reading comprehension by a significant percentage over non-serifed fonts. By "significant", I mean (from memory) typically a 30% to 40% (or greater) improvement in reading comprehension.

That body of research/knowledge covers the domains of communication and human visual perception. It's not so much that "Our eye will try to close the missing part of the letter..." (as the blog author had put it) as that our brains seem to be wired up in such a way as to:
   (a) try to make sense of whatever information the eyes give them, and
   (b) extrapolate using that information.

It was shown conclusively that serifs provide a lot more useful information for such extrapolation than non-serifs, which means that the brain gets more information in total from printed characters using serifed fonts. There are several reports summarising this research and its findings, in the public domain. You are welcome to download these two (click on the link to start a download):
It was just a lucky coincidence for us that the early printers standardised on the use of Roman serifs (e.g., Times Roman today), and an unlucky coincidence that they also standardised on black print on white paper - which has subsequently been found to be one of the most contrasting combinations they could have used and which can cause visual perceptual disorganisation and even eye-strain. If we invented writing today, we would probably standardise on something which did not cause visual perceptual disorganisation or eye-strain such as - for example - black on grey (I'm writing this in my editor, using that colour scheme), or black on pale green. People who have seen those old computer monitor screens with glowing green writing on a black background might now perhaps understand why they seemed so easy to read - your brain "liked" them. For a really useful take on this, refer Turn Google Docs into a Distraction Free Writing Tool, where the author provides a simple and nifty note-taking template for use in Google docs.

Sadly, on many (most?) web sites, the ubiquitous and de facto standard use of non-serifed fonts (e.g., the artistically lovely Helvetica and other "pretty" fonts) and meaningless clashing clours, graphic objects or "eye-candy" has become all the rage. These serve merely to minimise reading comprehension and even detract and distract from comprehension. Thus comprehension takes a nosedive and visual perceptual disorganisation is on the increase. It's all just bad ergonomics really.

When we are responsible for laying out a web page or a blog, and before we publish it, we need to consider the above factors. For example, do we want to help the viewer to achieve maximum reading comprehension of some important information that we desire to or are obliged to communicate, or do we not really care just so long as the viewer gets a vague warm and fuzzy feeling about our web page, regardless of the eyestrain?

Recent research on web-browsing habits shows that the user can be quite ruthless if they don't get something that they find useful/interesting from a web page in the first 3 or 4 seconds. If they get something useful, then they will stay and read on for a while - (say) maybe 20 seconds or more - because their attention has been captured and they are willing to give their cognitive surplus to the page. If not, then they will immediately skip that page and go to another page, and likely as not will never return to the page skipped. This could represent a lost opportunity for the web page publisher. You can see this principle - the power of the attention-grabbing effect - in operation, if you use a feed aggregator or subscription reader. For example, I use Google Reader. One of the subscriptions I have is to a blog called "Hot Air". That blog usually has very readable articles in it, but for some reason the publishers do not present punchy and infomative header lines for my Google Reader to display. For example, sometimes all they have is a "Read it all...", which I find an immediate turn-off.

I scan those header lines - they are each maybe about 20 words long, at most. If I do not see anything that grabs my interest in that header line, then I skip it. If you need to scan or get across thousands of items in a day, and do not have the time or patience to follow up bad attention-grabbers, you will never know how good or bad the skipped post was and Hot Air will never know how much they could potentially increase their readership by the simple expedient of writing informative, punchy and attention-grabbing header lines.

Now, assuming that what I have written above is largely valid, then why on earth does Google blogger.com have sans serif default fonts, and why does Google Reader present all of its text to the reader in a sans serifed font? Like I said above:
"Maybe modern educational standards just don't cut the mustard."
AddThis Social Bookmark Button

1 comment:

  1. Thank you for your long comment. I don't get it why you are taking this matter so personally? The point of one blog is probably to share your ideas with rest of the world. I do exactly that share my ideas. But your right I speck four languages but English is not my mother language. Yes I'm not graduate in psychology,I'm graduate in chemical engineering. Yes I did not conduct any field research. Everything is my personal point of view.

    But the question is what do you not agree with me?
    First PDF says:"People are able to read serif more quickly."

    That is exactly what I'm saying. Serif are more readable. My personal explanation that serif provide addition information that help close the character.In the example with C you can see that serif font stops and can't be confused with anything else. But the C of sans-serif can be confused because C doesn't stop and the logical line can continue with forming of the letter O.

    The second PDF: " Instead of the physical
    characteristics of the font influencing the readability of the text, an
    alternative explanation is that the results may have occurred because
    people are more familiar with serif fonts."

    In my second part of my post I'm saying:
    "Other interesting study( theory) was that we are fast reading Times New Roman because Times New Roman is everywhere specially in the newspapers and books.
    In other words we are use to read Times New Roman and with time we get better and faster."
    Witch is exactly the same conclusion like the second PDF.

    ReplyDelete