What do your format choices mean to readers?
Hyunjin Song and Norbert Schwarz summarize the implications of some recent psychological research on format this way:
Any variable that facilitates or impairs fluent information processing can profoundly affect people’s judgements and decisions. [Writers] are therefore well advised to present information in a form that facilitates easy processing: if it’s easy to read, it seems easy to do, pretty, good, and true.
Let’s explore this using a memo with a simple plan for reducing energy costs in an office as an example.
Simplicity of typeface determines perceived simplicity of content
The actual content — every single word — and its arrangement is the same in Memo A and Memo B.
But we can predict readers of Memo B will believe the content is difficult to understand or the recommendation is difficult to implement because of the typefaces chosen.
- Memo A uses MS Word’s old defaults: Arial for headings and Times New Roman for body text.
- Memo B uses Forte for headings and Mistral for body text.
If readers considering the proposed recommendation could choose between reading Memo A or Memo B, Memo A would win out nearly every time. That’s because typefaces influence how fluently information can be processed. We equate easy to read with easy to understand or do.
Familiarity of typeface and words determine perceived risk within content
Another problem with using the typefaces in Memo B has to do with their lack of familiarity. It turns out that we associate risk with the unfamiliar.
Here’s another example of this principle. The content of Memo A and Memo C is identical except Memo A uses the word “reduce,” while Memo C replaces that word with “curtail.” Nothing else is different between the two memos.
Yet we can predict that readers of Memo C will believe the recommendation is more risky than readers of Memo A because of the use of the less familiar “curtail.” In this case, we’re talking about a difference in style rather than format. The point is that we equate easy to read –whether because of familiar typeface or word choice — with safety.
Legibility determines perceived truthfulness of content
Memo D is exactly the same as Memo A except that it is less legible because of the low color contrast between the black text on the blue background. We can predict that readers of Memo D will believe less in the truth of its content than readers of Memo A. We equate easy to read with true.
In sum, your format choices influence what readers think of your content. Is it difficult or easy to use and understand? Is what you’re talking about risky or safe? Is it the truth? All excellent reasons to learn a few principles for choosing a successful format for your documents. My video tutorial (see the link below) covers the basics of using typefaces, page layout, and lists.
Thanks to Cheryl Stephens for making me aware of this article in The Psychologist. The original is worth a read to learn more about the research behind successful formatting choices.
Related articles
- The video tutorial on format (proswrite.com)
These all sound like old results. Are these old results, or are they really new research?
I’d be a lot more interested if someone did some research on whether #1 and #3 are the result of #2. I don’t find #1 makes any sense in a historical or cross-cultural context, so I suspect #2 might be the cause.
Ditto for #3. For example, embossed or sandblasted type are more difficult to read and have the “low contrast” quality mentioned in your post. Do readers really think fancy-looking embossed or relieved type really make readers think the contents are not truthful? Do readers really think a letter-sized sheet with your door number posted on your window really looks more truthful than lower-contrast permanent signage? If so a lot of people will be very interested.
The research cited is pretty recent: 12 of 18 studies were published since 2006. Those that are older are used to show the history of studies in an area. The Psychologist exists to make research findings meaningful and available to practitioners.
I think I understand your point that familiarity might influence perceptions of simplicity. However, if a document uses all capital letters in Times New Roman, the research predicts the lack of simplicity (fluency in cognitive processing) will lead readers to judge the content of the document as hard to understand. I don’t think you can argue that there is a familiarity effect in that case.
I don’t understand what you mean about historical and cross-cultural contexts.
I encourage you to follow the link to Song & Schwarz’s original summary as my post was purposefully reductive.
I think cross-cultural is clear enough. Imagine a culture where you must write in a formal register or will not be seen as credible. (Japan, I believe, is such a culture.) If someone from that culture moves to North America, will that person still automatically associate unfamiliar words with untruthfulness? Will that person ONLY make such an association when reading in English? If so what would it say about the English language, or rather how we treat writing in the English language?
Or contrast. If you look carefully, this whole comment section is low contrast, or at least at a lower contrast at your article. If you look more carefully you notice that a lot of big web sites (obviously wanting to project a credible image) use a lower contrast than the highest possible of black-on-white. This suggests that in the design culture (or, since those clients have approved the designs) certain parts of corporate culture, higher contrast does not necessarily mean better. Obviously, the lower possible contrast has now been raised as a result of disability legislation, but the ideal contrast is, as far as I can see, not generally perceived to be the highest possible.
Historically formal writing is used more often, including certain cultures where you MUST write in a formal form. (This is still true in a certain sense for my native language, where we are not supposed to write in the way we speek; we must write in a more formal form.) Did people in the past associate formal writing with untruthfulness? I highly doubt it. Do people in cultures where they MUST write in a formal form to be considered credible feel the same thing about formal forms in English?
In the past Fractur was used for writing German and there was a claim that Fractur, when people are familiar with it, actually speeds up reading. We now find Fractur almost impossible to read simply because it is no longer used. The same can be said of italics (they were initially designed to replace roman, not to complement it), cursive handwriting (everyone used to write in cursive and I was shocked when I moved here to find no one write in cursive), all caps, or the formal register.
By the way, I actually have anecdotal evidence that the claim that mixed case is easier to read is false. (I regurgitated the claim that mixed case is easier to read, and was summarily refuted.) Familiarity with the form is key, and all the advice saying avoid all caps will just cause people to become less and less familiar with all caps, causing it to become more and more difficult to read — this is basically just a self-fulfilling prophecy.
Really interesting thoughts here. I can’t believe how ethnocentric I sometimes am! I see how the connection of high formality level (lack of familiarity) with trust can be culturally specific. I’ll do some more thinking about this . . . I smell a new blog post in here somewhere.
By the way, the specific example in #2 (Memo C) is just a change in registry. This has been discussed in a forum in response to some government decisions, and the comments so far has been negative. Since a higher registry is associated with more formal writing which is in turn associated with the government, all the Memo C example suggests is that we distrust government.
When all government foregoes formal writing and use “plain language” throughout, if my hypothesis is correct, then people will start associating “not credible” with plain language and formal writing should gain credibility. IMHO it’s much more useful in the long term to teach people to think critically instead of blindly saying so-called unfamiliar words makes a piece of text look less credible.
While teaching people to think critically (I assume that means helping them expand their vocabulary?) is a laudable goal, it doesn’t help the writer whose task it is to deliver a recommendation to her boss.
I agree. But what will that do? In my opinion that just exacerbates #2. This is an artificially created feedback loop where readers are reinforced false ideas like that all caps is hard to read (they are not), or that handwriting is hard to read (they were not), or that formal writing necessarily suggests untruthfulness. Then we’ll get more research saying that all caps is hard to read and handwriting is bad and coloured paper is bad. This is just wrong.
A document written in all capital letters is accurately perceived as difficult to read. It is not just perception but actual reading speed and accuracy that is affected. Capital letters reduce fluency because they eliminate word shape. See a simple illustration here: http://uxmovement.com/content/all-caps-hard-for-users-to-read/
I know there is some controversy about reading by word shape. See http://www.blog.theteamw.com/2009/12/23/100-things-you-should-know-about-people-19-its-a-myth-that-all-capital-letters-are-inherently-harder-to-read/. But I remain skeptical that the fluency problem of all capitals is really about familiarity.
I know about this. But I’m not convinced. As I mentioned above, I have anecdotal evidence, and after I was refuted I honestly asked myself and I agreed that I didn’t actually believe in what I regurgitated (i.e., I didn’t actually personally find mixed case easier to read).
That anecdote was about a very specialized form of text (computer code), but if this was true in one specialized form, it is enough to make be suspicious of any general claim that all caps must be harder to read. There are scenarios where practical production constraints will make all caps more appealing or leave all caps the only practical choice, such as in cases where the height of the type body is fixed and using mixed case will cause the type size to become too small.
Might I also add that accuracy of perception is extremely important in computer code, yet mixed cases does not necessarily translate to easier-to-read. That speaks volumes.
One more counterargument from the cross-cultural perspective: In the Chinese language, all characters (including, in some typographic traditions, punctuation marks) are set within an imaginary rectangle. In extreme cases, the whole composition is monospaced, and the shape of every character approximates a grey square. If the all-caps-is-harder-to-read theory is correct, then Chinese must be exceptionally difficult to read because there is, according to this theory, no possibility to differentiate word forms. Yet this is obviously untrue and people obviously can still differentiate word forms even though every word looks like a grey square.