UI Design Newsletter – February/March, 2009

May you live in interesting times

Step away from the edge.

A small detail in the Kindle ad tipped my decision to try it. Unless you've spent years thinking about the psychology of reading, you probably won't notice it.

The text on the right side of the "page" is not right-justified. It has a ragged-right edge. And that, for me, was tantalizing.

In right-justified text, the size of spaces between words is varied to make the lines come out even. The size of the spaces is irregular but not meaningful. The goal is just to make the lines come out even. But your brain registers that the spaces are different sizes. And it tries to sort out what that information conveys. Trying to interpret signals that are noise slows you down and makes reading feel more effortful.

You say red herring. I say red herrings.

To be sure, part of the fun of Kindle is that Amazon had to balance a lot of design options. For instance, they tried to create an unpacking-the-product-should-be-emotional experience. For an Apple native, that was a bit weird. Somehow, the pull-off paper zipper (think FedEX envelopes) sets the wrong tone. But, they tried.

Out of the box, my first impression was positive: It's smaller than I thought. The text resolution is better than I had hoped.

But then, there is no backlight. This means the battery lasts a really long time. But it also means the screen is surprisingly grey and the text contrast is low. And you still need a nightlight to read.

The keyboard lets you annotate while you read. But it's awkward. Big. That choice seems odd since the bigger keyboard means a smaller reading screen. We were all trained to type on phone-sized keyboards, weren't we?

Navigating isn't bad. The fact that the Menu button takes you to Shop At the Kindle Store is irritating. Even if I understand why it's so. The joystick offers a few surprises, such as, you can't turn pages with it.

But these problems are red herring(s). Kindle isn't really about unpacking and navigating. It's about reading. Unfortunately, the reading part is where Amazon goes wrong.

Just because you can, doesn't mean you should.

It may seem counter intuitive that a small detail like where lines end would make text easier to read. If it's true, why do publishers of books, journals, magazines, and newspapers right-justify everything? It's a question that a lot of us who think about reading think about a lot.

And the first publication I opened (the New Yorker) lived up to the promise: Hertzberg, in ragged right. With the cartoons thoughtfully aggregated into one section. Joy.

But the second one I opened (Nudge; Improving Decisions about Health, Wealth, and Happiness), and the third, fourth, fifth, and sixth (White Tiger: A Novel; I was told there'd be cake (Essays); Technology Review; and the Wall Street Journal, respectively) all have right-justified text. Gain the buyer's trust. Violate the buyer's trust.

To be fair, it may not be Amazon that is making the choice. But, they could. And if any single group can make reading better, it is Amazon. Well, maybe Amazon and ReadSmart.

One stop past the end of the line...

Actually, there's more to enhancing readability than where lines end. Remember how the random size of spaces between words in right-justified text undermines reading? The reverse is also true: Bever and colleagues, Linguists and Psychologists at the University of Arizona, have shown that when line-ends and space-sizes offer clues to how words should be grouped, reading is faster and feels easier (Bever, Jandreau, Burwell, Kaplan & Zaenan, 1990; Jandreau & Bever,1992, among many others.)

To show this, Bever and team engineered (and patented) a text processing/formatting algorithm (which they call ReadSmart) that "reads" text input and adjusts inter- and intra-word spacing based on psychologically tested, linguistic rules. The new, meaningful spaces guide readers' eyes and helps them to group the words correctly even as they read. "ReadSmarted" text is easier to read because part of the work of reading is already done for you. But unlike other text formatting algorithms, ReadSmart improves readability without changing the length of the text, or the way it looks on the surface.

Bever's early studies of linguistic formatting (including more than 500 students in the U.S. and abroad) showed when the spaces between/within words predict the structure, comprehension and reading speed increase up to 20%. Similar comprehension improvements have been documented for readers under duress and second language readers.

It's much more persuasive if I read it.

More recent industry studies suggest that improving readability and comprehension may also increase persuasiveness. The scenario supposedly goes something like this:

1. Potential donors receive solicitation letters. If they like the cause, they may open the letter.
2. If they open them, they tend to skim for about as long as it takes to get from the front door / mail box to the kitchen / trash can.
3. Since ReadSmarted text reads faster and more easily, potential donors get further into the message to the emotional hook that removes the block or amplifies their drive toward commitment.
4. Along the way, since the cognitive burden of "reading" is reduced by ReadSmarting text, readers can commit more mental energy to processing and remembering the message. Remembered messages are more persuasive.

Even if the scenario is not exactly right, the effect of formatting text to improve readability is profound. Direct mail donor acquisition letters that are formatted by ReadSmart work better. In a meta-analysis of 5 direct marketing campaigns (over multiple charities) reaching 393,000 households, the formatted letters triggered 22% more responses than donor letters with standard text. And, people who responded to formatted solicitations ultimately donated more (48% more on average)1.

More than meets the eye.

Ultimately, linguistically-informed text formatting algorithms like ReadSmart make text more persuasive by reducing the mental burden of reading. Readers understand better with less effort. By extension, organizations and agencies that apply these algorithms can benefit, as well. And that promise makes ReadSmart tantalizing.

Sort of like the Kindle2 was. And could be again. If Amazon made "ReadSmart my book" the default menu item. Until then, I'm going back to paperbacks.

[The ReadSmart text formatting algorithm improves readability on paper, in fixed-width websites and mobile phones. To learn more go to www.readsmart.com.]

1 To be sure, the scale of the behavior change resulting from ReadSmarting text makes it feel a bit like a but-wait!-there's more! paid-for-TV commercial. I'd be far more skeptical, if I didn't have direct knowledge of the psycholinguistic research behind text-formatting generally and ReadSmart specifically. (Fair balance: Bever was my thesis advisor.)

References

Anglin, J. M., & Miller, G. A. (1968). The role of phrase structure in the recall of meaningful verbal material. Psychonomic Science, 10, 343–344.

Bever, T. G., & Robbart, J. (2008). System and method of determining phrasing in text. U.S. Patent No. 7,346,489. Washington, DC: U. S. Patent and Trademark Office.

Bever, T. G., & Robbart, J. (2008). System and method for formatting text according to phrasing. Patent pending. Washington, DC: U. S. Patent and Trademark Office.

Bever, T. G., & Robbart, J. (2006). System and method for formatting text according to linguistic, visual and psychological variables. U.S. Patent No. 7,069,508. Washington, DC: U. S. Patent and Trademark Office.

Bever, T. G., Nicholas, C. D., Hancock, R., Alcock, K. W., & Jandreau, S. M. (2007). System, plug-in and method for improving text composition by modifying character prominence according to assigned character information measures. Patent pending. Washington, DC: U. S. Patent and Trademark Office.

Bever, T. G., Jandreau, S., Burwell, R. , Kaplan, R., & Zaenan, A. (1990). Spacing printed text to isolate major phrases improves readability. Visible Language, 25, 74–87.

Coleman, E. B., & Kim, I. (1961). Comparison of several styles of typography in English. Journal of Applied Psychology, 45, 262–267.

Hartley, J. (1980). Spatial cues in text. Visible Language, 14, 67–79.

Hartley, J., & Burnhill, P. (1971). Experiments with unjustified text. Visible Language, 5, 265–278.

Jandreau, S., & Bever, T. G. (1992). Phrase-spaced formats improve comprehension in average readers. Journal of Applied Psychology, 77, 143–146.

Nicholas, C. D., Maher, J. Ashley, K. L., Berendt, L. H. (2009). System and method for converting the digital typesetting documents used in publishing to a device-specific format for electronic publishing. Patent pending. Washington, DC: U. S. Patent and Trademark Office.

Klare, G. R., Nichols, W. H., & Shufford, E. H. (1957). The relationship of typographic arrangement to the learning of technical material. Journal of Applied Psychology, 41, 41–45.

Mason, J. M., & Kendall, J. R. (1979). Facilitating reading comprehension through text structure manipulation. The Alberta Journal of Educational Research, 25, 68–76.

North, A. J., & Jenkins, L. B. (1951). Reading speed and comprehension as a function of typography. Journal of Applied Psychology, 35, 225–228.

We know that if users can't find it, they can't be persuaded by it. But this goes a bit beyond. It shows if the user must work hard to get to content (even in a small detail like text quality) it can make a material impact on persuasion. Usability is no longer enough, but it is absolutely needed.