A Fonty Conundrum

This page has now been moved to some thoughts on which font is BEST.

Researchers Get Their Solutions in a Twist, Font-Wise

On other pages, I have commented on how the question of ease of reading of typefaces or fonts raises more questions than the people who think they have the answers think it does. Now The Economist chips in with an article.

In The Economist of 16–22 October 2010 there was an article titled Learning difficulties in which the first paragraph said: ‘A paradox of education is that presenting information in a way that looks easy to learn often has the opposite effect. Numerous studies have demonstrated that when people are forced to think hard about what they are shown they remember it better’.

Daniel Oppenheimer, a psychologist at Princeton University and colleagues recruited 28 volunteers aged between 18 and 40 and asked them to learn, from written descriptions, about three “species” of extraterrestrial alien, each of which had seven features. This task was meant to be similar to learning about animal species in a biology lesson. It used aliens in place of actual species to be certain that the participants could not draw on prior knowledge. [I’m quoting here from the article]

Half of the volunteers were presented with the information in difficult-to-read fonts (12-point Comic Sans MS 75% greyscale and 12-point Bodoni MT 75% greyscale). The other half saw it in 16-point Arial pure-black font, which tests have shown is one of the easiest to read. [The article does not say which tests these were that showed Arial 16pt black easier to read than Comic Sans or Bodoni 12]

[Still quoting] Participants were given 90 seconds to memorise the information in the lists. They were then distracted with unrelated tasks for a quarter of an hour or so, before being asked questions about the aliens, such as “What is the diet of the Pangerish?” and “What colour eyes does the Norgletti have?” The upshot was that those reading the Arial font got the answers right 72.8% of the time, on average. Those forced to read the more difficult fonts answered correctly 86.5% of the time.

[Still quoting] The question was, would this result translate from the controlled circumstances of the laboratory to the unruly environment of the classroom? It did. When the researchers asked teachers to use the technique in high-school lessons on chemistry, physics, English and history, they got similar results. The lesson, then, is to make text books harder to read, not easier. (End of quote]

Now we know that researchers can find themselves barking up the bonkers tree from time to time, and one has to be careful therefore with taking the results of research at face value. This particular piece of research, at least as it is reported in The Economist, is pregnant with questions that would need answering before we decide upon its worth.

The first question is whether Comic Sans and Bodoni are actually harder to read than Arial, for if they are not, if the reverse is the case, then the conclusions of the research will be the exact contrary.

Similarly with type size. I would have thought that 12 point in blocks of text was easier to read than 16 point, but perhaps the experts know better (ahem).

And similarly with greyscale. Many people say they find black on white too harsh. Possibly the 75% greyscale was a defining factor in making the text in Comic Sans and Bodoni actually easier to read not harder.

And then there’s this: if we accept that people remembered better the harder-to-read font, is this not a justification for making the fonts that we use harder to read, on the basis that we want what we write to be better noticed?

On the topic of whether Arial 16 point is actually easier or harder to read than some other typeface and size, the actual answer is that it depends on context. I can categorically say, whatever any researcher may seek to prove with experiment, that if you went to the bookshop and picked up a paperback novel set in Arial 16 point you would find that very hard to read. As a mini-display, a leaflet or ad say, you could find Arial 16 point very comfortable. One does not need academic experiment to prove this, you just will, you know you will.

Is Mr Oppenheim barking up a wrong tree altogether? Or just barking? Could be both I suppose.

Next page in this set: The Academic Evidence Base for Typeface Readability .