An Illusion that Explains Why Typos Are So Hard to Catch

Illustration for article titled An Illusion that Explains Why Typos Are So Hard to Catch

Every time you type an email or a document, errors are likely to creep in — and no matter how carefully you proof read, you might not catch everything. Why do we have such a hard time noticing typos and repeated words?

It turns out this is partly a matter of the way your brain processes text — and partly something that's specific to English and a few other languages.

Top image: Pond5

We've established before that reading is so trained into our culture, we do it instinctively. Sometimes our eyes bounce off a page of dense, boring text without absorbing any content — but for the most part it's almost impossible to look at a word and not read it. In the past, we've shown how deep into our brain reading reaches with the Stroop Effect. The Stroop Effect happens when the words for one color are printed in ink of another color. Asked what color the text is printed in, people will get confused. If the word "green" is printed in blue ink, even though their eyes see blue, their brain thinks "green." It takes people a while to sort it out.


But, you ask, if we're so comprehensively steeped in the printed word that we just can't help but read what's in front of our faces, why do we so often get tripped up on typos and misspellings? It turns out, there's a whole class of word illusions clustered around our instinct to miss errors.

I had a teacher who put up a poster in class with a triangle printed on it. Inside the triangle were the words "Don't Take Anything For Granted." Except they weren't those words. They were "Don't Take Anything For For Granted," and the text was arranged so that one "for" was on one line, while then the next line had another "for." People invariably missed the double "for," and the teacher got to be smug about them taking things for granted.

Illustration for article titled An Illusion that Explains Why Typos Are So Hard to Catch

It's true that we miss those types of things routinely. But that's generally because we look for what writing is — and not what writing isn't. What writing is, generally, is a way to communicate meaning from one person to another. While it's fun to see what the brain trips over or forgets, there's a more interesting illusion that allows people to see how much they understand of a puzzling paragraph. The letters are scrambled, except the first and last letter of each word. While it can be difficult to read some of the words, it's fairly easy to breeze through the entire text and understand all of it.

How much does this work? The scrambled word illusion became so popular that researchers at the MRC Cognition and Brain Sciences Unit, which is closely linked with Cambridge, were asked about it and replied to it. And when they looked into the scrambling in English and other languages, interesting patterns emerged.


They noticed that it was actually possible to render the whole thing incomprehensible by putting the letters too far away from where they ought to be in a correctly-spelled word. The entire thing benefits from the fact that English has an abundance of medium-sized words, with a few redundant letters in them. It's possible to scramble those letters around while keeping them near where they would be, if the word were in its correct order. German, scientists found, is also a good language for that sort of thing.

When a colleague tried it with Hebrew, on the other hand, it became an unreadable mess. Apparently vowels are not written — not being a linguist I did not know this — and so the text as written already makes use of this tendency of people to fill in intuitively what they need to in order to understand the text. Scramble the letters and that intuition is derailed, making it impossible to understand. This trick relies on a language to be explicit in order to keep people understanding it when it's muddled.


Top Image: Tom Murphy VII

Via MRC and Josh Nimoy.


Share This Story

Get our `newsletter`


Thomas Skogestad

When did Cambridge research this?