Why we miss our biggest mistakes

Have you ever missed an obvious typo and wondered how on Earth you did it? Perhaps it was in a subheading in large, bold type. Or maybe it was something that should have been even more obvious. Say, a howler of a mistake in the title of a PowerPoint deck that only revealed itself when you presented it live on Zoom or at a conference before a packed auditorium.

I’m sorry if the memory just made you wince. But if it did, you may take just a little comfort from knowing that you are not alone. We all make these mistakes. In fact, I can hardly bear to type these words for fear that ‘finger trouble’ will lead me to make an embarrassing error that might undermine my claim to any expertise in this area.

That's why I’ve spent the last few weeks digging into the brain science behind something that we rarely talk about: the mechanics of how we read. (And by that, I mean how we really read – not how we’d like to think we do.) This is an area that's a lot more important than most people realise. In fact, I'd argue that life would run a lot more smoothly for all of us if we understood how our eyes detect words on a page and how our brains then process them. Ignorance and (arguably worse) myths surrounding this process stop us getting to the heart of understanding not only reading but why we don’t always get the reaction we were hoping for.

It also explains why we often fail to see our biggest mistakes until it’s too late. Usually, these mistakes cost us little more than a red face and perhaps a few snickers from whomever we happen to be writing to. But sometimes, the consequences are a little less trivial.

Take the case of the telephone company Pacific Bell, for example. It found itself at the end of a 10 million dollar lawsuit after it printed an ‘r’ instead of an ‘x’ in an ad placed by a travel agency for vacations to what should have read ‘exotic’ holiday destinations. The travel agency owner, Gloria Quinnan, claimed the mistake had scared off half her customers, as the Yellow Pages promotion of the company’s trips to ‘erotic’ destinations had alienated her elderly clientele. The error, she argued, had caused her mental anguish, not just through loss of reputation but because she had to fend off countless calls from a quite different market demographic.

The 331 million-dollar typo

Or, if you think that’s bad, consider the tale of a hapless trader working for Mizuho Securities, a Japanese investment bank.

It was 9 December 2005, a momentous day for a recruitment company called J‑Com, which was making its debut on the Tokyo Stock Exchange. As its price rose, the trader placed an order to sell one of the firm's shares when the price reached 610,000 yen.

At least, that’s what he meant to do. What he actually typed in was an order to sell 610,000 shares for 1 yen each. The trading system dutifully accepted and implemented his order, losing his employer the equivalent today of a third of a billion dollars in the process. As if that weren’t bad enough, his error undermined investor confidence so much that the company’s price collapsed completely.

As the calamitous cascade continued, it triggered a stock sell-off that wiped 300 points off Japan's Nikkei share index. It even prompted a statement from the country’s prime minister, in an attempt to reassure investors and calm the markets. And all because the trader failed to spot that he’d typed one thing but meant another.

I’ve witnessed such errors first hand, made by people who you’d think really should have known better. In the mid-1990s, I was working as a sub-editor on a magazine for a major publisher. Standard procedure at the time was for the graphic designer to create the page layout, using the images that would be in the published mag but filling in the text columns just with nonsense Latin text. (The text usually starts with the words lorem ipsum, and these days it’s often included in word-processing apps such as Microsoft Word.)

The designers were meant to use the same, meaningless content for the headlines and picture captions too, so that it would be obvious to those of us on the subs desk that we needed to type in the real content in its place.

Few of the designers did that, though. When they were in full, creative flow, it could be too much of a hassle to come out of the page-layout software they were using and go hunting for some lorem ipsum to copy and paste. Instead, they’d just key in some nonsense themselves that they were certain the subs would catch. Their faith in us was touching but in this case totally misplaced. We may have been able to work miracles with barely coherent copy from our contributors. But we were still as human as anyone else.

Honest mistake: even the Lincoln Memorial has a typo

So it was that an editor with whom I shared an office once signed off her magazine and sent it to the printing presses even though it still contained a photo caption that read: ‘No doubt Susan will type something jolly interesting here.’ Worse, one of the UK’s leading computer titles at the time – from the same publishing stable – ended up on the shelves of a national newsagent chain with ‘Type some bollocks in here!’ splashed across its centre pages in 96-point type.

Ouch. And yet, none of us is immune. We are all doomed to miss typos and other errors from time to time. There’s no shame in it. I even missed one myself in my recent post about the Sally/Anne test, for example. (Thanks again to the eagle-eyed Silent Influence subscriber who kindly point it out.)

Seeing what you expect

In fact, there is something unique about errors in written text that makes them especially hard to spot. It all comes down not just to the way our brains are wired but to how our eyes detect what’s in front of us. Once you understand these limitations, you’ll understand not just why we’re often blind to our own mistakes but why we’re actually more prone to miss the biggest typos. In fact, the larger the type, the more likely we are to miss it.

Part of the problem is that the brain sees not what’s actually there but what it expects to see. As we read, we hear words in our head, but sometimes we get ahead of ourselves, hearing words before we read them and thinking they’re there when they’re not (or not seeing them when they’re right in front of our eyes). This is why reading out loud can be a good way to check what you’ve written, as it forces your brain to slow down. (I often catch errors when recording the audio versions of Silent Influence lessons.) Yet there’s a lot more to it than simply falling prey to wishful thinking. And at the root of it is not just neuroscience but mechanics.

As you are reading this, your eyes are continually focusing and refocusing on fragments of text. They are scanning the words, not in one smooth movement, but hopping from fragment to fragment, in a series of jerky movements called saccades. (‘Saccade’ means ‘jerk’ in French.) And in between each saccade, they fixate on a portion of text for half a second or less.

Narrow focus

What may surprise you is just how narrow your area of focus in each of these fixations is. Stop for a moment and focus on these three words. And when I say focus, I mean really focus, so that all three words are as sharp as a pin. It’s not easy, is it? As you look at each word, the other two keep drifting in and out of focus. That’s because the focus point in the retina in each of your eyes only has room for seven to nine letters of average-sized printed or on-screen text. So if you focus on the middle word – ‘three’ – you can also focus properly on the last letter of the word on the left and the first of the one on the right (‘e’ and ‘w’ in this case), and that’s about it.

You’ll notice that all the other words, in the line and the rest of the paragraph, will be a blur.

The focal point in your retina is called the fovea. It’s an area packed almost exclusively with the type of photoreceptors called cone cells (based on their shape). And it’s this type that specialises in processing detail and colour, and that’s responsible for visual sharpness.

It’s this area that also has the most nerve connections and on which the brain relies while reading. And it relies on it totally. For the 150-500 milliseconds that each fixation lasts, the only thing the brain can focus on is those few letters.

That’s assuming that they’re of average size – say 11 points or so. If they’re smaller (and assuming you can still read them), you’ll see more letters. But the bigger they are, the less you’ll be able to focus on. And that fact, right there, is why we miss the biggest, most embarrassing typos of all. If we’re faced with a sentence in 96-point text, we can’t focus even on a single letter if what we’re reading is at arm’s length. It’s simply too big for the fovea to cope with. Instead, we’ll only see a bit of it. What you need is more distance, to effectively make the text small enough to fit the fovea. Like, say, the kind of distance an audience of 1500 people will have when you’re delivering a keynote presentation.

You do see the letters either side of those you’re able to focus on, but not many of them. Researchers in this field have found that, on average, it’s four letters to the left and fifteen to the right (in languages that read from left to right). But what’s really interesting is what the brain does about the words that it can’t see. Because what it does, is it just makes them up. That’s right: what we think of as reading is mainly hallucination. We think we’ve checked the title on a PowerPoint slide. But we’ve only imagined we’ve seen it.

Back in 1975, two scientists named George McConkie and Keith Rayner, from Cornell University and the University of Rochester, conducted an almost fiendishly cunning experiment to demonstrate this. (That’s right, 1975 – this is not new science, even though most people have never heard of its discoveries or know the truth about how we really read.) They coupled an eye-tracking device with a computer program that replaced the words either side of wherever volunteers were looking with a series of x’s. Whenever the volunteer moved their eyes in the usual hopping motion along each line, the computer would refresh the display and show whatever letters should be there, but again replace everything else with x’s.

The result? The volunteers were none the wiser. They literally had no idea that anything had changed. As long as the correct letters were in place wherever they happened to focus, it didn’t matter that everything else had been replaced by gibberish. The volunteers were still able to read the text and they didn’t suspect a thing.

Love is blind

It’s no wonder we miss the odd typo from time to time. What’s more miraculous is that we spot any of our mistakes – especially given that we’re often so relieved at having finished writing that we just assume that it will be as clear and perfect to the reader as it seems to us.

Love really is blind, in this case literally.

Sometimes, when your writing is in full flow, it can be useful to use placeholder text. Adding temporary content in the place of a key fact keeps you from having to stop and risk losing momentum. It can also help you focus on creating the layout of, say, a PowerPoint slide deck, without having to get distracted from the task by what you’re actually going to say.

But if you do that, make it obvious that whatever you type should not be there. And I don’t mean obvious now, when you know it’s just a placeholder, but obvious when you come back to it a week later, full of confirmation bias that everything looks pretty good and ready to go. (The more times you read something, the more inured you become to its imperfections – it’s called the mere exposure effect.) Even better, use nonsense lorem ipsum Latin text, so that you’ll spot it when you run your spellchecker for the last time. There are even websites that will generate that text for you.

Remember, the biggest type is too big for your eyes to read when it’s on a screen right in front of you. So they don’t see what’s there but what you think is there.

This fact has wider implications. It means that whoever is reading what you have written will also see what they expect to see, too. They may well have already made up their mind about what you’re going to say before they even open your text, email or proposal. If that's about to lead them to a decision that's not in your favour, what you write is going to have to do an awful lot of heavy lifting.

This is a big enough topic in itself to merit a post of its own and one for another day. But in the meantime, just remember: whatever you choose as placeholder text, try to make sure you’d be happy to see it in 96-point type.

Sources for this post

Keith Rayner – he of the cunning experiment that replaced words with x’s – co-wrote the book Psychology of Reading with three of the other most prominent scholars in this area of research. It’s the key academic text on this topic. Running to almost 500 pages, it’s not exactly a light read. But each page is packed with references to the most important research and it’s the go-to book for answers to most questions about how we make meaning out of squiggles and dots on a page. Complementing it perfectly is neuroscientist Stanislas Dehaene’s Reading in the Brain, which focuses more on the neuroscience of reading. It’s also superb, as well as being more accessible to the lay reader.

Rayner and McConkie’s original 1975 paper on their experiment is here [PDF]. And finally, the website loremipsum.io includes not just a useful lorem ipsum generator but details of where the nonsense text originated and what it actually means.

Photo credits: ViDI Studio/Shutterstock, National Mall and Memorial Parks.