Why You Miss Typos in Your Own Writing

The best plain-language explanation comes from psychologist Tom Stafford of the University of Sheffield, interviewed in WIRED's 2014 piece on typo blindness: missing your own typos is not carelessness. It is a side effect of competence. Writing is a high-level task. To do it at all, your brain automates the low-level components (spelling, letter order, function words) so it can spend attention on the hard part: meaning. Once those components are automated, your brain stops inspecting them.

Then comes the cruel part. When you proofread your own work, you are not reading fresh text. You are re-running a text you already know the meaning of. The version in your head competes with the version on the screen, and because reading is heavily predictive, the head version fills in whatever the screen version is missing. You do not see “hte” because your brain hands you “the.” You do not notice the missing word because the sentence was complete in your mind before your eyes reached it.

Because you have no inside knowledge to betray you. Reading a colleague's draft, every sentence is new information: your brain cannot predict the next clause, so it actually reads the words on the page instead of skimming past them. That is the whole asymmetry: their typos live in text you process; yours live in text you remember. The author is, structurally, the worst-placed person to proofread the work. Every trick that follows is an attempt to turn you back into a stranger to your own draft.

They work to the degree that they make the text unfamiliar, which is exactly the advice Stafford gives in the WIRED piece: change the font, change the background color, print it out and edit by hand.

• Waiting is the gentlest version. Familiarity decays with time, so a draft you have not seen for a week reads almost like someone else's. The catch: deadlines.
• Font, color, paper all break the visual memory of the draft. Cheap and worth doing, but the effect is partial, since the words still arrive through the same predictive reading loop.
• Reading backwards is the extreme case: it destroys meaning so each word gets inspected alone. That catches misspellings, and it structurally cannot catch a missing “not,” because meaning is what a missing word breaks.

All three attack familiarity from inside the same channel: your eyes. There is a fourth option that switches the channel entirely.

A text-to-speech voice is the stranger you cannot become. It has no intended meaning to defend and no memory of your draft. It just reads the page, verbatim. Listening moves proofreading out of the predictive visual loop where typo blindness lives: a missing word becomes an audible pothole, a doubled word gets dutifully spoken twice, and a sentence you mangled while rearranging clauses sounds exactly as mangled as it is.

This is why a listening pass and a spellchecker are complements, not rivals. The spellchecker catches non-words; the voice catches the perfectly spelled sentence that is not the sentence you meant.

If the draft is private (an unpublished manuscript, a grant application, a personal essay), the tool matters. LoudReader reads your exported PDF or EPUB with natural offline voices and is fully on-device and private, your library never leaves your device. It also highlights each word as it speaks, so when your ear flinches, your eye is already on the offending word.

The version that sticks is small:

1. Finish the draft. Run the spellchecker first and let it eat the easy errors.
2. Export to PDF and open it in LoudReader (free, no account).
3. Listen with the text in view. Pause at every flinch. The flinch is the finding.
4. Fix, then re-listen to just that paragraph before moving on.

The full technique guide is proofread by listening; for the essay-specific version (what to listen for, how to loop on fixes), see read my essay out loud: edit your essay by ear. And if you are wondering what LoudReader does and does not do, the FAQ covers it honestly.