Why Do I Read So Slowly? How to Find Out Which of the Four Brakes Is Yours

"I read slowly" is a symptom, not a diagnosis

It's a lot like a headache: it tells you something is wrong, not what.

Two people with the same number on the stopwatch can have completely unrelated problems. One hears every word spoken aloud in their head and can't outrun the pace of their own inner voice. The other reads fluently and effortlessly, but by the third page realizes they've been thinking about something else for two minutes. The same course helps the first and costs the second six weeks.

This piece isn't about solutions — it's an inventory. By the end you should know what to fix, not how. The brakes run automatically, below the threshold where you'd notice them, so you have to catch yourself in the act, and that takes a procedure.

The baseline you start from

Without a number, "slow" doesn't mean anything. A meta-analysis covering 190 studies and more than 18,000 adults puts silent reading at 238 words per minute for non-fiction and 260 for fiction (Brysbaert, 2019). These figures aren't broken down by native language, so treat them as the best available anchor, not a passing grade.

Measure before you diagnose: many people who call themselves slow are actually right at the average and reading something hard. They have no brake to fix — they have the wrong expectation.

The four brakes, seen from the inside

The voice that reads along with you

You hear the words pronounced in your mind, with intonation. On longer ones there's sometimes a small movement in the tongue, lips, or jaw, as if you were mouthing them without sound.

The outside sign is different: your speed sticks to one value and won't climb, no matter how hard you concentrate. Reading aloud sits, in the same meta-analysis, at 183 words per minute, and if the voice in your head pronounces everything, you're anchored somewhere near that number.

It isn't a parasite you simply get rid of, though: suppressing it can be trained, but it comes with a measurable cost to comprehension, and the method and its price deserve a separate discussion.

The eye that doubles back

This brake is almost imperceptible, which is exactly why it slips past you.

Roughly 10 to 15 percent of eye movements across a page are regressions onto text already read (Rayner, 1998). They last under a second and go unnoticed precisely because they're fast, so you only remember the big regressions — the ones where you consciously reread a paragraph.

The nuance matters a lot here: on hard text, a regression is a repair, not a flaw, and blocking it artificially lowers comprehension (Schotter, Tran, and Rayner, 2014). When it's reflex and when it's useful work is a whole discussion on its own.

The mind that wanders off the page

Its signature is the total absence of effort, which is exactly what makes it so hard to catch.

In a study cited by Brysbaert (2019), readers reported task-unrelated thoughts during 42 percent of the sentences in a difficult text and 36 percent in an easy one (Feng, D'Mello, and Graesser, 2013). In those moments, reading times rose by nearly a second per sentence.

It isn't a personal flaw — everyone reads this way. What differs is how fast you notice and come back to the page.

The words that cost you too much

Here you feel slow only on certain texts. You fly through a novel, then get stuck on a contract or a textbook, and conclude you have a reading problem.

You don't. You have a vocabulary problem in that domain, and it shows directly in your eyes: the time spent on a word reflects the processing it demands, and rare words hold your gaze longer (Just and Carpenter, 1980).

Out of an entire battery of cognitive skills tested, the one that best predicts reading speed isn't eye strength or the width of your visual span — it's how fast you recognize words (Kuperman and Van Dyke, 2011). It's the brake nobody wants, because the fix isn't a technique.

Three minutes of watching yourself

You need an unread book of medium difficulty and a stopwatch.

There's one hard rule: read at your normal pace. You want a representative sample, not a performance.

While you read, split your attention in two: one part follows the text, the other watches the process, like a silent observer that doesn't comment. Move it through four zones, about 45 seconds each:

1. Mouth and jaw. Do you feel any movement, however small, on certain words?
2. Inner ear. Do you hear the words pronounced, with intonation, or just understand them?
3. Eye path. Do you drift back to the left? Do you reread the end of a sentence "just to be sure"?
4. Presence. At the end of every page: can you say what was on it?

Note all four at the end, including the weak ones, then pick the dominant one — the one that, if it vanished, would free you up the most.

Two mistakes ruin the test. The first is correcting yourself while you observe: if you tell yourself "stop subvocalizing," your attention jumps into control mode and the report turns false. The second is only checking the brake you expected to find. Check all four, especially the regressions.

The tests that tell them apart

Self-observation gives you suspects. The tests below confirm them, in under ten minutes each.

The voice test. Read a passage aloud and time yourself, then read a comparable one silently from the same book. If the two numbers come out close, your inner voice is pronouncing everything and it's your ceiling. If silent reading beats it by a hundred words per minute, the voice isn't the main brake.

The easy-text test. Repeat the measurement on a page from a novel you'd read for pleasure. If your speed jumps, the brake is the material, not the habit: habits follow you everywhere, text density stays on the shelf.

The pencil test. One page, one pencil, one tick mark in the margin every time you catch yourself going back. Where you go back to matters too: to the word right before, or to the section heading? The second isn't a brake, it's navigation.

The end-of-page test. Say out loud, in one sentence, what was on the page you just finished. If you struggled and still can't say it, it's the text. If you felt no effort at all and still can't say it, you weren't there.

What you measured	Likely brake
Reading aloud and reading silently give close numbers	inner voice
Fast on fiction, slow on any technical text	vocabulary and density
Lots of tick marks in the margin, at the word right before	reflexive regressions
Zero effort felt and zero content retained	attention

What shows up on lists and isn't a brake

Almost every speed-reading course lists "excessive fixations" too — stopping on every single word instead of groups of words.

Except that isn't a habit you can change by deciding to. The zone from which you can actually identify letters extends about 3 to 4 characters to the left of your fixation point and 14 to 15 to the right (Rayner, 1998). The rest of the line is shape, not text. The number of stops comes out of how fast you recognize words, so it's a consequence of the fourth brake, not a fifth one. The most complete review of the field says it without hedging: there's no silver bullet (Rayner, Schotter, Masson, Potter, and Treiman, 2016).

It's worth saying what you actually gain, too. In a recent controlled study with a small sample, a commercial speed-reading app produced about 42 words per minute more than the control group, while a group that only listened to an hour of explanations about how reading works gained 29, with no statistically significant difference between the two (Klimovich, Tiffin-Richards, and Richter, 2023). Dozens of words per minute, not a doubling.

The most uncomfortable part comes last: you're an unreliable witness in your own case. In a report commissioned by NASA on photoreading, a certified instructor had rated her own comprehension of the material 4.5 out of 5 before the exam, then scored 35 percent on the quiz, where pure guessing gives you 25 (McNamara, 2000). The feeling that you understood is the worst measuring instrument you own, and the feeling that you read slowly is just as unreliable. That's why the procedure ends in numbers, not impressions.

What to do with the answer

Each brake leads somewhere different, and there's no point attacking all four at once.

If the voice is dominant, work on articulatory suppression and accept the trade-off in comprehension. For reflexive regressions you need a tool that blocks them on easy text, not more willpower. When your attention leaves, the answer isn't an eye technique — it's a question you actually want the answer to, and shorter sessions. And for vocabulary, the only shortcut is that there isn't one.

Redo the diagnosis in a few weeks, on comparable text. The dominant brake changes, and it's a good sign when it does. But start with the number — otherwise you have nothing to work from.

Measure your reading speed and comprehension, free