Common Mistakes When Learning Braille

The most common mistake when starting braille is not about memory: it is pressing too hard. Pressing harder gives you no extra information about where the dots are (that has been measured), and it tires your hand and slows your finger. The eleven mistakes that follow are just as fixable, and most are made without knowing it.

This article is not a list of warnings. It is a diagnosis: each mistake has a card with what it is, why it happens, how to spot it and what to do instead. If you want the whole path from the start, it is in how to learn braille; that guide lists these mistakes, and this one takes them apart.

Key points

• Pressing does not help. Quadrupling contact force, from 50 to 200 grams, improved nothing in touch tasks that depend on where raised features are, which is exactly what reading braille is (Gibson and Craig, 2006).
• Read with the pad of the finger, not the tip. The European Blind Union recommends it explicitly in its 2025 fast reading recommendations: the pads of the index and middle fingers, not the fingertips.
• Stopping and rubbing is the habit that most gives away a slow reader. In a study of blind children, slow readers changed the direction of their finger's acceleration 13.94 times per centimeter, against 5.23 for fast readers (Nonaka, Ito and Stoffregen, 2021).
• Four pairs of letters are exact mirror images in English braille: d/f, e/i, h/j and r/w. We computed them from the braille tables, not from a list.
• What confuses the eye and what confuses the finger are different. Mirror pairs confuse the eye; the finger is confused by a one-dot difference, and the measured groups are different (Özkan et al., 2026, 24 blind readers).
• The numeric indicator is not optional. The first ten letters share their cells with the ten digits, with no exceptions: without the indicator, the text is genuinely ambiguous.

The twelve mistakes at a glance

The first five are technique mistakes: they show in the hand and are fixed in days. Six to eight are method mistakes: they show in the study plan and cost weeks. Nine and ten are code mistakes. The last two depend on who you are in relation to braille.

Technique mistakes: what the hand does

1. Pressing your finger into the dots

What it is. Putting weight on the fingertip, as if the dots would become clearer the harder you push.

Why it happens. Because it is what anyone does with a small raised feature, and because braille feels like a weak signal you have to "pull out". A reasonable intuition, and a wrong one.

The evidence. In 2006, Gregory Gibson and James Craig measured three touch tasks at two contact forces: 50 grams and 200 grams. In the task that depended on the intensity of the stimulus, quadrupling the force improved performance a lot. In the two tasks that depended on the spatial layout of raised features (grating orientation and gap detection), force made no difference (Exp Brain Res 170(2), 172–181). Reading braille is the second kind of task: the question is not how strongly you feel a dot, but where it is. Extra force buys no information. It does cost something: the European Blind Union recommends a light touch because pressing slows recognition and can strain the hands.

How to spot it. The fingertip is marked when you lift it. The hand tires within ten minutes. The forearm tenses. And above all, the finger slows down: it is very hard to glide smoothly while pushing down, so pressing nearly always drags in mistake number 3.

What to do instead. Rest the finger with just enough weight for the skin to touch the dots, and no more. A calibration trick: rest the pad on the page and reduce the pressure until you stop feeling the dots; then add a little. That little is reading pressure. The EBU recommendations add two things that help keep it: relaxed wrists, forearms nearly parallel to the table, and slightly curved fingers resting naturally on the line.

2. Reading with the fingertip instead of the pad

What it is. Touching the paper with the very end of the finger, almost the edge of the nail, instead of the soft flat pad underneath.

Why it happens. Because pointing is something we already know how to do, and we point with the tip. Nobody decides to read with the tip: it is inherited from pointing. It usually comes with a finger held too upright.

How to spot it. Look at the angle of the finger. If the nail is almost perpendicular to the paper, you are on the tip. If the finger lies fairly flat and the part touching is soft and wide, you are on the pad. Another sign: with the tip, it is especially hard to tell apart two dots in the same column, because the contact area is too small to cover the whole cell.

What to do instead. Lay the finger flatter. The EBU's 2025 recommendations are unambiguous: the dots should be touched by the pads of the index and middle fingers, not the fingertips, because that gives more touch surface and spreads the effort instead of concentrating it on one point.

3. Staying on a cell until it "resolves"

What it is. Finding a letter you do not recognize, stopping the finger on it and rubbing it up and down until you identify it. It has a name in the literature: scrubbing.

Why it happens. For the same reason we press: the feeling that the information is there and that persistence will get it out. It is also what we do when reading by sight and a word does not fit (we look back), except that with the eyes going back costs milliseconds and with the finger it breaks the whole movement.

How to spot it. It is the most visible mistake from outside: the finger stops and oscillates. And it is the one that best separates fast readers from slow ones. Takeshi Nonaka and colleagues recorded the exploratory movements of eight blind children and counted how often the finger's acceleration changed direction per centimeter: 13.94 times per centimeter for slow readers against 5.23 for fast readers. Mean finger speed, 3.17 cm/s, correlated with reading performance at r = 0.87 (Scientific Reports 11, 7182). The sample is small (eight children) and the data are correlational, so they do not show that scrubbing causes slowness; but the association is strong enough to use as a warning sign.

What to do instead. Keep going and come back. If a cell does not come on the first pass, carry on to the end of the word and come back. Often you no longer need to, because the context has solved it. It is what the EBU calls "reading with the brain": anticipating from context instead of decoding cell by cell.

A point about the mechanism, because it is often told wrong. People say "touch responds to movement" and conclude that the finger must move. The precise finding is stranger: in a classic experiment, recognizing raised letters with an active finger and with a still finger while the letter moved underneath gave practically the same result, 49.0% against 50.7% correct, with a 0.94 correlation between letters (Vega-Bermudez, Johnson and Hsiao, 1991). What is needed is not that the finger moves, but that there is relative movement between skin and dots, and that the cell is crossed in one sweep instead of sitting under the finger indefinitely.

4. Moving the paper instead of the hand

What it is. Keeping the hand still and dragging the sheet underneath to move along the line.

Why it happens. Because it works for the first five minutes. Sliding a sheet is easier than learning to glide a finger in a straight line, and with single cells or short labels the trick does not fail.

How to spot it. The sheet ends up turned relative to the edge of the table. You lose the line when moving to the next. And one very reliable sign: the reader cannot say where they are on the page without searching again, because their hand has not built any map of the space.

What to do instead. Fix the paper (with the other hand, with clips, with anything) and move the hand. This is teaching judgment, not an experimental result, so we say it as such; but the reasoning is concrete. Braille reading involves two different movements: the forward sweep and the return to the next line. In a study of 32 blind readers, they were linked to different errors: the return sweep to skipping whole words and phrases, the forward sweep to letter and syllable errors (Papadimitriou and Argyropoulos, 2017). A reader who moves the paper cannot automate either, because the reference point shifts underneath on every pass.

5. Reading with one hand and never learning the hand-over

What it is. Using one index finger for everything: sweeping the line, going back to the start of the next and starting again. The other hand rests.

Why it happens. Because at first coordinating one finger is hard enough. The second hand looks like a luxury for advanced readers, and it gets postponed. The problem is that it gets postponed forever: the more the one-hand pattern becomes automatic, the harder it is to replace.

How to spot it. Time the end of the line. If there is a clear pause (the finger finishes, lifts, travels back, feels for the left margin, finds the line), you pay that toll on every line. On a thirty-line page, thirty times.

What to do instead. The hand-over, which the EBU calls the scissor technique: the left hand starts the next line while the right finishes the current one. In Nonaka and colleagues' study, the most experienced readers used that pattern, while slow readers read with one hand or moved both together without varying their speed. The same recommendations point to something almost nobody teaches: fluent readers use not only the index fingers but four or even six fingers, the extra ones previewing what comes next and keeping track of the page. Start by resting the middle finger next to the index and letting it follow.

Method mistakes: what goes wrong in the plan

6. Learning the alphabet in alphabetical order

What it is. Studying a, b, c, d… as a list of 26 unrelated symbols.

Why it happens. Because that is how we learned the print alphabet, where the letters really are unrelated: the shape of m does not follow from c. In braille it does, and that fact changes the whole study plan.

How to spot it. You flow from a to j and stop dead from k on. If, to recall a letter, you have to recite the alphabet from the start until you reach it, you are memorizing a list.

What to do instead. The three series, checked against the braille tables:

And w, ⠺, always presented as the loose exception because Louis Braille's French barely used it, has a handle: w is j plus dot 6. Compare j ⠚, dots 2, 4 and 5, with w, dots 2, 4, 5 and 6. How to turn this into a study method is in how to memorize the braille alphabet.

7. Not knowing which grade your material is in

What it is. Struggling with a text without realizing it is written in grade 2, contracted braille, where one cell can stand for a whole word or a group of letters.

Why it happens. Because almost all English braille in the wild (books, signs, menus) is grade 2, while most beginners start with the grade 1 alphabet. The two look identical until a cell turns up that is in no alphabet chart.

How to spot it. Cells you cannot decode and that are not in any letter chart. A short word taking fewer cells than it has letters. A single letter standing alone, like ⠽, where you expected a whole word. You are not short of memory: you are reading another level of the system.

What to do instead. Check the grade of the material before fighting it. If you are learning on your own, get the letters automatic in grade 1 first, then add contractions in small groups; many children in English-speaking schools learn contractions from the start, which is a different, longer path built for them. The full list is in braille contractions, and the braille translator shows any text in both grades.

8. Practising for two hours on Sunday and nothing all week

What it is. Packing study into long, spaced sessions instead of short, frequent ones.

Why it happens. Because that is how exams are passed, and because a long session feels productive. The feeling of progress within a session is real; what does not survive is the result a week later.

How to spot it. One unmistakable sign: every session starts by catching up on the last. If the first twenty minutes of each session go on recovering ground, the gap between sessions is too long for what you are learning.

What to do instead. Short, daily sessions, ideally at the same time. And a nuance that matters more in braille than elsewhere: the slow part of braille is a motor and perceptual skill, not a list of facts, and for skills, very long gaps reduce the benefit of spacing. The evidence, with the meta-analyses and effect sizes, is in how long it takes to learn braille. In practice: better not to skip a day than to do a long session to make up for it.

Code mistakes: what goes wrong on the page

9. Mixing up mirror letters

What it is. Reading f where it says d, or i where it says e. It is not a memory slip: those cells are exact mirror images of each other across the vertical axis.

Why it happens. Because of a property of perception called mirror invariance: by default we treat a shape and its reflection as the same object, because in nature a tiger seen from the left or the right is still the same tiger. Learning to read forces us to break that invariance (it is why children confuse b and d in print), and breaking it takes effort and is learned. Adélaïde de Heering and Régine Kolinsky showed that braille readers break it for braille cells too, and that the effect grows with experience of that material (Cognition 189, 55–59, 2019). In other words: you are not clumsy; your brain arrives treating those two cells as one, and experience is what separates them.

Which ones. This table is not copied from anywhere. We computed it by mirroring every letter (dot 1 ↔ 4, 2 ↔ 5, 3 ↔ 6) using the braille tables. There are four pairs, no more:

Three letters are immune by construction, because they are identical to their own reflection: c ⠉, g ⠛, x ⠭. For 15 of the 26 letters, the mirror image is not a letter at all.

How to spot it. By the shape of your mistakes, not their number. A memory mistake is random: today you miss p, tomorrow s. A mirror mistake is systematic and clusters: you miss d and f, both ways, again and again, while the rest of the alphabet is fine. The practical test: for a week, note which letter you read and which it was. If your mistakes fall in the pairs in that table, you do not need to review the whole alphabet; you need to work on four pairs.

What to do instead. One rule, which we checked against all four pairs: look at which column is heavier. Count the dots in the left column and in the right; the side with more dots identifies the letter. If they tie, the side whose top dot is higher wins. Mirroring swaps the columns, so this comparison always separates them. Applied: d is heavier on the right and f on the left; e and i tie at one dot each, and e wins on the left because its dot is in the top row, while in i the top dot is on the right.

Two honest caveats about this rule. It is a rule for analysis, not for fluent reading: it helps you get unstuck on a pair while learning it, not to read with. And a fluent reader does not count dots.

And now the part that changes the advice depending on how you read. The pairs above are a geometric fact, and they are the real problem for people who read braille by sight. But the finger is confused by something else. A research team measured real tactile confusion: 24 fluent blind readers, on a motorized braille display, telling pairs of letters apart. The most confusable groups were d–f–h, u–v–x, m–n–p, n–y–o–z and s–t–r–w–p–q, and what groups them is not reflection but the particular arrangement of dots, especially in positions 2, 3 and 5 (Özkan, Baciero, Perea and Gómez, 2026). Checked against the braille tables, most pairs within those groups differ in a single dot (m/n in dot 5, o/z in dot 6, p/q in dot 5, u/v in dot 2), and only two of the pairs mentioned are also mirror images.

The practical conclusion is awkward and useful: if you learn by sight, your enemy is the four mirror pairs; if you learn by touch, your enemy is the pairs that differ by one dot. They are different lists, trained with different exercises.

10. Forgetting the numeric and capital indicators

What it is. Writing "15" as the letters a and e, or a name without marking the capital. It is the most frequent writing mistake, and the one that makes text unreadable.

Why it happens. Because they are two signs that do not exist in print. There is nothing in the print alphabet they translate, so no existing habit brings them to mind. They are forgotten the way anything you have never done is forgotten.

Why it is not optional. The first ten letters share their cells with the ten digits, with no exceptions: a=1, b=2, c=3, d=4, e=5, f=6, g=7, h=8, i=9, j=0. They are not similar: they are the same cell. The only thing that tells them apart is the numeric indicator, ⠼, dots 3, 4, 5 and 6, in front. The capital indicator, ⠠, is dot 6; two of them at the start of a word make the whole word capitals.

How to spot it. The symptom is that you cannot reread what you wrote without remembering what it said. If, rereading your own label, you hesitate between "5" and "e", the indicator is missing. A second symptom: you write lone numbers correctly and forget the indicator inside a sentence, because it is remembered when the number is the point and lost when it is in passing.

What to do instead. Practise with texts that mix: addresses, dates, shopping lists, prices, not isolated numbers. Write your own name, which needs a capital indicator from the first cell. And check: type anything into our translator and see where the indicators appear.

One detail the translator will show you: when a letter from a to j comes right after a number, the number would swallow it. 5a written as ⠼⠑⠁ would read as 51. UEB solves it with the grade 1 indicator, ⠰, which ends the number: 5a is ⠼⠑⠰⠁. Letters from k on need no indicator (5k is ⠼⠑⠅), and a space ends the number cleanly. The number rules are in braille numbers.

Mistakes that depend on who you are

11. If you can see: looking

What it is. Learning the code with your eyes and believing that trains touch reading. It does not, at all.

Why it happens. Because looking is faster and the brain picks the faster route. It is not a lack of discipline: as long as sight is available, it solves the task before the finger does, and the finger learns nothing from a task already solved.

How to spot it. A thirty-second test. Recognize ten letters by sight, then the same ten without looking. If you get nearly all of them by sight and collapse without it, you have learned the code and not started on touch. They are two skills and one does not bring the other.

What to do instead. It depends on your goal, and here is the good news: for most reasons a sighted person learns braille (writing notes, labeling the house, checking a child's homework) you do not need to read by touch. If you do, it is trained without looking from day one, on real raised dots, starting with telling one dot from two before letters. It is all in learning braille as a sighted person.

12. If you support someone: correcting every slip while they read

What it is. Interrupting the reader at every mistake, with the best intentions in the world.

Why it happens. Because the supporter almost always learned the code by sight in two weeks, so they catch mistakes instantly and from a comfortable position. Pointing them out feels like help; in fact it is an advantage of channel dressed up as correction.

How to spot it. When the reader stops on their own before you correct them, anticipating the interruption. At that point they are no longer reading: they are waiting to be graded.

What to do instead. Note mistakes and talk about them later, not over the reading. The reason is mistake number 3 seen from outside: what has to become automatic is the finger's continuous movement, and an interruption breaks it just as scrubbing does. Changing roles helps more than any technique: from corrector to correspondent. Write to them, and have them write back.

What we could not confirm

As in the rest of this blog, here is what the material we could read does not support.

And one worth saying separately: we do not know which mistake on this list is the most costly. We grouped them by type, not severity, because we have no data to rank them by impact.

How to fix them, in order

1. Start with the hand. Minimal pressure, finger flat on the pad, continuous movement. The first three mistakes are the quickest to fix and the ones that block everything else.
2. Fix the paper and move the hand. And do not leave the second hand idle for more than a few weeks.
3. Check your plan. If you are memorizing in alphabetical order, stop and learn the three series. It saves weeks.
4. Check which grade your material is in before fighting cells that are not in the alphabet.
5. Spread your practice. Fifteen minutes a day, not two hours on Sunday.
6. Sort your mistakes before reviewing. If they cluster in pairs, it is geometry, not memory.
7. Write something real with numbers and capitals, and reread it the next day.
8. If you support someone, write to them instead of correcting them.

You can check any transcription in the translator and keep the braille alphabet chart at hand. What no screen will do is train your touch: that needs real raised dots.

Frequently asked questions

What is the most common mistake when learning braille?

Pressing the finger into the dots. It is the most frequent and the most counterproductive, because extra force adds no information (quadrupling it does not improve spatial touch tasks, according to Gibson and Craig, 2006) and in exchange it tires the hand and slows the finger, which is what actually matters.

Which braille letters are most often confused?

It depends on whether you read with your eyes or your fingers. By sight, the four pairs that are exact mirror images: d/f, e/i, h/j and r/w. By touch, the confusing ones differ by a single dot: m/n, o/z, p/q, u/v.

Do you read braille with the tip of the finger?

No: with the pad, the soft part just behind the tip. The European Blind Union's 2025 fast reading recommendations point to the pads of the index and middle fingers, not the fingertips, because they give more contact area and spread the effort.

Should you read braille with both hands?

Eventually, yes. Two hands allow the hand-over (one starts the next line while the other finishes the current one), which removes the pause at the end of every line. You can start with one, but do not leave the hand-over for "later": the more automatic the one-hand pattern, the harder it is to change.

Why do I mix up d and f in braille?

Because they are the same cell mirrored: d is dots 1, 4 and 5, and f is dots 1, 2 and 4. Our perception treats a shape and its reflection as the same object by default, and telling them apart is learned. The rule that separates them: d has two dots in the right column and one in the left; f the reverse.

Is the numeric indicator really necessary?

Yes. The first ten letters use exactly the same cells as the digits 1 to 0, so without the numeric indicator in front (dots 3, 4, 5 and 6) the text is ambiguous and cannot be reread.

I have been at it for weeks and I am still slow. Am I doing something wrong?

Probably not. Recognizing the code and reading it with your fingers are two skills on very different scales: the first takes days or weeks, the second months or years. Being slow by touch after a few weeks is normal, not a sign of a mistake. The timelines, with the studies behind them, are in how long it takes to learn braille.