012 min
What Bionic Reading actually does
Bionic Reading changes only how a text looks. The words, their order and the line breaks stay the same. A converter takes plain text and sets the opening letters of each word in bold. The reader's eye is meant to land on those bold letters and let the rest of the word be filled in without a direct look.
The method's own description is that the first few letters of words are emphasised by using bold type, to create "artificial fixation points." A fixation is the short pause when the eye holds still and takes in letters. The idea is that a text with ready-made fixation points needs fewer pauses of the eye's own choosing.
The converter exposes three settings, which also show how the method is built:
| Setting | What it changes |
|---|---|
| Number of bold letters | How many letters at the start of each word are bold |
| Frequency of fixation points | How often a word gets bold letters at all, from every word to every few words |
| Opacity of the bold letters | How strongly the bold letters stand out from the rest of the word |
The same method is also sold as a font feature. The Bionic Reading Font Converter adds the bolding to a font as an OpenType Feature, which is a built-in option a font can switch on in software that supports it. That means a designer can apply the effect to a typeface they already use.
Fixation point: the spot on a line of text where the eye stops and takes in letters before the next jump.
The method sits between typography and reading research. It borrows the look of a bold lead-in, which designers use on purpose to mark where a reader should start. But it bolds every word, or a regular share of them, so the bold marks no one idea. It marks a position inside a word. That difference is why the method behaves very differently from the emphasis tools covered in visual hierarchy.
022 min
How the eye moves when reading
To judge any claim about faster reading, you need to know what the eyes do on a line of text. Reading is not a smooth glide. The eye jumps from place to place, and it takes in text only while it is still.
The jumps are called saccades. The still periods between them are fixations. A recent eye-tracking paper summarises typical values for adults reading English: the length of a typical fixation is 200 to 250 milliseconds, and a saccade generally spans 7 to 9 letters. So a line of text is read in a series of stops, each about a fifth of a second long, spaced a few letters apart.
This gives Bionic Reading a clear thing to change. If bold letters pull the eye to the start of each word, the reader could make fewer stops per line, or shorter ones, or longer jumps. Any of those would raise reading speed. The claim is therefore testable. You record where the eyes land and how long they stay, with and without the bold letters, and compare.
The claim also has a built-in difficulty. A fixation does not only gather letters. During the pause the brain also recognises the word and starts to work out its meaning. Putting a bold marker on the first letters could change where the eye lands without changing how long recognition takes. Reading research has a name for the best landing place in a word, the optimal viewing position, which sits near the middle of the word rather than at its first letters. A marker at the first letters pulls the eye away from that position.
032 min
Where Bionic Reading comes from
Renato Casutt gave the technique its name, Bionic Reading, and it reached the design press in May 2022. Creative Bloq published a report on 23 May 2022 under the question of whether the method would turn readers into speed readers.
Casutt's account of how he found it is a personal one. He says he discovered the phenomenon by accident while working on a book in a foreign language, when he found he could relate the foreign words to words in his own language when parts were sectioned off. This is a description of a single designer's experience. It is not a study, and it does not show that the effect holds for other readers or for text in a language they already know well.
The method spread fast because it is easy to try. A converter turns any text into the bolded version in a moment, and an example paragraph is easy to share on social media. Casutt has also made the tool's API available, which is a programming interface that lets designers and developers add the conversion to their own apps. By then the method had a commercial form: apps, browser extensions and licensed fonts sold under the Bionic Reading name.
There is no peer-reviewed research behind the launch. The first large public test came from a company that makes reading software, not from a university, and the first published experiments appeared later (see the sections on evidence and history below).
043 min
What the tests of Bionic Reading found
Three separate tests, run in different ways, point to the same result. None of them found that Bionic Reading speeds up reading.
The Readwise experiment, 2022
In June 2022 the company Readwise posted an online experiment to Hacker News, Reddit and Twitter to test the claim that Bionic Reading lets people read faster without any loss of comprehension. Readers read two articles, each split into halves, with one half in Bionic format and one in plain format, and answered three questions after each half. Daniel Doyon published the results on 19 July 2022.
The team found no evidence that Bionic Reading has any positive effect on reading speed. In the group analysed, readers were on average 2.6 words per minute slower with the bold format than without. That is a difference of less than one percent. Comprehension was also virtually identical across the two formats.
The design has limits. The readers were volunteers who found the test through technology sites, and they read online at home with no control over their screen or setting. It is a large, useful sample, not a laboratory trial.
The reading-time study, 2024
Joshua Snell published a four-page paper in the journal Acta Psychologica with a direct title: "No, Bionic Reading does not work." He noted that empirical tests of the claim were lacking, and that theoretically one might expect a cost rather than a benefit. His reasoning is the viewing-position point above: bold first letters draw the eye to the part of the word where recognition is hardest.
Participants read 100 paragraphs, some in Bionic format and some in normal font. Statistical analyses revealed no significant difference in reading times between Bionic and normal reading. Snell concluded that the method does not facilitate reading. He did not find the cost he had predicted either.
The eye-tracking study, 2025
T. R. Beelders of the University of the Free State recorded eye movements while participants read one passage in standard font and one in Bionic font. The paper appeared in the Journal of Eye Movement Research in October 2025. It reported that fixation durations, number of fixations and reading speeds were not significantly different between the two formats.
This study tests the mechanism itself, not only the speed. If the bold letters worked as fixation points, fixations would cluster on them. They did not: fixations were spread throughout the word and not only on leading characters, so readers were not able to complete a word from its bold start alone. The bold font also did not help with low-frequency or unfamiliar words, which are the words where a hint should matter most.
Claimed chain: bold word starts, then fewer or shorter fixations, then faster reading. Measured result: the first link does not occur, so the chain stops there.
052 min
How Bionic Reading shows up in real products
Bionic Reading appears in three real forms today, and each one raises a different design decision.
Where it is offered
The company behind the method sells apps and extensions under the Bionic Reading name: a web app, apps for iOS, Android, macOS and Windows, a Chrome extension and a Word add-in. In each, the reader converts a text and adjusts how much of each word is bold. The choice sits with the reader, who can switch it off. This is the form with the least risk, because a reader who finds it unhelpful loses nothing.
It is also sold as a font feature, through the converter described earlier, and as a licence for publishers. Here the choice moves to the person who sets the type. A publisher who applies it to a whole book makes the decision for every reader.
A case where the idea fits
The strongest version of the idea is not the full method but a smaller practice that designers already know: a bold lead-in at the start of each list item or short paragraph. In a long set of release notes, a bold first phrase on each entry lets a reader decide in a glance whether the entry matters. The bold part carries meaning, and it appears once per entry. This works through visual hierarchy, not through eye guidance inside words. It matches how people scan web pages, a behaviour documented in the F-pattern research.
A case where it backfires
Apply Bionic format to a full article, and every word has a bold start. A reader scanning the page now sees a uniform texture, and no word stands out from the others. The bold letters mark position, not importance, so they cannot rank anything. This is a reasoned prediction from how emphasis works, not a measured result. What was measured is that reading is no faster. Say a staff designer in a design review adds Bionic format to the body text of a documentation site. Readers get no speed benefit, and the scanning cues that headings and bold key terms gave them are now weaker because the page is full of bold marks.
062 min
Applying Bionic Reading
Treat Bionic Reading as a reader preference, not a performance feature. These moves follow from the evidence above.
Make it optional.
Offer it as a switch that the reader controls and that the reader can turn off at any time. Do not apply it by default to all text.
Do not promise speed.
The tests found no gain in reading speed or comprehension, so any copy that says "read faster" is a claim the evidence does not support. Describe it as a display style that some readers find easier to look at.
Keep the text's real structure.
Headings, lists and bold key terms do the work of marking importance. Bionic format should never replace them.
Spend the effort on settings that have a record.
Line length, type size, line spacing and contrast are documented factors in readability. A change to any of these is more likely to help than a change to the bolding.
Test with your own readers.
Give a sample of your users a task with questions about the text, with and without the format, and compare time and answers. A result from your own audience outweighs a general claim.
When following this advice backfires
Some readers say the format helps them focus, and a reader's report of comfort is real information even when the speed numbers are flat. Removing the option from people who like it is a cost. The point is to offer the format as a choice. It is not to forbid it.
Check one practical detail before shipping: bold letters inside a word are usually produced by splitting the word into separate pieces of markup. Test the page with a screen reader to confirm each word is still announced as a whole word, which matters for accessibility.
071 min
How to spot Bionic Reading going wrong
Because the format has no measured speed benefit, the useful signals are about use and comprehension, not about speed alone. Each of these can be checked in analytics, a usability session or a support queue.
- The off-switch rate. If the format is a setting, count how many readers who turn it on later turn it off.
- Time per article against questions answered. Compare average time on a page and the share of correct answers on an end-of-article question for readers with and without the format. Readwise used three questions per article half, which is a simple model.
- Scroll-back behaviour. In session recordings, a reader who scrolls up to re-read a line is having trouble. Check whether it happens more often on the formatted version.
- Support tickets that mention the page looking cluttered or hard to scan. These point to the uniform bold texture described earlier.
- Complaints from readers who use a screen reader. These point to markup that splits words.
081 min
Bionic Reading vs. nearby concepts
Bionic Reading is often grouped with several nearby ideas. They differ on the axis of what the emphasis marks.
| Concept | What the emphasis marks | Evidence of faster reading |
|---|---|---|
| This pageBionic Reading | ||
| Not in the library yetBold lead-ins and keywords | ||
| DesignVisual hierarchy | ||
| DesignF-pattern | ||
| Not in the library yetSpeed-reading courses |
The closest neighbour is the ordinary bold keyword. Both use bold type. A bold keyword says "this idea matters". A bold word start says "look here, at this part of this word". The first changes what a reader notices on the page. The second tries to change how the eye moves along a line, and the studies above say it does not.
091 min
Where the evidence on Bionic Reading is contested
The main dispute is between the designer's account and the three tests. The designer's account is that the format guides the eye and the brain completes each word. The tests say the eyes do not behave that way. The strongest version of the designer's side is a fair one: the tests measure averages over short passages, and a format could suit a smaller group of readers or work after long practice.
The evidence leaves room for that. Beelders states plainly that, in the short term, Bionic font does not affect reading, and that further investigation is needed to determine whether a long-term intervention with Bionic font is more meaningful than standard interventions. None of the three tests followed readers over weeks or months.
Snell's paper also stops at one measure. He measured reading time. Reading time does not capture whether a reader feels less tired or more willing to keep going, which are the claims many supporters make. Those claims are not refuted by the speed result, and they have not been tested in the studies cited here.
For a practitioner, the position while this is unsettled is straightforward. The measured claims, speed and comprehension, have been tested and not supported. The unmeasured claims, comfort and focus, belong to individual readers. Offer the format as a choice, and do not use the unmeasured claims in marketing copy.
101 min
How Bionic Reading changed since 2022
Four dated events show how the discussion changed.
- May 2022. The method reached the design press with claims about faster reading. The only support was the designer's own account.
- June and July 2022. Readwise ran its public test and published the results on 19 July 2022. It was the first large measurement, and it found no speed benefit.
- May 2024. Snell's paper went online. Its abstract says empirical tests of the claim were lacking before his, which also tells us the Readwise test was not a published scientific study. The paper appeared in the July 2024 issue of Acta Psychologica.
- October 2025. Beelders published the first eye-tracking test, which also examined the claim that the bold letters work as fixation points.
The meaning of the term changed along the way. In 2022 "Bionic Reading" named a reading method with a claimed effect on the reader. After the tests it is better described as a typographic style: a way to set text, with the claimed speed effect unsupported. Marketing for the product has not always followed that change, so the term can still refer to the claim in some places and to the style in others.
?6 questions
Questions people ask
What is Bionic Reading?
Does Bionic Reading make you read faster?
Does Bionic Reading improve comprehension?
How do I apply Bionic Reading to a design?
Is Bionic Reading the same as bold keywords?
Are there exceptions where Bionic Reading helps?
▶1 video
Watch
Understanding Bionic Reading
§5 sources
Sources
Show all 5 sourcesHide sources
Joshua Snell, "No, Bionic Reading does not work," Acta Psychologica, volume 247, article 104304, July 2024. Open the record at and the paper at .
T. R. Beelders, "Guiding the Gaze: How Bionic Reading Influences Eye Movements," Journal of Eye Movement Research, volume 18, issue 5, article 49, October 2025.
Daniel Doyon, "Does Bionic Reading actually work? We timed over 2,000 readers and the results might surprise you," Readwise blog, 19 July 2022.
Joe Foley, "Will 'Bionic Reading' turn us all into speed readers?" Creative Bloq, 23 May 2022. A news report on the launch, which quotes the designer's own account.
Bionic Reading, the company's own product site, for the apps, fonts and converter.





