Bionic Reading Converter

What is bionic reading?

Bionic reading is a typographic treatment that bolds the first portion of each word — usually 40 to 60 percent of the letters — so that the bold stem of each word stands out and the unbolded tail recedes. The idea is that the eye’s saccade (the quick jump from fixation point to fixation point as you read) lands more reliably on the bold stem, and the brain fills in the rest of the word from context and partial shape.

The technique was popularised by Renato Casutt, who registered the Bionic Reading trademark and built a paid SaaS around it. In mid-2022 a series of viral demos on Twitter / X and TikTok claimed huge reading-speed improvements; the demos racked up tens of millions of views and several major publications wrote it up. Subsequent academic scrutiny was less flattering — see the FAQ above. The technique survives as a preference: a meaningful fraction of readers find bionic-formatted text more comfortable on long pages, even if the speed boost is not statistically robust.

How it works

The algorithm is short enough to describe in one paragraph. The input text is split into tokens on whitespace boundaries. For each token, the tool computes N = ceil(length × fixation_ratio) — where fixation_ratio is the slider value, 0.20 to 0.70, defaulting to 0.50. The first N characters are wrapped in <strong> for the HTML preview and Markdown copy (where **stem**tail is the syntactic equivalent). Punctuation attached to a word is treated as part of the tail — Hello, becomes **Hel**lo, — so the comma sits in the unbolded portion where it belongs visually.

That’s the whole engine. There is no machine-learning model, no syllable parser, no part-of-speech tagger. Bionic Reading AG’s commercial product uses a slightly more sophisticated word-length table (short common words get lighter treatment), and this tool mirrors that behaviour for words of three letters or fewer — the, and, of, it get only the first character bolded regardless of the slider, because bolding “th” in “the” produces noticeably uglier text than bolding “t”.

Common use cases

• Skimming long-form articles. Paste a 3,000-word essay, render it bionic, and read the bolded result. Most readers report a small but real comfort gain on text they were going to skim anyway.
• Dyslexia-friendly reading aids. Some readers with dyslexia report that the bolded stems help them anchor on each word and reduce the “swimming” sensation of long blocks of uniform text. This is anecdotal, not clinical, but the technique is harmless and free to try.
• Attention-deficit reading aids. Similarly, some readers with ADHD report that the visual rhythm of bold/non-bold helps them resist drifting off mid-paragraph. Treat as a preference, not a treatment.
• Note rewriting for review. Convert your own meeting notes or study material to bionic format before re-reading; the visual change can refresh attention on text you’ve already seen.

How to use this Bionic Reading converter

1. Paste your text into the Input textarea, or accept the built-in sample paragraph to see what the output looks like.
2. Adjust the Fixation % slider. 50% is the default and matches the most common published examples; 30–40% is a lighter touch; 60–70% is heavy, useful for very dense material.
3. The Preview updates live with the bolded HTML. Read it directly on the page if that’s all you need.
4. Use Copy as Markdown to copy the result with **bold** markers — paste into Obsidian, Notion, GitHub issues, or any Markdown-capable destination.
5. Use Download as HTML to save a standalone .html file with the bolded text and a minimal stylesheet — useful for sharing, archiving, or sending to a Kindle / e-reader that accepts HTML.

Quality tips

The technique works best on flowing prose. Code, structured data, tables, and lists of nouns produce visually noisy results — the algorithm can’t tell that getUserById is one identifier, so it bolds the first half and the result looks broken. Strip code blocks and tables from your input before converting, or accept that they will look odd.

Short, choppy text (chat logs, one-line bullet points) doesn’t benefit much either — the eye doesn’t need help finding fixation points in short tokens. Aim for paragraphs of 50 words or more.

Privacy

Everything happens in your browser. The tokeniser, the bolding loop, and the HTML / Markdown export are all JavaScript on your device. Pasted text — which might be a personal journal entry, unpublished writing, or someone else’s confidential document — never travels across the network. Confirm in the browser Network panel, or just disconnect from Wi-Fi after the page loads; the converter keeps working.

Compatibility notes

The HTML output uses only <strong> tags and standard inline styling — it renders identically in every modern browser, in Apple Mail, Gmail, Outlook, Notion, Obsidian, and any other tool that displays HTML or Markdown. The Markdown output uses the universal **stem** syntax accepted by CommonMark, GitHub Flavoured Markdown, and every Markdown renderer in common use. There are no external font requirements; the bold weight is the bold of whatever font surrounds the pasted text.