Blog · Typography & Layout
Two-column layouts and dyslexia
If reading a two-column academic PDF or a newspaper page feels significantly harder than reading a normal webpage, you're not imagining it. Multi-column layouts impose a specific set of reading demands that interact badly with how dyslexic readers process text. This is not a vague complaint — there's a typography-research explanation, and there are practical ways to work around it.
What makes two-column text harder to track
Reading is not a smooth left-to-right scan. Your eyes make a series of short, rapid jumps called saccades, each one landing at a new fixation point roughly every 200-250 milliseconds. At the end of a line, the eye has to make a large return sweep to find the start of the next line — and that sweep is one of the places where dyslexic readers are most likely to land in the wrong place.
Two-column layouts make this worse in two ways that compound each other.
First, the columns are narrow. A typical academic PDF column runs around 70-80 millimetres wide, which at normal reading distance translates to roughly 50-60 characters per line. That sounds like a comfortable measure — typographers often cite 60-75 characters as ideal for single-column body text — but the ideal range assumes the reader's eye can return smoothly to the left margin after each line. In a two-column layout, "the left margin" is not the left edge of the page; it's the left edge of the column, which changes position depending on which column you're in. That ambiguity adds a small but consistent cognitive load to every line return, across hundreds of lines in a long document.
Second, the gutter between the columns is a visual trap. When the eye finishes a line in the left column, it has to make a return sweep that stops at the column's left edge — but the gutter and the start of the right column are right there in peripheral vision. Many dyslexic readers report that their eye slides across the gutter into the right column, particularly on long pages where the columns are in the same vertical zone. This is not a perception error so much as a tracking error: the eye is very good at jumping to salient features, and a column of text immediately to the right of your reading position is salient.
The left column ends here, and the eye must return to the left edge of this column — not the page. After several hundred such returns, any hesitation or misalignment compounds into skipped lines and lost place.
The narrow line length also means more line returns per paragraph, which multiplies the number of times the eye must navigate back to the correct margin.
Meanwhile the right column sits in peripheral vision throughout the reading of the left column. Its text exerts a pull on the eye — particularly at the end of lines, when the brain is preparing to make a large return sweep.
The result is a layout that requires constant spatial recalibration, which is effortful for all readers and genuinely disruptive for readers whose tracking is already less automatic.
The research on line length and reading
The connection between line length and reading ease is one of the most replicated findings in typography research. Work by Kevin Larson at Microsoft Typography and by Keith Rayner (the leading researcher on eye movements in reading) consistently shows that very short lines — under 45 characters — increase the number of regressions, the backward eye movements readers make to re-parse text they didn't catch on the first pass. Short lines also interrupt phrase grouping: English syntax naturally chunks into phrases of five to eight words, and a line that breaks mid-phrase forces the reader to hold a partial phrase in working memory across the line boundary.
This matters especially for dyslexic readers because working memory load is already elevated. Research consistently finds that dyslexia involves not just phonological processing difficulties but also reduced capacity to hold and manipulate verbal information in short-term memory. A layout that adds extra working memory load — mid-phrase line breaks, ambiguous return points — is one that compounds an existing difficulty, not one that exposes a new one.
Our article on line length and dyslexia covers the optimal character-per-line range in more detail. The short version: for dyslexic readers, a wider single column (65-80 characters per line) tends to outperform a narrower two-column arrangement even when the total text area is identical.
Newspapers vs academic PDFs vs magazine websites
Not all two-column layouts are the same, and the reading experience varies significantly by context.
Newspaper columns are the most demanding. They are extremely narrow — broadsheet columns often run to 35-45 characters per line — and they are set with full justification, which introduces irregular spaces between words that further disrupt the rhythm of saccades. Newspaper typography was designed for ink on paper, where the physical page provides a stable spatial reference. On screen, scrolling changes the relationship between the column edges and the screen edges, removing that reference.
Academic PDFs are the most commonly encountered two-column format for adult readers. Most STEM journals (Nature, IEEE, ACM, and their kin) use two-column layouts in their PDF editions to match the print journal format, even though the PDF is almost always read on screen. The columns are somewhat wider than newspaper columns, but the font size is typically smaller (8-10pt is common in print PDFs), which reduces the effective measure further. Academic PDFs also tend to insert tables, figures, and equations mid-column, which interrupt the spatial flow the reader's eye has established.
Magazine-style websites — the kind that use CSS multi-column layouts or masonry grids — present a different problem. The column breaks are responsive, so where one column ends and another begins changes with window width. A reader who has found a comfortable reading position may inadvertently trigger a reflow by resizing the window, losing their place entirely.
Eye tracking and losing your place
The specific experience of "losing your place" in a two-column document is well documented in reading research on dyslexia. Studies using eye-tracking equipment have shown that dyslexic readers make more and longer regressions than typical readers, and that these regressions cluster at line returns. The beginning of a new line is consistently harder to find than the rest of the line, because the return sweep has to land accurately on a target — the first word of the next line — without visual guidance from adjacent text.
In a single-column document, the left margin is fixed and prominent. The eye develops a motor routine — return to the left edge — that becomes increasingly automatic over the course of reading a page. In a two-column document, the "left edge" changes from the page margin to the column gutter, and then back, and the routine never fully solidifies. Our piece on why dyslexic readers lose their place goes into the eye-tracking research in more detail.
What actually helps
The options depend on the format.
For academic PDFs
The most effective intervention is reflow: converting the PDF into a single-column format that can be read like a normal webpage. Several tools do this:
Adobe Acrobat's Reflow view (View > Zoom > Reflow, or Ctrl+4 on Windows) attempts to extract the text and reformat it into a single flowing column. Quality varies by PDF — documents that were scanned from print rather than generated digitally may produce garbled output. But for a well-formed PDF, Reflow is the quickest fix available without installing anything extra.
Unpaywall and Semantic Scholar both offer HTML versions of many academic papers. The HTML version is always single-column, and it renders in your browser where you can apply LexiFont or any other reading tool. Before wrestling with a two-column PDF, check whether an HTML version of the paper exists — it often does for open-access articles. Our guide to reading research papers with dyslexia covers this workflow in detail.
Browser PDF viewers with text selection allow you to copy the full text and paste it into a text editor or a minimal-layout page. This is a slower process but gives you full control over font and line length. Pasting into a Google Doc, then applying a dyslexia-friendly font, works reasonably well for papers you need to read carefully rather than scan.
For font control on the PDF itself — if you need to stay in the PDF rather than converting it — see our guide to dyslexia-friendly PDF reading in Chrome.
For newspapers and news sites
Most news sites serve their content in a single-column article view below the headline, even if the homepage uses a multi-column grid. The challenge is usually navigating from the homepage to the article, not reading the article itself. Once you're on the article page, reader mode (built into Firefox, Safari, and Edge, and available as an extension for Chrome) strips the surrounding columns and serves the body text in a single clean flow.
If you're reading a newspaper in PDF form — the kind that replicates the print layout exactly, as many regional papers do — the Reflow approach above applies. Alternatively, most newspapers also publish a website version of each story, which you can find by searching the headline. The website version is almost always single-column and more amenable to font overrides.
The comparison of reader mode versus reading extensions covers the trade-offs in more detail, including which tools work best when the original layout is multi-column.
For magazine-style websites
CSS multi-column layouts are common on feature-magazine sites and some blogs that use a "newspaper-style" template. Reader mode handles most of these well — it extracts the article body and ignores the surrounding column structure. If a site's reader mode is inconsistent, narrowing the browser window often forces the layout into a single column, because responsive breakpoints usually collapse to single-column below about 700px of width.
Changing the font
Changing the font does not fix a two-column layout, but it reduces the difficulty of navigating it. Dyslexia-friendly fonts with strong letter differentiation — OpenDyslexic, Lexend, or Atkinson Hyperlegible — reduce the letter-confusion load, which means more of the reader's attention is available for the tracking problem that the column layout creates. It's a partial fix, not a full one, but in practice readers often find that the combination of a better font and a slightly larger text size makes a two-column layout manageable even when full reflow is not available.
LexiFont applies font overrides to HTML pages in the browser, including HTML versions of academic papers opened via Unpaywall. It won't apply to the inside of a PDF rendered by a PDF viewer, but it covers any text the browser renders directly. The Pro tier gives access to the full font library including Atkinson Hyperlegible and Lexend in a single click.
A note on line spacing in narrow columns
One thing that consistently helps in narrow-column documents, even when you can't change the layout, is increased line spacing. Research from the BDA and from independent typography studies suggests that line spacing of 1.5 to 2.0 times the font size helps dyslexic readers keep their place on the line — and the benefit is larger in narrow-column text, because the larger vertical gap between lines makes it easier for the eye to distinguish the line it just left from the one it's returning to. Our article on line and letter spacing for dyslexia covers the exact values that the research supports.
In a PDF viewer you can sometimes force line spacing through accessibility settings or through the Reflow view, though the results are inconsistent. In an HTML version of a document, LexiFont's spacing controls let you dial in the line height that works for you.
The broader point: layout is not neutral
Two-column layouts are not bad design in some absolute sense. They exist because they let publishers fit more text on a page, and they work reasonably well for readers whose eye-tracking is automatic and whose working memory load during reading is low. The problem is that those conditions don't hold universally, and the readers most likely to find narrow-column layouts difficult are also the ones most likely to be reading complex material — academic papers, detailed news reports — that they genuinely need to understand.
The practical implication is worth stating plainly: if you find two-column academic PDFs genuinely hard to read and you have dyslexia, that is not a reading comprehension problem. The layout is the problem. Converting to a single-column format is not working around your difficulty; it's removing an unnecessary obstacle.
Quick checklist for two-column documents:
1. Check whether an HTML version exists (Unpaywall button, Semantic Scholar, publisher website).
2. If reading the PDF directly, try Reflow view (Ctrl+4 in Adobe Acrobat).
3. If the layout remains two-column, increase zoom to at least 150% — this narrows the visible page and often forces you to scroll rather than track across columns, which is easier.
4. Use a reading ruler or focus mode to isolate one line at a time.
5. Apply a dyslexia-friendly font to any HTML text via LexiFont to reduce letter-confusion load before tackling the layout challenge.