Frequently asked questions

What is Fovea?

Fovea is a speed reading app for iPhone and iPad that shows a book one word at a time, holding each word’s optimal recognition letter at a fixed point on screen so your eyes never have to track a line or find their place again.

It also carries a catalog of 605 original book summaries, each with an English and a French edition. Importing and reading books you already own is free forever; the summary catalog, the sepia and night themes, finer pacing control, sync and reading statistics are the paid part.

The technique is called RSVP, for rapid serial visual presentation. It has been studied since the 1970s, which means there is real evidence about what it does and does not do, and we publish that evidence rather than talk around it.

Does speed reading work?

Not in the way it is usually sold: you can learn to skim faster, but no technique lets you read several times faster while understanding the text as well as you would at normal speed.

The largest review of the evidence concludes it is unlikely readers can double or triple their reading speed while still understanding the text as well as if they read at normal speed (Rayner, Schotter, Masson, Potter and Treiman, 2016). Reading speed and comprehension trade against each other, and every product claiming otherwise is claiming something the literature does not support.

The reason is mechanical. Moving your eyes takes roughly a tenth of reading time, around 30 ms per saccade against fixations of 200 to 250 ms. The other nine tenths is your brain recognising words and building meaning, and no display trick shortens that.

What does work is unglamorous: practice, and a larger vocabulary. Skimming also works, for the jobs skimming is for. It is just not the same thing as reading, and we are not going to relabel it.

Does reading one word at a time hurt comprehension?

It can, and the clearest measured cost is to literal recall: in a direct comparison, RSVP readers scored 60 percent on questions about what the text explicitly said, against 72 percent for people reading the same passage normally.

That study also found the RSVP group read at 209 words per minute against 200 for ordinary readers, a difference that was not statistically significant, and that their blink rate more than halved, from 8.5 to 4.67 blinks per minute, alongside a measured rise in visual fatigue (Benedetto and colleagues, 2015). Interestingly, inferential comprehension was not affected: what suffers is the detail, not the argument.

Speed matters more than the format. With 209 participants across five rates, comprehension under RSVP matched ordinary reading up to 350 words per minute and dropped significantly above it (Di Nocera, Ricciardi and Juola, 2018). So a moderate speed is where RSVP behaves itself.

Fovea’s answer is not to deny the cost. It is to remove the two things that cause most of it: the app starts you at 300 wpm rather than at an impressive number, and it lets you go back, which the apps in those studies did not.

Can you go back and re-read something?

Yes, and that is the single most important difference between Fovea and the RSVP apps the research criticises.

Let go and the stream stops. Tap the left edge and it steps back a few words, one to ten, as you set it. Swipe left to scrub back word by word, and swipe further to move a sentence at a time. Chapter steppers in the header go both ways. When a passage needs more than a rewind, switch to the paginated view: it opens on the exact word you were on, with the paragraph around it, and you read it the ordinary way for as long as you want.

This matters because it is the specific thing the evidence says goes wrong. When researchers masked each word once the reader’s eye had passed it, so text could be read but never re-read, comprehension dropped, on ordinary sentences and not only on tricky ones. The authors concluded that regressions support comprehension, and that this calls the viability of eye-movement-eliminating speed reading apps into question (Schotter, Tran and Rayner, 2014).

About one eye movement in seven goes backwards during normal reading. Some of that is idle doubt and some of it is repair: you took a sentence the wrong way and you go back to fix it. A forward-only stream does not stop you making that error, it only stops you correcting it.

If eye movement is only a tenth of reading time, why bother?

Because a tenth is a real saving, and because the point of a fixed reading point is not only speed: there is no line to track, no place to lose, and nothing to find again.

We should be careful about how far we push that second half. The time saving is arithmetic and we can show our working. The claim that a single fixed point is easier to stay with than a page of text is not something we have controlled evidence for; it is what readers report, and we present it as that.

The honest summary is that Fovea changes how reading feels and modestly changes how long it takes. If you want a number, at the default 300 words per minute you are reading at roughly the pace of a competent reader on easy text, with the tracking work removed. Pushing the speed up is available, and the app shows you where the evidence puts the comprehension ceiling for the specific text you are reading.

How long is a Fovea summary, and how long does one take to read?

A typical summary runs about 6,000 words, which takes 24.0 minutes at the app’s default 300 words per minute, 18.1 minutes at 400, and 16.1 minutes at 450.

Those are median times across the 605 English summaries, measured by running the app’s actual pacing curve over each summary’s real text rather than by dividing words by speed.

The distinction is not pedantry. Fovea does not show every word for the same length of time: long words are held longer, commas and sentence and paragraph endings add a pause, and the first few words after you start are eased in. That makes real reading time about a fifth longer than words divided by speed would suggest, which is why 6,000 words at 300 wpm takes 24 minutes and not 20.

At 700 words per minute the same summary takes 10.5 minutes, but that is above where the evidence puts the comprehension ceiling, and at that speed you are skimming.

Are the summaries the actual books?

No. Every Fovea summary is an original work written from scratch in our own words, and it reproduces no part of the book it discusses.

A summary carries the ideas and the argument, not the sentences that expressed them, and runs at roughly a sixth of the length of the book behind it. No summary is written by, affiliated with, or endorsed by the author or publisher. Cover art is generated for Fovea and is never a book’s real jacket.

If you want the actual book, buy it, then import it into Fovea and read it here for nothing. That will never be behind a subscription.

Is Fovea good for dyslexia?

Some readers with dyslexia find it helps them, but Fovea makes no therapeutic claim and we will not tell you the research supports one.

Dyslexic readers do make longer fixations, shorter saccades and more regressions than typical readers, which makes it tempting to conclude that removing eye movements addresses the problem. Rayner’s review of the field argues the opposite: eye movements are generally not a cause of reading disability but a reflection of other underlying difficulties (Rayner, 1998). Removing a symptom is not treating a cause.

Fovea bundles OpenDyslexic and Atkinson Hyperlegible, and every typeface in the app is free forever, because charging for an accessibility affordance is not something we are willing to do. We will not oversell them either: controlled studies of dyslexia-specific typefaces have not found reading rate or accuracy benefits over ordinary fonts (Wery and Diliberto, 2017; Kuster and colleagues, 2018), and where a benefit has appeared it traced back to letter and word spacing rather than to letterform design.

The practical answer is that reading your own books in Fovea is free, so you can find out whether it suits you without paying anything or taking our word for it.

What does Fovea cost, and what is free?

Importing and reading your own books is free forever, with no page limit, no daily cap and no account required; a Pro subscription unlocks the summary catalog, the sepia and night themes, finer pacing control, sync and reading statistics.

There is no artificial cap on the free tier. If you own an EPUB, a Markdown file or a plain text file, you can read all of it, at any speed, with every typeface including the dyslexia-friendly ones, for nothing.

Fovea is finishing up for the App Store, so final pricing is set there.

Is my reading data private?

Yes: the app works with no account at all, and with no account nothing about you leaves your device.

If you never sign in, the books you import, your reading position, your settings and your reading statistics live only in the app’s local database and are removed when you delete the app.

If you do sign in, so your library follows you across devices, the server stores your account identifier, library metadata such as each book’s title, author and source, your reading position, and your reading preferences. The text of your books is never uploaded, and reading statistics are never synchronised and stay on your device.

There is no advertising in Fovea, no profiling, no sale of personal data, and no third-party analytics in the app itself. This website uses Google Analytics to count visits, which is described in full in the privacy policy.

Does Fovea work offline, and what languages does it support?

Reading works fully offline, including everything already in your library; only browsing the summary catalog and signing in need a network connection.

Every summary in the catalog exists in both English and French, so 605 works means 1,210 editions. The app picks the reading language from your device locale, and you can override it in Settings or switch edition from a book’s detail screen.

Books you import are read in whatever language they are written in. The pacing engine adjusts for English and French word frequencies, and simply contributes nothing rather than guessing when it does not know a language.