Does Photoreading Work? What the Only Independent Test Found

You "photograph" the page in a single glance. One page per second. 25,000 words per minute. Three books in an afternoon.

Search for it, and you'll find roughly eight results claiming it works. Not one of them cites a study.

There is exactly one. It's public, it's free, it runs 22 pages, and NASA Ames paid for it. The conclusion is short, and it isn't pleasant.

The problem isn't that you didn't try hard enough

You did the steps. The preparation, the breathing, the defocused gaze sweeping the page, the incubation pause, then the activation pass.

And something happened. The material felt familiar. Words seemed to jump out as if you'd already met them. You had the clear sense that you knew what was on the page.

Then someone asked you what chapter three said. Or you sat an exam. And the feeling turned out to be empty.

It's the same as recognizing a song from its first three seconds but not being able to sing two lines of it. The recognition is real. The knowledge isn't.

If you paid for a photoreading course, you weren't naive. You were persuaded by a mechanism that sounds plausible, taught by people who genuinely believe in it, and you had an experience that really does produce something.

It just produces a feeling of knowing. And the feeling of knowing and actual knowledge part ways exactly when you need the second one.

What your eye actually does when it crosses a page

You don't read fluidly, even though it feels that way. Your eye jumps.

It stops somewhere on the line for roughly 200 to 250 milliseconds. That stop is called a fixation, and it's the only moment new information gets extracted. Then the eye makes a short jump, a saccade of about seven or eight letters, and stops again.

The zone from which your brain can actually identify words is small. Roughly seven letters to the right of the fixation point — the current word plus one more, maybe two.

Think of a flashlight in a dark room. The circle of light is sharp. Outside it you can make out shapes, contrasts, maybe blocks of text, but no letters you could actually name.

Photoreading assumes you can widen that circle until it covers the whole room. This is where it falls apart: researchers tested it, and you can't.

In a 2009 experiment, researchers artificially enlarged the letters in peripheral vision so they'd be as visible as the ones at the center. The useful zone didn't grow at all.

And in the other direction, masking a single letter right at the point of fixation cut reading speed in half. So peripheral vision can't take over the job, not even under perfect conditions.

The team that synthesized the entire literature on the subject, in a 2016 paper, put it bluntly: reading large chunks of a page simultaneously through peripheral vision "is not possible, biologically or psychologically."

The limit isn't in the eye. It's in how fast the brain recognizes words. It isn't a training problem, it's the architecture of the system.

What you were told, and why it's wrong

Paul Scheele, the method's creator, claimed the figure of 25,000 words per minute back in 1993. The same number still circulates today, word for word, everywhere the method is sold.

Danielle McNamara, a researcher in reading comprehension, tested the technique with NASA Ames funding. The final report came out in January 2000 and is still publicly available.

Two people were tested. One was the investigator herself, after a two-day workshop followed by about two months of practice. The other was a certified instructor, trained directly by Scheele, with over three years of practice and roughly 150 students taught.

It needs to be said plainly: two subjects is a small study. I'll come back to that limitation in a moment.

On expository texts of about 1,500 words each, with 24 open-ended questions per text, comprehension dropped 20% for both readers. Broken down, conceptual questions lost 30%, text-based questions only 10%.

In other words, deep understanding is exactly what collapsed first.

On speed, the number that matters is even simpler. 112 words per minute with photoreading, 114 at normal reading speed. No difference. Not a trace of 25,000.

The gap between the two figures has an explanation the instructor gave herself, during the workshop. The numbers in the ads are calculated only from the "photographing" phase of the page. The other four phases, which take up nearly all the time, don't get counted.

The exam that settles it

The hardest part of the report isn't a table. It's an actual exam.

The expert instructor photoread three physiology chapters, about 26,350 words, in 73 minutes. Then she sat the real exam from a master's-level course.

Before the exam, she rated her own understanding of the material at 4.5 out of 5. She estimated she'd retain about 68% of the content — a solid grade.

Here's what she got:

• 0% on the conceptual questions (she didn't even attempt the short-answer and fill-in-the-blank items)
• 35% on multiple choice, where pure guessing gets you 25%
• 67% on true/false, where pure guessing gets you 50%
• 29% on the prior-knowledge questions

She failed the exam. After seeing the result, she revised her self-rated understanding from 4.5 down to 2.

This is the heart of the whole story. The method doesn't produce knowledge, it produces confidence, and confidence is exactly what stops you from checking.

McNamara identifies two sources of this overconfidence. The first is priming. After the preview pass, concepts "feel familiar" on the second pass, and the brain reads that familiarity as memory.

The second is the self-suggestion affirmations taught as part of the method, things like "everything I photoread leaves a lasting impression on my inner mind." These build a personal stake in the technique's success, which makes the result hard to judge objectively.

What a two-person study actually tells you

A skeptic might say two subjects prove nothing. They're right in principle, but not here, for three reasons.

First: the claim being tested is 25,000 words per minute. Disproving it only takes the most advanced available practitioner failing to hit it — and that's exactly what happened.

Second: 26 years have passed. Nobody has published a contrary result, not even the organization that sells the method.

Third: the claimed mechanism — "preconscious" processing of the page through a divergent, unfocused gaze — is incompatible with everything that's been measured about how people identify words. It isn't just untested, it's impossible given the architecture described above.

The report's conclusion, in the author's own words, runs two lines: the results clearly show no benefit to using the PhotoReading technique, and the extremely high rates claimed by practitioners were not observed.

What actually works instead

You can't absorb a page in a single glance. But you can decide what's worth reading slowly and what you can skip — which is closer than it sounds to the promise that drew you in.

The best phrasing I know of comes from a speed-reading course, not a lab: "Skimming doesn't replace reading. Skimming decides WHAT deserves to be read closely."

The irony is that the only part of photoreading that survives scrutiny is the preview pass. Table of contents, headings, the first and last paragraph. That's structural skimming, and structural skimming actually has evidence behind it.

So if you paid for a course, there's something worth salvaging. Keep the preview, drop the page-photographing phase with the defocused gaze. The first is a real triage tool. The second is the one that produced 112 words per minute.

In a 2009 experiment, under time pressure, skimming an entire text quickly produced better memory for the important ideas than reading half of it carefully. Another study found that readers who paid attention to headings wrote the most accurate summaries.

In practice, that means going through:

• the table of contents and section headings, to see the map
• the first and last paragraph of each section
• whatever is bolded, boxed, or charted

And now the honest part, the one the hype skips: skimming has a cost, and the cost has been measured. In a study of 330 students, reading rate rose from 225 to 600 words per minute. Memory for important information dropped by exactly as much as memory for unimportant information.

In other words, the eye doesn't select what matters on its own. The selection works because it's structural, not perceptual. Headings do the triage, not your gaze.

Which gives us a simple metric, one that deserves to replace the words-per-minute number in your head.

Effective Speed = words per minute × comprehension percentage.

A reader at 380 words per minute with 85% comprehension beats a reader at 500 with 55%. And photoreading, measured honestly, comes out to 112 words per minute with 20% comprehension lost. Below normal reading, on both counts.

There's one more piece of news, and it's the good kind. Across an entire battery of cognitive skills tested, the strongest predictor of reading speed wasn't any eye technique — it was how fast you recognize words. And word recognition rests on vocabulary and on how much you already know about the topic.

You can see it in the extreme cases too. When speed-reading graduates were given technical texts they knew nothing about, their edge over plain skimmers disappeared completely. Their speed had, in effect, been knowledge all along.

And knowledge is built by reading. It isn't the shortcut you were looking for, but it's the only one that actually leads anywhere.

How to practice this

Start with a measurement that includes the part photoreading hides.

Our public test takes a few minutes. It gives you your speed in words per minute, but it also asks what you actually read, so you can see what's left of that speed once you subtract what you didn't understand.

The adult average, from a meta-analysis of 190 studies, is 238 words per minute for nonfiction and 260 for fiction. These are general adult benchmarks, not a personal baseline — the best anchor available until you measure your own.

After the test, the rest is simple and unglamorous. Read a lot, choose deliberately what you read fast and what you read slowly, and test yourself from memory after every chapter.

It won't give you 25,000 words per minute. It will give you something photoreading never produced, in any test: knowledge you can actually use when someone asks.

Measure your speed and comprehension