Audiobooks vs. Reading: Does Listening Count?

For stories and general non-fiction, listening counts. When Rogowsky, Calhoun & Tallal (2016) compared adults who read an e-book, listened to the audiobook, or did both at once, they found no significant difference in comprehension among the three groups. For dense academic material, the answer flips: students who listened to an academic article as a podcast scored worse on a quiz than students who read the same text (Daniel & Woody, 2010). That's the whole verdict — the medium barely matters for narrative, and matters a lot for hard material. Everything below is the detail.

What does the research actually say?

Two studies carry most of the weight in this debate, and they appear to disagree until you notice what people were reading.

The first is Rogowsky, Calhoun & Tallal (2016), published in SAGE Open. Adult participants took in the same general non-fiction material three ways: reading it as an e-book, listening to the audiobook, or doing both simultaneously. On the comprehension tests that followed, no group had a significant edge. Not the readers. Not the listeners. Not even the belt-and-suspenders group doing both at once — which is worth pausing on, because "read along while you listen" gets sold as a comprehension booster, and in this study it bought nothing measurable.

The second is Daniel & Woody (2010), in Computers & Education. College students got an academic article — the dense, argument-heavy kind you'd meet in a course — either as text or as a podcast. The students who listened scored worse on the quiz than the students who read.

So: is listening the same as reading? For a memoir, a thriller, a pop-science book, the evidence says the comprehension you walk away with is broadly the same. For a journal article, a textbook chapter, anything you'll be tested on — text wins, and not by accident.

Why the split? Narrative is forgiving. Stories and well-written general non-fiction are built with redundancy — context, recaps, momentum — so a moment of drift costs you very little. Dense argument is the opposite: nearly every sentence is load-bearing, and the format that lets you slow down for the load-bearing ones wins.

Why does text win when the material gets dense?

The difference isn't magic ink. It's control of the pace.

When you read, you regulate speed sentence by sentence without ever deciding to. Rayner and colleagues' eye-tracking research (2016) found that roughly 10-15% of eye movements during normal reading are regressions — small jumps back to something earlier in the text. On an easy sentence your eyes sail; on a hard one they slow down, loop back, re-check which noun that pronoun was pointing at. All of this happens in fractions of a second, below conscious awareness. Rereading is not a failure mode of reading. It's a built-in feature.

A narrator gives you none of that. Audio moves at one speed whether the current sentence is connective filler or the crux of the entire argument. You can pause and scrub back fifteen seconds, but that's a deliberate act with real friction — clumsy enough that most of us rarely bother, while our eyes would have regressed over the same passage a dozen times without breaking flow.

There's a second issue: structure. Dense material leans on things narration flattens — headings, equations, diagrams, the visible shape of a numbered list. A narrator reading a table aloud is a small tragedy.

And a third: hard material is slow on purpose. Brysbaert's 2019 meta-analysis of 190 studies — the same one that put average adult reading speed at about 238 WPM for non-fiction — found that technical material is read slower still. That slowdown is the comprehension work happening. A fixed-speed audio stream doesn't let a hard paragraph take the extra time it needs. It just keeps going.

See how fast you read text — free 2-minute test →

Are audiobooks cheating?

No, and this framing deserves to be retired.

There is no evidence behind it. The strongest direct test we have — the Rogowsky study above — found listeners comprehended general non-fiction as well as readers did. If listening were a shortcut that skipped the actual work, that result would be impossible. The work of a book — following the argument, tracking the characters, connecting chapter nine back to chapter two — happens after the words get into your head, and it's the same work whichever door they come through. Once you're a fluent reader, eyes and ears are two input cables running into the same comprehension machinery.

What the "cheating" charge usually is, underneath, is status anxiety about format: the sense that a book only counts if it involved a certain posture and a certain amount of eye strain. The point of a book is what ends up in your head, not the eye movements you performed to get it there.

One honest caveat, and it's about attention, not audio. Most audiobook listening happens while doing something else — driving, dishes, a run. Attention is the actual currency of comprehension, and a lane change or a boiling pot will quietly take a paragraph from you. That's not the format cheating you; it's multitasking doing what multitasking does. Reading with the TV on has the same problem, and nobody calls that cheating. They call it not really reading.

When does listening genuinely win?

There are hours in your week that print physically can't reach, and those hours are the audiobook's home turf.

Commutes and chores. A 40-minute round-trip commute, five days a week, is 200 minutes — more than three hours of potential book time that would otherwise go to the radio. (That's an illustration; the arithmetic is yours to adjust.) Folding laundry, cooking, walking the dog: same category. For narrative and light non-fiction, comprehension holds up fine, and three found hours a week is a serious amount of reading. If your real bottleneck is finding time at all, this is one of the honest answers in how to read more books.

Tired eyes. After a full day of screens, "read at 9pm" often loses to "stare at nothing". Ears don't have that problem.

Accessibility. For people with vision impairment or reading differences, audio isn't a compromise format — it's a front door to books, full stop. (And if reading itself has always felt unusually hard, that's worth a professional assessment rather than another productivity hack; we say more in why do I read so slow.)

Books you keep not starting. Some books sit on a shelf for years, then get finished in two weeks of commutes. A finished audiobook beats an unread paperback every single time anyone keeps score honestly.

When should you pick text instead?

Anything you'll be tested on. This is the Daniel & Woody result in one line: for academic material, the listeners scored worse. If there's an exam, a certification, or a meeting where you'll need the details, read. Students juggling heavy reading loads have more tactical options in speed reading for students.

Dense arguments, math, anything with notation. If understanding sentence twelve requires having genuinely gotten sentences one through eleven, you want the pause-and-reread control only text gives you.

When time matters and you read fluently. Spoken narration runs slower than most fluent adults read silently. At the average 238 WPM, an 80,000-word book is about 336 minutes of reading — call it five and a half hours, give or take (arithmetic you can check). Most fluent readers will finish the print version in less time than the audio takes at normal speed. If you don't know your own silent reading speed, the free reading speed test takes about two minutes and measures comprehension along with it.

When you plan to remember it long-term. Here the medium matters less than what you do afterward. Retrieval practice — closing the book and testing yourself — beats rereading for delayed retention (Roediger & Karpicke, 2006), and practice testing plus spaced review came out as the two highest-utility study techniques in Dunlosky and colleagues' 2013 review. Both work identically after a listening session, by the way. The full method is in how to remember what you read.

Should you listen at 1.5x or 2x speed?

Carefully, and with lower expectations than the productivity internet suggests.

The tradeoff mirrors reading exactly. Rayner and colleagues' 2016 review found that comprehension declines as reading speed is pushed, with true reading-with-comprehension topping out somewhere around 400-500 WPM — a ceiling set by the brain's meaning-building step, not by the eyes. There's no reason to expect the listening version of that bottleneck to be more generous. It's the same machinery. Speeding up narration doesn't upgrade your comprehension; it just delivers words faster to the same fixed-capacity process. We've made the longer version of this argument in the speed reading science piece — the short version is that every medium eventually hits the same wall, and the wall is understanding.

In practice: a modest speed bump on light, familiar material — a genre you know well, a re-listen, a breezy memoir — usually costs little, since narrators pace for clarity across all listeners rather than for you specifically. Doubling the speed on new or dense material is the audio version of skimming: sometimes exactly what you want, dishonest only when you then count it as having read the book.

If you want to push the dial anyway, push it the way you'd push reading speed: in small steps, with a check. Go up one notch, listen for a chapter, then ask whether you could explain that chapter to someone. Hold the new speed only if the answer is yes. The dial going up is not the goal. The book going in is.

Measure the reading half — 2-minute test with comprehension score →

Use both — deliberately

This was never a war between formats. It's a matching problem.

• Narrative and general non-fiction: either medium, whichever fits the hour you actually have.
• Dense, technical, will-be-tested material: text, at whatever pace the material demands.
• Hours when your eyes are busy or spent: audio, at a speed you can still retell.

That's also how we build ReadingFast: alongside the training drills and the speed test, the app includes a Listen mode where neural voices read your own books and uploads aloud, so one subscription covers both halves of this article.

"Does listening count?" usually hides a different question: am I allowed to get books this way? You are. Count the books, and count what stayed with you afterward. Nobody is keeping score of your eye movements — and after the Rogowsky result, they'd have nothing to score anyway.