Does Reading Out Loud Help You Remember? The Short Answer
Yes. Reading information out loud reliably improves your memory for it compared with reading it silently, a finding psychologists call the production effect. In classic word-list experiments, people recognize words they read aloud roughly 10 to 20 percentage points more often than words they read silently, and the advantage has held up after a week and with full paragraphs of text.
There are two catches, and they matter more than the headline. First, the gain is in memory: in a 2024 series of experiments, reading aloud lifted scores on memory questions but not, reliably, on comprehension questions. Second, it works best when you use it selectively. The boost comes from making some material stand out from the rest, so reading a whole chapter aloud dilutes it.
That makes reading aloud a precision tool. It's free, it needs no app, and it pairs naturally with highlighting: you mark the lines that matter, then give those lines your voice.
What Is the Production Effect?
The production effect is the memory advantage for material you produce while studying (say aloud, mouth, write, or type) over material you only read silently. Colin MacLeod and his colleagues at the University of Waterloo named it in 2010, in a paper reporting eight recognition experiments in the Journal of Experimental Psychology: Learning, Memory, and Cognition.
The idea is much older than the name. In 1917, the psychologist Arthur Gates noted that his participants "reported that practice in accurate pronunciation of the material was an aid in learning." Hopkins and Edwards confirmed the effect with word lists in 1972. Then, as MacLeod and Glen Bodner wrote in a 2017 review, it "largely escaped attention" for decades while related findings, like the generation effect, became staples of memory research. Their explanation: "Perhaps the effect merely needed a name to make a name for itself." Once it had one, at least 40 articles followed by 2017, including a special issue of the Canadian Journal of Experimental Psychology.
A typical experiment is simple. Words appear on a screen one at a time in one of two colors. Blue means read it aloud; white means read it silently. Afterward, people pick the studied words out of a list mixed with new ones, and the words they spoke are recognized more often.
There's some irony in the timing. For most of history, reading aloud was simply how people read, a story we told in Reading Was Always Social. Science got around to naming the memory benefit long after silent reading had become the default.
Why Saying Words Aloud Makes Them Stick
The leading explanation is distinctiveness. When you read a line aloud, your memory of it picks up features that silent reading doesn't supply: the motor act of speaking, the sound of the words, and the fact that the voice is yours. Noah Forrin and Colin MacLeod (2018) put it this way: oral production helps "because it entails two distinctive components: a motor (speech) act and a unique, self-referential auditory input." At test, remembering the act of saying a line is evidence that you studied it, a shortcut researchers sum up as "if I recall saying it aloud, then it must have been studied."
Several findings point the same way:
- The production has to be unique to each item. In MacLeod's 2010 experiments, saying the same word ("yes") for every item marked for production gave no benefit. Mouthing each word, even without sound, did: recognition rose by about 18 percentage points over silent reading.
- Contrast matters. The effect is largest in "mixed lists," where some items are read aloud and others silently. When one group reads everything aloud and another reads everything silently, a 2013 meta-analysis by Jonathan Fawcett found a moderate recognition benefit, which MacLeod and Bodner describe as "markedly smaller" than the mixed-list effect. On free-recall tests, that between-group effect is consistently absent. For whole texts, the question is still open: "We also do not know whether reading all of a text aloud will result in a memory advantage," MacLeod and Bodner wrote in 2017.
- Rarity matters too. In a small 2014 study by Michal Icht and colleagues (20 people per condition), reading only 20% of a word list aloud widened the free-recall advantage to 26 percentage points (46% vs. 20%), compared with 8 points when half the list was spoken. When 80% was spoken, the effect flipped, and the few silent words were remembered better than the spoken ones.
- Part of the advantage is a cost to the silent material. In a mixed list, the lines you read silently can end up remembered worse than they would be in an all-silent list, as if your mind treats the spoken lines as the important ones. For free recall, the 2017 review concludes the mixed-list effect "appears to be almost exclusively due to a cost."
- It isn't just extra time. Reading aloud is slower, but the effect held up even when silently read words were deliberately shown for longer, and a 2012 study with text found no significant correlation between reading time and the size of the benefit.
In practice, reading aloud is less a way to learn more than a way to choose what you'll remember best. Point it at a small set of the right lines and that trade works in your favor.
What the Research on Reading Aloud Shows
The production effect replicates well, with one caveat: most of the work uses word lists, and with real text the gains get smaller (the 2024 experiments below found 4 to 9 percentage points on memory questions).
| Study | What participants did | What happened |
|---|---|---|
| MacLeod et al. (2010), 8 experiments | Read some words aloud and others silently, then took a recognition test (two experiments compared all-aloud and all-silent groups instead) | In mixed lists, words read aloud were recognized up to 20 percentage points more often; mouthing worked too; saying "yes" to every word didn't; all-aloud vs. all-silent groups showed no reliable difference |
| Ozubko, Hourihan & MacLeod (2012) | Studied words, word pairs, sentences, and textbook-style essays | The benefit was still there a week later, and paragraphs read aloud scored better on a fill-in-the-blanks test |
| Forrin, MacLeod & Ozubko (2012) | Spelled, wrote, typed, mouthed, or whispered words | Every form beat silent reading; writing, mouthing, and whispering, each tested in the same list as speaking aloud, fell short of it |
| Fawcett (2013), meta-analysis | Compared groups who read everything aloud with groups who read everything silently | A moderate recognition benefit (g = 0.37), which MacLeod and Bodner call markedly smaller than the mixed-list effect |
| Icht, Mama & Algom (2014), 60 participants | Read 20%, 50%, or 80% of a word list aloud | The advantage grew when spoken words were a minority; when they were the majority, it reversed on free recall and vanished on recognition |
| Forrin & MacLeod (2018), 75 students | Read aloud, heard a recording of themselves, heard someone else, or read silently | A gradient: reading aloud, then hearing yourself (marginally lower), then hearing others, then silent reading |
| Roberts et al. (2024), 4 experiments, 393 participants | Read reading-test passages aloud or silently, then answered multiple-choice questions | Memory questions improved in all three experiments that tested them; comprehension questions didn't reliably improve |
The 2024 paper tested the effect on the kind of passages students face on reading tests. Brady Roberts and colleagues at Waterloo and Flinders University had people read passages from the Nelson-Denny Reading Test and from test-prep material, some aloud and some silently, then answer multiple-choice questions. Accuracy on the memory-focused questions looked like this:
- Experiment 1 (47 students, in the lab): 81% for passages read aloud vs. 72% for passages read silently
- Experiment 2 (64 adults, online): 81% vs. 74%
- Experiment 3 (151 adults, online): 85% vs. 81%
The gain was largest in the supervised lab (9 points) and smaller in the two online experiments (7 and 4 points), which the authors suggest may reflect online testing, but it never vanished. With whole passages, people didn't see it coming. In Experiment 2, participants predicted slightly lower accuracy for passages they read aloud (59%) than for passages they read silently (64%), a difference that wasn't statistically significant, while their actual scores went the other way. With single words, by contrast, people do expect saying them aloud to help (Castel, Rhodes, and Friedman, 2013).
The effect isn't universal, though. Older adults show a smaller production effect than younger adults (Lin and MacLeod, 2012), and saying first names aloud did nothing to help people remember which name went with which unfamiliar face (Hourihan and Smith, 2016).
Reading Out Loud vs. Silently: Does It Help Comprehension?
Roberts and colleagues split their questions into two kinds. Memory-focused questions asked about facts stated directly in the passage. Comprehension-focused questions asked about the passage's theme or tone, or required an inference, which you can't answer by recognizing a sentence.
Reading aloud helped the first kind and not, reliably, the second. Experiment 1 looked like a comprehension win (80% vs. 74%), but it didn't replicate. Experiment 2 came in at 82% vs. 81%, Experiment 3 at 87% vs. 85% (not significant), and a fourth experiment with new materials and 131 adults, built entirely around comprehension, landed at 54% vs. 52% (also not significant). The authors' conclusion: production improves accuracy "only when the tested information is a very close match to the studied information."
Research on testing accommodations agrees. In a 2004 study by Batya Elbaum and colleagues, 311 middle and high school students (230 of them with learning disabilities) took two forms of a reading comprehension test, one under standard conditions and one while reading the passages aloud to themselves. Average scores didn't differ, and the students with learning disabilities didn't benefit more than their classmates.
Reading aloud also costs time. Marc Brysbaert's 2019 meta-analysis estimated that adults read silently at about 238 words per minute (for nonfiction) and aloud at about 183, so speaking takes roughly 30% longer. A 5,000-word chapter goes from about 21 minutes to about 27. Slower isn't automatically better, either: the 2024 series also tested a deliberately hard-to-read font, Sans Forgetica, and it didn't improve accuracy.
So divide the work:
| What you want | Reading out loud | Reading silently |
|---|---|---|
| Remember facts and exact wording | Better: 81% vs. 72% on memory questions (Roberts et al., 2024, Experiment 1) | Baseline |
| Understand theme, tone, and inferences | No reliable gain | Just as good |
| Speed | About 183 words per minute | About 238 words per minute (nonfiction) |
| Stay focused | Least mind-wandering of the three ways to take in a passage (Varao Sousa et al., 2013) | Mind wandered on about 36% of probes |
| Best for | Definitions, formulas, names, dates, key claims, quotations | First reads, skimming, following an argument |
For deep understanding, reach for strategies that make you explain, like putting the idea into your own words and asking why it's true.
Aloud, Whispered, Mouthed, or Written: What Works Best?
You can't always talk out loud while you study, and you don't have to. Researchers have tested most of the alternatives against speaking aloud or against silent reading, though rarely against each other, so read the middle of this table as a rough order rather than a strict ranking.
| Way of producing | Benefit vs. silent reading | Evidence | Good for |
|---|---|---|---|
| Singing | Larger than speaking normally or loudly | Quinlan & Taylor (2013) | Word lists and vocabulary when you're alone |
| Speaking loudly | Larger than normal volume | Quinlan & Taylor (2013) | Studying alone at home |
| Reading aloud at normal volume | Large: the benchmark | MacLeod et al. (2010) | Your default for key lines |
| Hearing a recording of yourself | Marginally below reading aloud in one study | Forrin & MacLeod (2018) | Commutes and review sessions |
| Whispering | Smaller than speaking: 14 vs. 23 points in the same list | Forrin et al. (2012) | Shared rooms |
| Mouthing without sound | Smaller than speaking: 13 vs. 29 points in the same list | Forrin et al. (2012); MacLeod et al. (2010) | Libraries and trains |
| Writing or typing | Smaller than speaking | Forrin et al. (2012) | Notes and flashcards |
| Hearing someone else read | Small | MacLeod (2011); Forrin & MacLeod (2018) | Lectures and audiobooks, if you add your own voice |
| Reading silently | Baseline | Comparison condition | First reads, skimming, and comprehension |
The pattern mostly follows the distinctiveness idea: the more distinct ingredients a production has, the better it's remembered. Speaking aloud has three (the movement, the sound, and your own voice). Whispering keeps the movement and some of the sound. Mouthing keeps only the movement. Hearing someone else has only the sound. Forrin and MacLeod describe memory performance rising "in step with the number of distinctive components at study."
The evidence on singing is thinner. Quinlan and Taylor (2013) found that reading loudly beat normal speech and that singing beat both, but that's a single set of experiments, and MacLeod and Bodner's review calls it "some indication." Treat it as promising, not settled.
Reading Aloud vs. Listening: Audiobooks, Lectures, and YouTube
Most "learning by ear" puts you in the weakest audible condition on that table: hearing someone else. In MacLeod's 2011 study, where each participant studied beside a second reader, words they read aloud themselves beat silently read words by 29 and 33 percentage points across two experiments, while words they only heard the other person read beat silent reading by about 10 points. Forrin and MacLeod's 2018 study found the same ordering with smaller gaps: 12 points for reading aloud and 4 for hearing someone else.
Listening has a second problem: your mind drifts. Trish Varao Sousa, Jonathan Carriere, and Daniel Smilek had more than 200 Waterloo students take in three roughly 1,800-word excerpts from Bill Bryson's A Short History of Nearly Everything, one read aloud, one read silently, and one heard as audio. When probed during the task, students' minds were wandering about half the time while listening (51%), about a third of the time while reading silently (36%), and least often while reading aloud. Listening also produced the poorest memory and the lowest interest.
In that study, reading a passage aloud kept students focused, but their scores on a 10-item true/false test about it weren't significantly higher than for the passage they read silently, even though every student read one passage each way. Staying on task isn't the same as remembering more.
You don't have to give up lectures; just talk back to them. Pause a video after each section and say the main point in your own words, which turns listening into retrieval practice. Then give the exact lines you need your own voice: read the key points from a YouTube Summary aloud, or save transcript lines with Glasp's YouTube highlighter and speak those later instead of replaying the whole video. If you're learning a language from video, the shadowing technique adds speaking to listening in a more structured way, though its main payoff is your ear and fluency rather than memory for content.
How to Use Reading Aloud as a Study Technique
Here's a routine that uses reading aloud where the evidence says it works, and leaves it out where it doesn't.
- Read silently first, for understanding. Reading aloud doesn't reliably help comprehension and takes about 30% longer, so make your first pass silent. Your goal on this pass is to follow the argument.
- Highlight what you'll need to recall precisely. Definitions, numbers, formulas, names, dates, key claims, vocabulary. With Glasp's web highlighter you can mark these lines on web pages and PDFs, and Kindle highlights can sit in the same library.
- Read your highlights aloud. This recreates the mixed list from the lab: highlighted lines spoken, the rest silent. Keeping the spoken part a minority may even help, since the advantage in Icht's 2014 study was largest when only a fifth of the words were read aloud and vanished or flipped when most of them were. That's one more reason to highlight sparingly: a page where every line is marked has nothing left to stand out against.
- Close the page and recall. Reading aloud makes lines distinctive; retrieval practice is what makes them last. Write or say everything you remember, then check it against your highlights. The blurting method is a structured way to do this, and active recall explains why it works.
- Come back to it over several days. The production effect lasted a week in the lab, but exams usually need longer. Revisit the same highlights in spaced sessions and recall before you reread.
- Adapt to where you are. Whisper or mouth in shared spaces. You can also record yourself reading your highlights and replay the recording on a walk: in one study, hearing a recording of your own voice beat hearing someone else's, though no study has tested this exact routine.
A history chapter shows the split. You'd read it silently to understand the causes and consequences, highlight the names, dates, and the key claims you'd need to defend on an exam, read those aloud, and then try to reconstruct the chapter's argument from memory. Reading aloud handles the facts; the reconstruction handles the understanding.
Frequently Asked Questions
Is it better to read out loud or in your head?
Out loud is better for remembering specific facts and wording; in your head is faster and just as good for understanding. Words and passages read aloud are remembered better than those read silently, but silent reading is roughly 30% faster and reading aloud hasn't reliably improved comprehension. For studying, use both in sequence: silent first to understand, then aloud for the lines you need to keep.
Does reading out loud help you understand what you read?
Not reliably. In a 2024 series of four experiments, reading aloud improved answers to memory questions in all three experiments that asked them, while a gain on comprehension questions about theme, tone, or inference showed up only in the first and didn't replicate. A 2004 study of 311 middle and high school students found that reading passages aloud to themselves during a reading test didn't change their average scores. If reading aloud feels like it helps you understand, it may be because it keeps you on the page: in a 2013 study, students' minds wandered less often when they read aloud than when they read silently.
Why do I remember things better when I say them out loud?
Speaking adds distinctive features to the memory: the movement of your mouth, the sound of the words, and your own voice. Later, the sense that "I remember saying that" helps you recognize the information as something you studied. That's also why the effect is strongest when spoken material stands out against material you read silently.
Does whispering or mouthing the words work?
Yes. Both beat silent reading, but they fall short of speaking at normal volume. In a 2012 study that mixed them into the same lists as speech, speaking beat silent reading by about 29 percentage points and mouthing by 13 in one experiment, and speaking by 23 and whispering by 14 in another. That still makes them good options for libraries and other quiet spaces.
Is listening to an audiobook as good as reading aloud?
No. Hearing someone else read is the weakest of the audible options: it beats silent reading in word-list studies, but by much less than reading aloud. In a 2013 study, listening to a passage also produced the most mind-wandering (about half of all thought probes), the poorest memory, and the lowest interest. If you learn from audio or video, pause and say the key points back yourself.
Does reading out loud help long-term memory?
Yes, for at least a week. Ozubko, Hourihan, and MacLeod (2012) still found better recognition for words that had been read aloud after a one-week delay. For exam-length retention, combine reading aloud with spaced retrieval practice rather than counting on a single read-aloud session.
Does reading aloud help when learning a language?
It can help you recognize new word forms: MacLeod's 2010 experiments found a production effect even for pronounceable nonwords, which is roughly what new vocabulary looks like to a beginner. Those studies tested recognizing the words, not linking them to meanings, so say a new word and its meaning aloud together rather than the word alone. For pronunciation and listening, a dedicated speaking routine like shadowing is the better tool.
Conclusion
Reading out loud is one of the few study techniques with a strong lab record and no learning curve. The production effect is real: spoken material is remembered better than material read silently, the benefit lasts at least a week, and it works on paragraphs as well as word lists. Its limits are just as clear. It helps memory, not comprehension. It works mainly through contrast, so reading everything aloud blunts it. And listening to someone else is a weak substitute for your own voice.
So use it the way the evidence suggests: read silently to understand, highlight the lines you need to keep, say those lines out loud, and then test yourself. Glasp keeps your highlights from web pages, PDFs, and Kindle books in one library, which doubles as your read-aloud list, and YouTube Summary turns a long lecture into key points you can say back.
