Does Humming While You Read Actually Work? It's a Test, Not a Method

The advice circulates everywhere in the same shape. Hum silently while you read. Your inner voice no longer has room to pronounce the words, your eyes stay alone on the page, speed goes up. Sometimes it comes with a number stapled to the end, something like "double your speed in two weeks."

What's strange is that the trick actually does something. It's not a marketing invention.

It's a fifty-year-old laboratory manipulation, with its own name, designed specifically to break something. And the fact that it breaks something is exactly why it isn't a recipe for reading faster.

Humming Has Another Name in the Lab

The voice that pronounces words as you read them gets a full article of its own here, with everything that matters about it: why it shows up, what ceiling it imposes, and what you can do about it. In short, it's a trace of pronunciation running in parallel with your gaze, and the speed-reading niche treats it as the main culprit.

What the advice leaves out is where it came from. Humming layered over another task is called articulatory suppression, and it's the standard tool memory researchers use to check whether a process depends on inner speech.

The logic is surgical. You occupy the apparatus you'd use to speak — by repeating a syllable, counting, or humming — and then you watch what breaks. If the task still goes fine, inner speech wasn't involved. If performance drops, it was.

Baddeley, Thomson, and Buchanan used this procedure in 1975. Without suppression, lists of long words are recalled worse than lists of short words, because rehearsing them silently takes longer. Have people repeat a syllable out loud during the task and the difference disappears, because no one is rehearsing anything anymore. The system you block this way was later named by Baddeley (1986) the phonological loop.

So humming was born as a measurement instrument. Someone noticed it changes how you read and concluded the change was an improvement. It's a bit like pulling a part out of an engine to see what it did, noticing the engine screams louder and spins faster with no load, and selling that as a tune-up.

What Actually Breaks

When you hum over text, three things happen at once, and only one of them is what was promised.

Speed can go up. Nothing is waiting on pronunciation anymore, so your gaze can be pushed further along the line. The sensation is real, even from the first minute.

Comprehension drops, and not evenly. The synthesis Rayner and colleagues published in 2016, the most complete review of the speed-reading literature, reaches a conclusion that's inconvenient for the advice: inner speech isn't dead weight. It helps comprehension, especially where sentences are long, syntax is tangled, or you need to hold onto the start of a sentence to understand its end. It's the working memory that keeps the sentence together.

Hard text becomes unreadable. On a page of a novel you might not even notice. On three paragraphs of a contract, a technical article, or an anatomy textbook, you reach the end of them with nothing. Not less — nothing.

The second and third points aren't an accident of the method. They're the result the method was built to produce.

Carver (1990) had already spelled out the basic trade-off: past a certain pace, what you're doing stops being reading and becomes something else, with a different yield. Where that line actually falls shows up in a controlled evaluation of speed-reading training, Klimovich, Tiffin-Richards and Richter (2023): across thirty readers, the training did raise pace — by 35 to 42 words per minute, not the doubling the app promised — and comprehension was unchanged across the groups. Small gains cost nothing. What humming sells is the big jump, and that is exactly where comprehension breaks.

The Premise Behind the Advice Doesn't Hold Up Either

"Your inner voice caps you at speaking speed" sounds like a fact, but it comes with checkable numbers.

Brysbaert's (2019) meta-analysis, drawn from 190 studies, puts reading nonfiction aloud at around 183 words per minute and silent reading at 238. Silent reading is already almost thirty percent faster than your mouth. If inner speech were a metronome locked to speaking pace, that gap couldn't exist. It doesn't voice the words, it sketches them, and the sketch moves faster than the voicing. The full numbers, and what they mean for your own count, are in the article on average reading speed in words per minute.

And more important: what actually separates a fast reader from a slow one, once you measure both. Kuperman and Van Dyke (2011) tracked the eye movements of adult readers with different verbal skill levels and found that the best predictor of pace is word-recognition speed, which rests on vocabulary. Not on the mouth. Rayner (1998) showed the same thing from another angle: a fixation on a word lasts roughly 200 to 250 milliseconds, and that duration grows exactly when the word is rare or hard to integrate. The eye stays as long as the processing takes.

Humming doesn't speed up the processing. It just takes away one of its helpers.

The Honest Two-Minute Test

This is where the trick becomes useful. You use it once, to find out how much you lean on inner speech on your own material, not on a demo text.

You need the book you're currently reading, a pencil, and your phone's stopwatch. Pick two passages of roughly similar difficulty, about two hundred fifty words each, and mark them. At an ordinary pace, each one takes you around a minute, so the actual reading fits inside two.

Read the first passage normally, timed. Close the book and say out loud everything you remember, then write down how many distinct ideas you pulled out. Read the second passage while humming continuously, out loud, one held note or a wordless "la-la-la." Same timing, same recall out loud, same number written down.

The "out loud" detail isn't a whim. In the lab, suppression is done vocally precisely because that's the only way to verify from the outside that it's actually happening. Silent humming is impossible to verify, and it's the first thing you drop without noticing — exactly when the text gets harder, exactly where the measurement would matter.

You get speed by dividing word count by seconds and multiplying by sixty. And for the comprehension side, recalling from memory out loud is mandatory: the feeling of "I understood that" is the worst witness available, for the reasons described in the article on what happens when you read and remember nothing.

How to Read the Result

If speed went up and the idea count stayed about the same, you've learned something useful: on that text, you were pronouncing for nothing. The conclusion isn't to hum from now on, but that that kind of material can take a brisker pace anyway. Keep the pace then with something you can sustain for hours — a finger or a fixed time per page, not a tune.

If speed went up and the recall thinned out, you got the classic result from the literature. You bought speed with comprehension, at a bad exchange rate. The unpleasant part is that, without the two numbers on paper, you had no way to feel that.

And when the two figures come out identical, either you were already reading with very little inner pronunciation on that text, or you let the humming fade out along the way. Redo it, out loud, and pay attention to the moment you yourself stop humming: that's the sentence that demanded its voice back.

The result holds for that text and that day. On a novel and on an insurance policy you'll get two different numbers, and the gap between them is the most valuable piece of information in the whole exercise.

Why It Never Becomes a Habit

There are four reasons, and none of them come down to discipline.

You can't hum out loud in an open-plan office, a library, or a meeting. Silent humming, the only socially viable version, is precisely the one you can't verify.

Humming costs attention. The phonological loop has limited capacity, and you've just handed it a second task. You're paying resources for a probe, not for a tool.

Suppression takes out of play exactly the mechanism that helps you on hard text. It's like ditching the brakes because you go faster downhill.

And the last reason is the one the advice skips. It assumes a transfer: you hum for a few weeks, you stop, and the voice stays quiet on its own. That assumption is the unmeasured part of the story. What's been measured are the effects during suppression, and those include the loss of comprehension.

So use it once, pencil in hand, like a thermometer. It tells you how much you rely on inner speech, and on what kind of text. What you do with the answer gets handled differently after that — with pace kept from outside, with a conscious choice of what to read fast, and with a memory check after every chapter.

Measure your speed and comprehension on the same text