Reading and listening at the same time

Reading along with audio is one of those techniques that gets recommended everywhere and examined almost nowhere. The claim is usually that it doubles your input, or that it trains your ear and your eye at once, and it is presented as obviously good.

It is good. But the evidence for why is more interesting than the pitch, and getting it right changes how you should do it.

What the research actually shows

Start with the disappointing part, because it is the part nobody mentions.

For comprehension, a recent meta-analysis of reading while listening against reading alone found a trivial overall benefit, and the effect showed up mainly in studies where the reading was paced by the experimenter rather than by the reader. Where people read at their own speed, there was no reliable comprehension advantage.

So if you were told that reading with the audio running will make you understand more of what is in front of you, that is not what the evidence supports. Read a page silently or read it with narration and your grasp of that page is likely to be about the same.

For vocabulary, the picture is better. Studies comparing reading-only with reading-while-listening generally favour the paired condition, and assisted repeated reading has produced significantly larger vocabulary gains than reading unassisted. Where all three modes have been compared, reading while listening came out ahead of reading alone, which in turn beat listening alone.

The proposed mechanism is worth knowing: audio helps you cut the stream into meaningful chunks, and it can free up attention that would otherwise go on decoding the writing. You are not getting two inputs for the price of one. You are getting one input that is easier to parse.

The thing it is actually for

Here is what none of that measures, and what makes the pairing worth doing anyway.

You can know a word on the page and not recognise it when spoken. These are close to two separate pieces of knowledge, and for most learners the written one races ahead. You read ils ont comfortably, then hear something like “ilzon” and catch nothing. You know the character, and then somebody says it at speed with a tone you were not expecting.

That gap does not close by reading more, and it does not close by listening more either, because listening alone gives you no way to find out which word you just failed to recognise. It closes when both are in front of you at the same moment: you hear a sound you do not know, your eye is on a word you do know, and the two get stapled together.

This is why the pairing matters more in some languages than others. In French it is the single biggest obstacle, because liaison and elision mean words visibly change shape when they touch. In Japanese and Chinese, a character gives you no reliable guide to its sound, so the written and spoken forms have to be learned as a pair or they never connect. In Spanish, where spelling maps closely onto sound, the gap is much smaller and the pairing does correspondingly less.

What the pacing finding tells you to do

The most useful detail in that meta-analysis is the pacing one, and it is worth taking literally.

The comprehension benefit appeared when the reading was paced. Audio is a pacer. It will not let you stall on a word, drift back over the same line four times, or slow to the crawl that silent reading in another language tends to become. It drags you forward at the speed people actually talk.

That is uncomfortable, and it is most of the value. Reading at speech speed is a different skill from reading at your own speed, and it is the one you need if you ever want to follow a conversation. So:

• Let the audio set the pace and do not pause it constantly. If you stop it every few seconds you have turned the exercise back into ordinary reading with extra steps.
• Go through a piece once at speed and accept that you will miss things. Then, if it is worth it, go back and read the same passage properly, with lookups.
• Repeat the same piece rather than always taking a new one. The assisted repeated reading studies show the clearest vocabulary gains, and a second pass over something familiar is where the sound and the shape actually fuse.

Which one first

There is no settled answer, so pick by what you are trying to fix.

Listen first, then read if your problem is that you lean on the text. You find out what you genuinely caught, then the reading tells you what you missed and why.

Read first, then listen if the material is near the edge of your level. You get comprehension out of the way, then the audio does its real job of attaching sounds to words you have just understood.

Together, in one pass is the default and the one to use most of the time, particularly on repeat passes.

A caveat worth keeping

The research also finds that this interacts with the learner. Reading while listening is not uniformly better for everybody, and if you find the audio pulls your attention off the text and you end up following neither, that is a real effect rather than a failure of discipline. Try listening first instead, or drop the audio on the first pass and add it on the second.

The summary

Reading with the audio running will not make you understand today’s page better. What it does is stop the written and spoken forms of a language drifting into two separate languages in your head, and it makes new words stick better than reading alone does.

It is also, given that speech needs a smaller vocabulary than print, a better on-ramp than most learners think.

---

Sources: Hui, B. et al., Listening, Reading, or Both? Rethinking the Comprehension Benefits of Reading-While-Listening, Language Learning, for the comprehension meta-analysis and the pacing moderator; Comparing incidental vocabulary learning from reading-only and reading-while-listening, System; Exploring the effect of assisted repeated reading on incidental vocabulary learning.