Who was the second student for?

There is a moment in almost every Michel Thomas recording when one of the students gets it wrong.

Thomas teaches two beginners at once, out loud, while you listen at home as a sort of third student. When one of them stumbles, you hear the correction, the second attempt, and often the other student having a go as well.

On first listen it sounds like padding. It isn’t. We found that out by taking it away.

A series on how Echoa teaches

This is the first post in a series about the method behind Echoa: which ideas we’ve borrowed, where they come from, what the research says about them, and what we’ve had to change to make them work.

Two promises before we start.

We’ll show our sources, and say what each study did and didn’t test.

And we’ll keep two questions apart: whether an idea is well supported, and whether our course works. The research can answer the first. The second needs evidence we’re still collecting, and we’ll write about it when we have it.

The method doesn’t have a name yet. We’re working on it. The shortlist is not inspiring.

What did we borrow?

Very little in language teaching is invented from nothing, and we’re no exception. Three ideas do most of the work.

One new piece at a time

From Michel Thomas. Each new word or pattern is combined straight away with what you already know.

Learn “Ich möchte einen Kaffee” and you can ask for a coffee. Learn “keinen” and you can turn one down: “Ich möchte keinen Kaffee.” Learn “Kann ich … haben?” and you can ask whether you can have one, and then whether you can have one without milk: “Kann ich einen Kaffee ohne Milch haben?”

The sentences get longer. Each step adds one thing.

Coming back to it later

From Paul Pimsleur, who in 1967 proposed a schedule for returning to new words at growing intervals: seconds, then minutes, then hours, then days.

The idea underneath it, that practice spread out over time is remembered better than practice crammed together, is one of the most thoroughly studied effects in memory research. One review pooled 317 experiments on it. Cepeda and colleagues, 2006

The exact schedule is less settled. When researchers compared Pimsleur-style growing gaps with even ones, growing gaps did better on a test ten minutes later, but even gaps did better two days later. Karpicke and Roediger, 2007 That study used vocabulary pairs, not conversation. So we keep the spacing and hold the schedule loosely.

Saying it, not just hearing it

Pulling something out of memory tends to strengthen it more than going over it again.

In one study, students who were tested on a passage remembered more of it a week later than students who simply re-read it. The re-readers, meanwhile, felt more confident. Roediger and Karpicke, 2006 The passages were prose, not spoken language, but the point carries: the version that feels easier is not always the one that sticks.

So in Echoa you build each sentence yourself before you hear it.

What happens without the second student?

Echoa is one learner and one teacher. There is no second student.

That removes the stumbles and the corrections, and with them a good deal of repetition. In a Thomas recording, a sentence might be tried, corrected and tried again by two people before the lesson moves on. You hear it several times without anyone setting out to repeat it.

With one learner, a sentence that goes well is said once. Tidy, and gone.

The obvious fix is to add the repetition back as drills: now say it again. That gets tedious quickly. And saying something you heard a moment ago is closer to echoing than remembering.

So the repetition has to come from building. When you say “Kann ich einen Kaffee ohne Milch haben?”, you’re making a new sentence, and you’re also bringing back “Kann ich”, “einen Kaffee” and “haben” from earlier. Older pieces return as working parts of new sentences, and the gap before they’re needed again grows as the course goes on, across lessons as well as within them.

The second student’s job, it turns out, was repetition. In Echoa, the sentences do it.

That is the part we’d call ours. It’s an idea about teaching, and the principles it rests on are well supported. We haven’t shown that it does better than the alternatives, and we won’t pretend otherwise.

What’s coming in the series?

The next posts will take one part of the method at a time:

• why a verb gets far more practice than a noun
• what counts as “one new thing”, and why that matters more than it sounds
• mistakes, nerves and how corrections are given
• why you build the sentence before you hear it
• what we can and can’t yet say about whether it works

Where does Echoa fit in?

Echoa is a speaking course built on these ideas. You hear an English cue, build the sentence in the language you’re learning and say it aloud, and Sam, your teacher, responds to what you actually said. Pieces you learnt earlier come back inside new sentences, lesson after lesson.

No second student required. You’re welcome to bring one.