Easy at home, hard in the exam room: make your practice harder than the exam

The problem

At home on a Sunday evening, you do a listening sample paper. You listen, you answer, you replay when a passage escapes you, you check. Twelve out of fifteen. You feel ready. On exam day, same kind of recording, same length. But the audio does not come back, the clock is running, and the room is silent in a way that weighs on you. Eight out of fifteen. You go home wondering what vanished between the two. Nothing vanished: what you were measuring at home was not what the exam measures.

What the research found

Nicholas Soderstrom and Robert Bjork (Soderstrom & Bjork, 2015) reviewed decades of experiments on memory and motor learning and drew one simple distinction from them. Performance is what you get right during or just after practice. Learning is what remains days later, in another setting. The two do not always go together. Conditions that push performance up in the moment, such as rereading, redoing straight away, keeping supports within reach, often leave little lasting trace. Conversely, conditions that lower immediate performance can produce more learning.

Elizabeth and Robert Bjork (Bjork & Bjork, 2011) gave those conditions a name: desirable difficulties. Spacing sessions out instead of doing everything in one block, mixing types of exercise, and above all testing yourself instead of rereading. A worked example with numbers comes from Henry Roediger and Jeffrey Karpicke (Roediger & Karpicke, 2006). Students read short passages; some reread them, others had to write them out from memory. Five minutes later, the rereaders did better. A week later, the tested students remembered 56% versus 42% of the ideas for the rereaders. The method that looked least effective in the moment was the most effective at a distance.

The mechanism in plain words: when a support is available, your brain leans on it and does not need to build the path itself. When the support is missing, it has to search, and that search strengthens the path. The exam room is a place without supports. If your practice is full of them, you are training a different skill from the one about to be assessed.

Do this today

Take a short audio in French. Listen once, then answer the question or write down what you understood, before any replay. Only then listen again and correct. If you have time, add a constraint the exam does not have: turn the volume down a little, or answer in less time than allowed. Five minutes, more uncomfortable than usual. That is the point.

How our lessons use it

On the HumanFrench YouTube channel, the video « L’ascenseur » lends itself to this move. Listen once and, before pressing play again, say aloud what you understood of the story. Then listen again to check. The shadowing version adds the next difficulty: pauses where you have to repeat each sentence, with no text in front of you. The activity « Je m’appelle Nadia » (A1) works the same way: you hide the text and recall three details before checking. Start with « L’ascenseur ».

The honest limit

Soderstrom and Bjork offer a review, not a new experiment, and the studies they cite mostly concern memory for content in the first language, among students. The 56% versus 42% comes from a single experiment with short passages. None of this work concerns the TCF, the TEF or exam stress as such. The logic is sound; whether it transfers to your case is for you to test.

Your questions