Imitation versus Retrieval Practice in Foreign Language Pronunciation Learning
By Carolina Kuepper-Tetzel
A couple of months ago I have started re-learning French with a language learning app. I studied French in school for many years and was quite fluent in it at the end of my schooling. Unfortunately, I then stopped speaking or engaging in the language. Fast forward to today, I thought it would be great to refresh my memory of French, so I downloaded an app and started practicing.
In the app there are different activities asking you to match French to English vocabulary, identify the correct word from listening to an audio recording, fill-in-the-gap tasks, listening comprehension tasks, pronunciation tasks, and so on. The pronunciation task requires you to say back a sentence after listening to a recording of it. More specifically, that task is broken down in the following steps: (1) You see the sentence, (2) the audio recording is played instantly, (3) then you click and hold a button that records you saying the sentence, and (4) you receive feedback on your pronunciation performance. When I was doing the pronunciation task, it felt very easy to do because it was mainly a low-effort imitation task and I wondered whether that was the best way to practice pronunciation. The practice of imitation in foreign language learning is one that may also be happening in the classrooms where the teacher pronounces words out loud and students repeat them.
Image by Nothing Ahead on Pexels
The question then is whether imitation is the best approach to learning pronunciation or whether an alternative approach is equally or more effective. Kang et al. (1) investigated exactly this research question. They looked at comprehension and production (i.e., pronunciation) of foreign language words and pitched an imitation approach against a retrieval practice approach. In two experiments, undergraduate students in the US studied 40 Hebrew nouns in 3 or 6 learning cycles in one of two conditions:
Imitation condition: The image of the word was presented and the audio file pronouncing the word was played simultaneously. The learner then repeated the word aloud.
Retrieval practice condition: The image of the word was presented, and the learner was prompted to attempt to pronounce the word. Then, the audio file was played.
Final comprehension and production tests were given immediately after practice or two days later. For the comprehension test students heard each of the words and had to pick the correct picture for them. They found no difference between the imitation condition and retrieval practice condition on this test given immediately after practice. However, on the delayed test students performed better after having studied the material using retrieval practice than imitation. For the production test students were presented with images and asked to say the word out loud. Here, retrieval practice outperformed imitation on both immediate and delayed tests.
Importantly, performance in pronouncing the word correctly and overall quality of pronunciation during the final practice cycle was significantly lower in the retrieval practice condition compared to the imitation condition. This in itself is not surprising because one would assume that a learner’s performance to pronounce the right word correctly will be higher in a condition where they have just heard an audio file of it compared to a scenario where they have to retrieve it from memory. What is surprising is this additionally invested effort in attempting to retrieve the word from memory to pronounce it during practice pays off by resulting in better comprehension and pronunciation scores on the final tests after practice. This results pattern aligns well with the idea of desirable difficulties (2) that states that introducing effortful processes during practice (e.g., retrieval practice) is more beneficial in the long term even though during practice they may result in lower performances.
What to take away from this? After reading this study, I changed my behavior slightly when engaging with the pronunciation task on the language learning app: Before the audio file plays, I press the recording button and attempt to pronounce the sentence first. It does feel a bit more effortful, and I do make more mistakes than with the imitation approach, but that additional challenge makes it also more fun and hopefully more effective in the long run. In the classroom, teachers could embed activities that prompt students to attempt pronouncing words or reading out sentences first before offering them the correct pronunciation as feedback. That said, I would not completely discontinue imitation as a strategy. I think it has its place and can help scaffolding processes, too. Maybe mixing it up with retrieval practice is a good middle ground as it offers variation in practice which can be motivating.
(cover image by Kaboompics.com on Pexels)
References
(1) Kang, S. H., Gollan, T. H., & Pashler, H. (2013). Don’t just repeat after me: Retrieval practice is better than imitation for foreign vocabulary learning. Psychonomic bulletin & review, 20, 1259-1265. https://doi.org/10.3758/s13423-013-0450-z
(2) Bjork, R. A. (1994). Memory and metamemory considerations in the training of human beings. In J. Metcalfe & A. P. Shimamura (Eds.), Metacognition: Knowing about knowing (pp.185-205). The MIT Press, Cambridge, Massachusetts, USA.
