Roy Lyster: Oral corrective feedback as a catalyst for second language development

 

First International Conference on Written Corrective Feedback

University of Vic, 14-15 April, 2023

Roy Lyster

Oral corrective feedback as a catalyst for second language development

This was the evening plenary of the first day of the 1^st International Conference on Written Corrective Feedback, held at the University of Vic, Catalonia, 14^th April, 2023.

Roy Lyster is Professor Emeritus of Second Language Education at McGill University in Montreal (Canada). He is best known in the field of language teaching education for his work on corrective feedback and learner uptake, e.g. Lyster & Ranta, 1997.

 

Complementary approaches to L2 teaching

Professor Lyster explained that the vast majority of his research, training and indeed his own teaching has taken place in an EMI (English as Medium of Instruction) context in bilingual education in Canada (= CLIL in Europe). Based on his extensive observations of ELT in these classes, he suggests that effective L2 teaching should use these two approaches in tandem:

Proactive approach	Consists of planned content-based  or meaning-oriented instruction designed to meet curricular objectives while fostering language use and language awareness
Reactive approach	Ensures that oral interaction is a key source of language learning though scaffolding techniques such as teacher questions and corrective feedback

 

To correct or not to correct?

Errors are a part of learning and indicate that learners are formulating and testing their hypotheses about the target language. However, in order to make sure learners benefit from this trial and error, these hypotheses need to be confirmed or dismissed. In other words, they need corrective feedback (CF)

CF has an essential scaffolding role, because it allows teachers to provide language support during meaningful interaction. There is a large body of research to support the effectiveness of oral corrective feedback (OCF), including at least four meta-analyses, e.g. Mackey & Goo, 2007. Basically, providing correction helps!

Teachers are understandably reluctant to interrupt the oral interaction of learners making an effort to talk in order to provide immediate error correction. Furthermore, teaching manuals and training courses generally discourage intervention during fluency activities in favour of noting errors and providing delayed corrective feedback (DCF) after the activity has concluded. However, Lyster & Ranta (1997) found that none of the feedback types (see below) disrupted the flow of interaction. On the contrary, students expected to be corrected and were not frustrated by it.

Types of oral corrective feedback

CF can be explicit – you tell the student you are correcting them and provide the correct version, or implicit – the learner has to do the work, noticing and/or producing the correct version. The implication is that the more work the student does (Depth of Processing), the more memorable it is and more learning occurs.

There are two families of OCF:

·         Those which provide the correct form: Reformulations. These are explicit correction (e.g. no – don’t say that, say this) and implicit - recasts (repeating the learner’s utterance correctly before continuing with the next conversational move)

·         Those which withhold the correct form: Prompts. These ‘push’ the students to self-repairs. They are receiving feedback and producing output (Swain, 2005). These include: clarification requests; metalinguistic clues, elicitation, repetition of error.

Teachers love recasts

In their observations of French immersion classes, Lyster and Ranta found that well over half of CF episodes were recasts.

There are good reasons for this. They preserve the student’s original meaning and move the conversation along naturally. Also, if noticed, they free up the learner’s attentional resources (Long, 1996). Teachers justify their use of recasts as a way of correcting students without embarrassing them in front of others. However, this contradicts the message that ‘It’s OK to make errors, it’s a normal part of learning’. Also, in order to be effective, the recasts need to be noticed as CF. Students may interpret them differently!

What do teachers need to know?

Lyster believes that teachers need to be aware not only of the types of OCF they can provide, but also of the learners’ responses following the feedback. These responses (uptake) basically fall into two categories: learner repair or utterances still in need of repair. This uptake provides an indication of the learner’s current level of performance. This is particularly important in contexts where ‘dynamic assessment for learning’ is used to monitor students’ current levels in order to ‘push’ them to the next level (e.g. Lantolf & Poehner, 2010)

“ The purpose of OCF is not only to provide students with information about accuracy but also the means for them to move towards more advanced levels of performance” (emphasis in original)

In praise of prompts

Lyster argues that prompts are more effective: they help students to notice errors and generate correct forms. Also, learners remember information better when they take an active part in producing it rather than having it provided for them by an external source. He recalls an instance in his own teacher training when he was told that under no circumstances should a teacher repeat a student error as it may be interpreted as the correct form. However, learners are intelligent enough to pick up the intonational and non-verbal clues that a teacher uses when repeating an error as a prompt.

We watched two short videos of classroom interaction (same teacher, same pupils, grade 5 in an English immersion programme for L1 French speakers). The class was content-based and meaning-focused, discussing a class reader. In the first, the teacher used recasts and the interaction moved along effectively. In the second, the teacher used a variety of prompts, and Lyster highlighted the affordances that these provided (see his comments on the transcript below)

He pointed out that some of the most effective teaching he had observed over the course of his career was when a teacher ‘smartly played dumb’.

Implications for teacher training

In their seminal paper on classroom discourse, Sinclair and Coulthard (1975) identified the most typical classroom questioning exchange, consisting of three moves:

Initiating (I) move by the teacher Responding (R) move by the learner Evaluative (E) by the teacher

T: What would Max be feeling? S: He would be very worried T: Yes, very worried. That’s a really good word. (Tedick & Lyster, 2020)

This IRE pattern generally concludes with a teacher move which effectively shuts down the interaction. Teachers can be trained to ask follow-up questions to scaffold continued discourse.

Furthermore, as research has not yet provided conclusive evidence in favour of indirect v direct or comprehensive v selective correction (either in oral or written corrective feedback), this effectively leaves the door open for teachers to use a variety of feedback types according to their specific language learning objectives.

Prof. Lyster concluded his presentation by appealing to teachers to use the repertoire of techniques available to them, and to reconsider our fondness for recasts

References

Lantolf, J.P. & Poehner, M. E. (2010) Dynamic assessment in the classroom: Vygotskian praxis for second language development. Language Teaching Research, 15(1):11-33.

Long, M. H. (1996). The role of linguistic environment in second language acquisition. In W.Ritchie and T. K. Bhatia (Eds.), Handbook of second language acquisition (pp. 413-468). San Diego: Academic Press.

Lyster, R. & Ranta, L. (1997) Corrective feedback and learner uptake. Negotiation of Form in Communicative Classrooms. Studies in Second Language Acquisition, 19(1), pp. 37 – 66.

Mackey, A. & Goo, J. (2007) Interaction research in SLA : A meta-analysis and research synthesis. In Mackey, A (Ed.) Conversational interaction in second language acquisition: a series of empirical studies. Oxford University Press, p. 407-453.

Swain, M. (2005). The output hypothesis: Theory and research. In E. Hinkel (Ed.), Handbook of research in second language learning and teaching. Mahwah, NJ: Lawrence Erlbaum.

Tedick, D. J. & Lyster, R. (2020) Scaffolding Development in Immersion and Dual Language Classrooms. Routledge