Corrective feedback (CF) refers to responses to a learner's non-targetlike L2 production. It is a form of form-focused instruction; specifically, it falls under Type 3 (incidental focus-on-form) or sometimes Type 2 (planned) in Rod Ellis's taxonomy. Crucially, CF is a form of explicit instruction that is not PPP.
Lyster & Saito (2010) distinguish three broad categories based on the type of evidence they provide:
The teacher reformulates the student's utterance, minus the error.
Student: "Yesterday I go to the store." Teacher: "Oh, you went to the store?"
The teacher provides the correct form and clearly indicates the student's utterance was incorrect.
Student: "He goed home." Teacher: "Not 'goed': the past tense of 'go' is 'went'."
The teacher withholds the correct form and pushes the learner to self-repair. Includes elicitation, clarification requests, and repetition of error.
Student: "I have 25 years." Teacher: "You have 25 years? How do we say that in English?" (elicitation)
| Finding | Detail |
|---|---|
| Overall effect | Medium, maintained over time |
| Implicit vs explicit CF | Implicit CF effects better maintained long-term |
| Lab vs classroom | Lab studies show larger effects |
| Treatment length | Shorter treatments → larger effect sizes |
| Context | Foreign language contexts → larger effects than L2 contexts |
| Finding | Detail |
|---|---|
| Overall effect | Significant and durable on target language development |
| CF type | Prompts > explicit correction > recasts |
| Outcome measure | Effects largest on free constructed response measures |
| Setting | Instructional setting not a contributing factor |
| Treatment length | Long treatments > short-to-medium |
| Age | Younger learners benefit more from CF |
CF demonstrates that explicit instruction works through mechanisms completely different from PPP:
When Jason Anderson cites evidence that "explicit instruction works" to defend PPP, the corrective feedback literature is part of that evidence base, yet CF operates on entirely different principles.