The article reports on a study investigating the comparative effects of immediate and delayed corrective feedback in learning the English past passive construction, a linguistic structure of which the learners had little prior knowledge. A total of 120 learners of English as a foreign language (EFL) from 4 intact classes at a Chinese middle school were randomly assigned to conditions: immediate feedback, delayed feedback, task-only, and control. The 3 experimental groups attended a 2-hour treatment session where they performed 2 dictogloss (narrative) tasks in groups, each followed by a reporting phase in which they took turns telling the narrative to the class. The 2 feedback groups received either immediate or delayed corrective feedback in the form of a prompt, followed by recasts of utterances containing errors in their use of the target structure. No effect for the corrective feedback was found on elicited imitation test scores, but both the immediate and delayed feedback resulted in gains in grammaticality judgment test scores, with immediate feedback showing some advantage over delayed feedback. We interpret these results as showing that the feedback only aided the development of declarative/explicit knowledge and that the advantage found for immediate feedback was due to the learners using the feedback progressively in the production of new past passive sentences, whereas this did not occur in the delayed feedback condition.