Previous task repetition studies have largely overlooked the second language learners’ development of linguistic knowledge as well as written accuracy. Furthermore, sufficient attention has not been paid to the role of written corrective feedback (WCF) in task repetition to reinforce attention to form. Moreover, studies exploring task repetition effects on learners with different prior knowledge of the target structure are rare. This study attempted to bridge these lacunas. Seventy-nine upper-intermediate learners in Iran were divided into four groups: task repetition with no feedback (TR), task repetition with metalinguistic feedback (TR+M), task repetition with direct feedback (TR+D), and task repetition with mixed direct metalinguistic feedback (TR+DM). All groups performed an error correction test that measured explicit knowledge, an elicited imitation test that tapped automatized explicit knowledge, and a picture-cued written production test that measured written accuracy. Participants performed a dictogloss task and received WCF before repeating the same task. Subsequently, they performed another dictogloss task with different content. Results revealed that the +Prior Knowledge learners in the TR+DM group gained explicit knowledge and proved slightly better than the TR+D regarding written accuracy. None of the groups, however, could develop automatized explicit knowledge.