The objective of this thesis is to diagnose the strategies of metacognitive control in writing of pre-service teachers (PSTs) in the deferred revision of two writing tasks in L1 (Spanish/Catalan) and English as a foreign language (EFL). Two exploratory studies were designed. In the first one, ninety-eight (N=98) PSTs took part they were split into higher (B2&C1) or lower (A1-B1) EFL proficiency and wrote either an essay (knowledge-transforming) or a summary (knowledge-telling) in their L1 or EFL. Texts were revised two days later so that text quality could be improved. The metacognitive control was assessed following mostly Allal's (2000) taxonomy. Each aspect of those changes or transformations in the texts were validated and coded and so were the criteria to measure the quality of the texts and types of mistakes made before and after revision. The results suggest deferred revision was carried out depending on the language, task and level of English. Transformations in L1 texts had greater impact in the content and text organisation, in EFL were significantly more numerous but affected more superficial levels, mostly word-based substitutions in both texts and groups. Revision of essays in EFL seemed to have added more meaning: substitutions had a greater impact than they did on the summaries. Participants with lower EFL proficiency focused on aspects of superficial level with little influence on the content or the text structure. Participants with a higher EFL proficiency focused on longer structures which had a greater impact on the text. Higher EFL proficiency participants improved the quality of their texts more, the differences in L1 were slight between both groups. The second study replicated partially the first one with four (N=4) experts and eight (N=8) PSTs with elementary or intermediate EFL proficiency. Participants completed self-perception questionnaires about metacognitive control before and after deferred revision. They were screen-recorded while revising their texts and a think-aloud protocol was followed. The verbalisations were transcribed and organised for analysis. Participants faced deferred revision of the essays differently in both languages. They began the revision mostly without planning it. Experts displayed greater confidence in their first versions and carried out fewer transformations. Intermediate PSTs showed the most dynamic mental representation of their texts in both languages. The level of perception of the use of metacognitive strategies on the aspects of the task varied depending on EFL proficiency of participants. Differences were found in reading aloud their EFL texts and the use of the translation and L1 depending on the EFL proficiency. The results suggest the need to promote the development of strategies of metacognitive control in the PST education and all areas of curricula and languages at any educational stage through evidence-based teaching interventions.