1.1. Defining self-regulated learning

The purpose of this research timeline is to review salient studies on feedback and self-regulation in L2 writing. Self-regulation has long been prominent in the field of education (Hattie & Timperley, 2007*; Panadero, Reference Panadero2017). This is hardly surprising, as the goal of all education is to support students in becoming self-regulated, lifelong learners (Schunk & Greene, Reference Schunk and Greene2018), and feedback is a key tool for achieving that goal (Butler & Winne, 1995*; Nicol & Macfarlane-Dick, 2006*). The literature postulates that self-regulation is not a trait that students either possess or lack. Instead, it means using specific processes in learning, such as monitoring learning, setting proximal goals, and using time efficiently (Zimmerman, Reference Zimmerman2002). More specifically, according to Zimmerman*Footnote ¹ (2000, p. 14), self-regulated learning (SRL) refers to “self-generated thoughts, feelings, and actions that are planned and cyclically adapted to the attainment of personal goals.” This sociocognitive view of SRL conceptualizes SRL as a self-directed process in which mental abilities are transformed into academic skills (Zimmerman, Reference Zimmerman2002).

A widely cited model by Zimmerman (2000*, 2002) posits that this SRL process comprises three phases: forethought (task analysis and self-motivational beliefs), performance (self-control and self-observation), and self-reflection (self-judgment and self-reaction). During the forethought phase, students set goals and plan what strategies they need to achieve their goals. During the performance phase, students deploy the strategies chosen in the previous phase and monitor their progress. Lastly, students assess the performance and make decisions regarding future tasks in the reflection phase (Zimmerman, 2000*, Zimmerman, Reference Zimmerman2002). Feedback plays a crucial role in this process, as input from teachers and peers on previous tasks and performances helps students adjust their approach to current tasks (Zimmerman, 2000*). Zimmerman’s view of SRL concurs with those of other SRL models, which typically consist of a preparatory phase, a performance phase, and an appraisal phase (Panadero, Reference Panadero2017).

Research on SRL has traditionally evolved within educational psychology, with influential frameworks proposed by Zimmerman (2000*), Butler and Winne (1995*), and Hattie and Timperley (2007*). However, these models were primarily developed outside the domain of second language learning and writing. This timeline bridges that gap by synthesizing how principles from educational psychology translate into L2 pedagogy and writing assessment. Specifically, we highlight how constructs such as feedback cycles, goal-setting, and metacognitive monitoring – originally conceptualized in general SRL research – have been adapted and extended within applied linguistics to inform practices that support L2 writers.

In the literature, several scholars have shown that SRL is conducive to language learning (Teng & Zhang, Reference Teng and Zhang2022; Zhang & Zou, Reference Zhang and Zou2024). More specifically, many studies have examined the relationship between SRL and feedback in language learning. For example, self-regulated writers engage with feedback, which enhances L2 writing performance (Jiang & Ju, 2022*; Yang & Zhang, 2023*). Being a self-regulated language learner also contributes to a feedback-seeking orientation in L2 writing (Xu, 2021*, 2022*). To support SRL in an L2 context necessitates a dynamic interplay between the learning environment, tasks, strategies, and learners (Teng et al., Reference Teng, Wei and Zhang2024; Zhang & Zou, Reference Zhang and Zou2024). This can be achieved, for example, by creating a classroom environment where peers and teachers offer students scaffolding and mediation, thereby facilitating the students’ journey to becoming autonomous, self-regulated writers (Teng et al., Reference Teng, Wei and Zhang2024). How students perceive their self-regulation is important in language learning; according to Vattøy and Smith (Reference Vattøy and Smith2019), if students feel they are capable of self-regulating their learning, they are more inclined to find English teachers’ feedback useful. Despite the centrality of SRL in education, the concept has not been adequately deployed in foreign language learning and applied linguistics (Teng & Zhang, Reference Teng and Zhang2022).

A concept closely related to SRL is metacognition. Metacognition is a prerequisite to SRL, and refers to “the awareness of and knowledge about one’s own thinking” (Zimmerman, Reference Zimmerman2002, p. 65). SRL and metacognition have distinct developmental trajectories (Dinsmore et al., Reference Dinsmore, Alexander and Loughlin2008). A major difference between SRL and metacognition lies in monitoring and controlling; metacognition stresses control and monitoring of cognition, while SRL emphasizes control and monitoring of motivation, behavior, and cognition (Dinsmore et al., Reference Dinsmore, Alexander and Loughlin2008). Moreover, models of SRL consider metacognition, while metacognition and self-regulation have “developed in parallel with little observable cross-fertilization” (Dinsmore et al., Reference Dinsmore, Alexander and Loughlin2008, p. 394). Despite subtle differences, the definitions and scopes of metacognition and SRL are interrelated and partially overlap (Zhang & Zhang, Reference Zhang, Zhang and Gao2019).

Although Zimmerman’s (2000*) model dominates the research on SRL in the context of language learning (Teng & Zhang, Reference Teng and Zhang2022), other conceptualizations have also been implemented. Thus, it is imperative to understand the various conceptualizations of SRL. Fox and Riconscente (Reference Fox and Riconscente2008) provide a detailed discussion of how metacognition and self-regulation are viewed by James, Piaget, and Vygotsky. Vygotsky stressed language-based social interactions and formal schooling, while Piaget emphasized peer-level interactions and suggested that self-regulation operates in two realms: intellect (intention, direction of thoughts, and problem-solving actions) and affect (control and will). From James’s perspective, self-regulation refers to habit and will; more specifically, to activities that are automated (habit) or require effort (will) (Fox & Riconscente, Reference Fox and Riconscente2008). Furthermore, from a social cognitive perspective, Schunk and Zimmerman (Reference Schunk and Zimmerman2007) describe four partially overlapping levels of SRL development; observational and emulative (focus on social factors) as well as self-controlled and self-regulated (focus on the learner). Although SRL has been theorized differently, various models and theories also include similarities, such as the importance of interaction (Clark, 2012*; Fox & Riconscente, Reference Fox and Riconscente2008). Moreover, given the pivotal role of self-regulation in sociocultural theory and the emerging research on learner agency from a neo-Vygotskian perspective, it can be argued that self-regulation is also fairly close to learner agency (Dufva & Aro, Reference Dufva, Aro, Deters, Gao, Miller and Vitanova2014). Although Dufva and Aro’s (Reference Dufva, Aro, Deters, Gao, Miller and Vitanova2014) approach was in many ways consistent with the neo-Vygotskian conceptualization of agency, their approach was dialogical, which deviates from the neo-Vygotskian approaches to understanding and studying agency. By integrating both sociocognitive and sociocultural perspectives, this research timeline acknowledges the theoretical diversity underpinning SRL and positions the selected studies within a broader conceptual landscape.

1.2. Formative assessment and SRL

Feedback is a central practice within formative assessment. This study adopts the definition of formative assessment proposed by Black and Wiliam (Black & Wiliam, Reference Black and Wiliam2009, p. 9). According to them, “Practice in a classroom is formative to the extent that evidence about student achievement is elicited, interpreted, and used by teachers, learners, or their peers, to make decisions about the next steps in instruction that are likely to be better, or better founded, than the decisions they would have taken in the absence of the evidence that was elicited.” In other words, formative assessment should not be conflated with teacher feedback, peer feedback, online feedback, or self-assessment, as the mere use of these practices does not in itself constitute formative assessment (Black & Wiliam, Reference Black and Wiliam2018). Ultimately, formative assessment, also referred to as Assessment for Learning, relates to optimizing and making adjustments to teaching and learning (Black & Wiliam, Reference Black and Wiliam2009; Wolsey et al., Reference Wolsey, Lenski and Grisham2020). To achieve this, several conditions must be satisfied, such as providing clear learning goals and assessment criteria as well as using multifaceted tasks in which students can show their proficiency (Black & Wiliam, Reference Black and Wiliam2018; Wolsey et al., Reference Wolsey, Lenski and Grisham2020). Moreover, if assessment is conceptualized as formative, it needs to be continuous throughout the learning cycle; if exams and tests are used continuously without using them to enhance students’ learning, the assessment practice is not formative (Davison & Leung, Reference Davison and Leung2009). Similarly, the use of alternative assessments, such as portfolios or oral exams, does not automatically refer to formative assessment (Davison & Leung, Reference Davison and Leung2009).

In line with Hattie and Timperley’s (2007*) framework, this research timeline considers the four levels of feedback addressed across the reviewed studies: task-level feedback (focused on accuracy and completion), process-level feedback (supporting writing strategies), self-regulation-level feedback (enhancing learner monitoring and control), and self-level feedback (personal judgments of ability). This distinction not only clarifies how feedback functions across studies but also helps identify trends in research emphasis, particularly the growing focus on process-oriented and self-regulatory forms of feedback.

Scholars argue that feedback is an effective means of enhancing SRL in educational settings (Paris & Paris, Reference Paris and Paris2001). Thus, it is unsurprising that several studies have examined feedback and SRL across various levels of education, employing diverse methodological approaches, such as qualitative, quantitative, and mixed methods (Schunk & Greene, Reference Schunk and Greene2018). These studies highlight that teacher feedback, peer feedback, and self-feedback (self-assessment) encourage students to take a proactive role in learning, thereby fostering their development as self-regulated learners (Nicol & Macfarlane-Dick, 2006*). More specifically, if feedback is targeted at the self-regulatory level of learning, students will be equipped to monitor their learning and develop as autonomous learners (Hattie & Timperley, 2007*). Feedback also strengthens self-efficacy, a key component of SRL (Nicol & Macfarlane-Dick, 2006*; Lee & Evans, 2019*). Moreover, earlier studies indicate that when students engage with feedback from various sources and act upon it, they simultaneously become more self-regulated (Hattie & Timperley, 2007*; Xiang et al., Reference Xiang, Yuan and Yu2022). However, not all students are receptive to feedback on their writing, which often depends on their beliefs about their ability to accomplish writing tasks (Zumbrunn et al., Reference Zumbrunn, Marrs and Mewborn2016). Another explanation may lie in earlier experiences; students with positive feedback perceptions exhibit higher writing self-regulation aptitude compared to students with negative feedback perceptions (Ekholm et al., Reference Ekholm, Zumbrunn and Conklin2015). To enhance students’ positive views on feedback, it is important to discuss with students what the relevance of feedback for writing is and what feelings feedback evokes in them (Ekholm et al., Reference Ekholm, Zumbrunn and Conklin2015; Zumbrunn et al., Reference Zumbrunn, Marrs and Mewborn2016).

In addition to enhancing SRL, feedback supports the development of students’ writing efficiency and quality (Xiang et al., Reference Xiang, Yuan and Yu2022; Yang et al., 2022*). In writing, self-regulation is important (Graham et al., Reference Graham, Harris, MacArthur, Santangelo, Schunk and Greene2018; Santangelo et al., Reference Santangelo, Harris, Graham, MacArthur, Graham and Fitzgerald2016), particularly when revising a text (Nicol & Macfarlane-Dick, 2006*). Skilled writers invest additional time in making revisions and make more substantive revisions than do less-skilled writers (Graham, Reference Graham, Alexander and Winne2006). Research indicates that when students use criteria such as rubrics to self-assess their writing, their work quality often improves (Santangelo et al., Reference Santangelo, Harris, Graham, MacArthur, Graham and Fitzgerald2016). Therefore, in many educational contexts, teachers are urged to support their students in becoming self-regulated writers (Santangelo et al., Reference Santangelo, Harris, Graham, MacArthur, Graham and Fitzgerald2016).

1.3. Research timeline on feedback and SRL

In light of the discussion mentioned earlier, we propose a research timeline that examines the interplay between formative assessment and self-regulation in L2 writing, incorporating the key aspects identified. Specifically, we will analyze how formative assessment – through teacher feedback, peer feedback, self-assessment, and AI-driven feedback – facilitates students’ development as self-regulated L2 writers. Reviewing the existing research on feedback and writing in self-regulated L2 learning is essential, as review studies on this topic remain scarce (Zhang & Zou, Reference Zhang and Zou2024). Thus, this research timeline contributes to the existing literature by detailing trends and gaps in earlier studies.

We identified 51 studies (1995–2026) that provide insights into how teacher and peer feedback, self-assessment, and AI-driven feedback contribute to self-regulation development in writing. The studies examine the interconnectedness of feedback and SRL in L2 writing. The studies span educational levels from primary school to higher education. Regarding the selection process, several criteria were used to select the studies. They were published in English in reputable journals or books. The papers provided theoretical and practical insight from the perspectives of the seminal papers reviewed in this timeline, such as Butler and Winne (1995*), Zimmerman (2000*), and Hattie and Timperley (2007*). Moreover, the papers explored formative assessment practices (teacher, peer and AI-driven feedback as well as self-assessment) and self-regulation in L2 writing. Empirical papers were conducted in the context of English (L2) teaching in primary, secondary, or higher education.

The categorization adopted in this review is grounded in established SRL-oriented frameworks in writing pedagogy (e.g., Lam, 2014*; Nicol & Macfarlane-Dick, 2006*; Paris & Paris, Reference Paris and Paris2001) and shaped our selection of studies. Explicitly delineating these categories enhances transparency and allows readers to trace thematic patterns across the timeline.

The studies reviewed highlight the complexity of research on feedback and SRL in English language teaching. For example, a variety of research designs have been employed, with several empirical studies collecting data through longitudinal, intervention-based approaches. The lengths of the longitudinal studies varied considerably, from a few weeks to several months. Intervention studies often included training sessions on various aspects of feedback and SRL. Qualitative, quantitative, and mixed methods approaches have been used fairly evenly, with minor variations. For example, quantitative and mixed-methods studies seemed to be slightly more prevalent than qualitative studies. In recent years, some trends are visible in the dataset, such as validating SRL questionnaires and investigating mindsets and AI-driven feedback. Data collection was typically on writing courses, where students produced diverse texts, such as argumentative, expository, and narrative essays. However, most studies focus on higher education contexts, with comparatively limited research on primary and secondary school settings. Although the emerging role of AI-driven feedback technologies constitutes a significant methodological innovation, it remains underexplored relative to traditional teacher and peer feedback practices. Furthermore, despite growing recognition of the importance of inclusive assessment practices, learners with special educational needs remain underrepresented in SRL and L2 writing research. Writing is a cognitively demanding skill, making SRL strategies even more critical for learners with special needs. Addressing these gaps presents opportunities for advancing pedagogical innovation, informing institutional policies, and supporting equitable learning outcomes in diverse L2 contexts.

On the basis of the above considerations, we will use the following categories to group the studies:

1. Level of education

1.1 higher education

1.2 secondary education

1.3 primary education

2. Types of feedback

2.1 teacher feedback

2.2 peer feedback

2.3 self-assessment

2.4 AI-driven feedback

3. Methodological orientations

3.1 qualitative

3.2 quantitative

3.3 mixed methods

4. Main implications for SRL in formative assessment of L2 writing

4.1 theoretical implications

4.2 pedagogical implications

Regarding the fourth category, it is important to note that all the papers of the research timeline have both theoretical and pedagogical implications. However, we have grouped the papers according to their primary contribution to the field.