THIRTY-TWO YEARS OF CORRECTIVE FEEDBACK IN LANGUAGE TEACHING AND LEARNING: A REVIEW OF THE LITERATURE

Corrective feedback has been one of the controversial topics in second language acquisition, L2 learning, and teaching. Corrective feedback could be in written or oral forms. Much has been written and published on this debatable topic. The purpose of the current study is to review the most cited published articles from 1990 to 2022. Reviewing research on CF has several benefits. First, researchers who are interested in investigating CF gain access to a synthesized and chronological review of the topic, and it provides insights to conduct future research on CF from different perspectives and theories. Secondly, L2 teachers can gain a clear understanding of CF, and learn about different methods and strategies for providing it to their learners alongside the results and conclusions from the previous studies. Finally, it is a ready-made review for those who cannot obtain the available studies on CF, and they have the access to refer to the previous studies once they conduct further research on CF.


The purpose
As an EFL learner and now EFL L2 university instructor, I have got interested in investigating corrective feedback (CF) while I was working on my MA project in the USA. After returning to my teaching career, I taught research writing class and communication (i.e., reading and writing). There, my interest in learning more about CF was increased. I wanted to learn more about how much has been published on CF and what results and conclusions other researchers have reached and found. As I was reading, I thought writing a paper on reviewing previous studies can benefit my knowledge on the topic and those who are also interested in investigating CF. I decided to collect those studies which have been cited the most from the last thirtytwo years. After collecting the studies, I started reviewing them based on some criteria: the context, the participants, the used tools to collect data, the findings, and conclusions. In the final section, I have briefly mentioned the conclusions I have reached and proposed some suggestions for future directions on CF.

Review of selected articles
In this article, I thoroughly review the existing literature on WCF in the last thirty-two years up to 2022. The main reason for providing and reviewing this long history of WCF in the research world is to obtain a very deep and comprehensible understanding. Another reason for looking into previous research on WCF is to have a clear direction for future studies. Therefore, I have decided to divide this literature based on decades.
Searching for the available research on WCF, one can collect research on WCF even before the 1990s, but as time passes, as an EFL/ESL learner, and now, as a second language acquisition researcher, reviewing thirty-two years of research on WCF makes the current study one of the few studies to map out what has been done and found on WCF.

Theory Application Corrective feedback: From 1990 to 2000
The ultimate goal of L2 teachers is to help to improve their learners' writing skills and accuracy. Thus, both researchers and teachers have been researching to achieve this goal. Here, the first reviewed article was on types of written feedback (WF) on the development of second language writing skills (Kepner, 1991). In her study, the researcher conducted the study on Spanish intermediate-level participants. The participants received two types of feedback on their written journal entries for eight assignments; message comments and surface error corrections. The results showed that providing the mentioned types of CF neither improved nor enhanced the quality of L2 students' writing. After some years, Truscott (1996) conducted a study entitled "The Case Against Grammar Correction in L2 Writing Classes". He argued that grammar correction has to be abandoned. Although he stated that grammar accuracy is important, he provided evidence from previous research that CF has little or no effect on enhancing learners' writing abilities. Truscott suggested that instead of WCF, "accuracy is improved through extensive experience with the target languageexperience in reading and writing" (p. 34). However, after Truscott's (1996) case against CF, Ferris (1999) carried out a study as a response to Truscott (1996). She responded to Truscott's statement that "grammar correction has no place in writing courses and should be abandoned" (1996, p. 328) is premature and overly strong" (p.2). She also provided evidence from other studies (e.g., Ferris, 1995) that CF can and does help some students' writing abilities. In the study by Ferris (1995), the researcher found that in a semester-long ESL composition class, the participants were taught how to identify, prioritize, and correct their errors. The results showed that most of the learners were able to correct their errors successfully. After the published article by Ferris (1999), once again, Truscott (1999) published another article in response to Ferris (1999) because she had rejected Truscott's case against grammar correction. Truscott put forward his responses and argument against what Ferris (1999) had said in her research. Based on the above-reviewed studies, some researchers are advocates for CWF; whereas, some others such as Truscott strongly against it.
After investigating WCF from one narrow lens without basing it on any theoretical frameworks, Nassaji and Swain (2000) stepped out further and conducted a study on corrective feedback in L2 from A Vygotskian Perspective. They investigated whether negotiated help is more effective for the learner's zone of proximal development (ZPD) or randomly provided help. The participants of the study were EFL Korean learners. The results showed that explicit CF was more effective when the learners were provided with random help. In addition to that, the results also revealed that the help that is provided within the ZDP was more effective than the help provided randomly.

Corrective feedback: From 2001 to 2010
After WCF had been investigated to some extent, from 2001 to 2010, WCF and its types have been researched extensively. Ferris and Roberts (2001) attempted to examine the degree how which explicit WCF should be given in L2 writing classes. The study was important because research on types of WCF was scarce. In their study, 72 university ESL students were observed to investigate their self-edit. The results showed that the two groups who received CF significantly outperformed the no-feedback group. Regarding, how much the feedback has to be explicit, two groups received "codes" and the other group "no-codes", the results showed that there were no significant differences among the groups. Hence, we can see that CF worked with ESL students when the types of provided CF were selected.
In another research, Havranek (2002) examined when CF is most likely to succeed. He researched 207 learners with different proficiency levels. The results showed that some factors affect the success of CF, and among those are both situational and linguistic factors play an important role, but one of the most important factors is learners' contribution to the correction sequence. Learners have to be able to correct the error once CF is provided with it.
Since their responses to each other on the efficacy of WCF, Truscott, and Ferris have conducted more research on the same topic. In 2004, Ferris wrote research on CF and the debates related to it and its current and future position of CF. In her study, Ferris outlined the available research on CF and how much they help learners to improve their writing accuracy. In the same study, she believed that more longitudinal studies were necessary because the ones she reviewed were not adequately provided evidence of the usefulness of CF.
From 2007 to 2009, several studies have been selected such as (Bitchener, 2008;Bitchener & Knoch, 2008;Bitchener & Knoch, 2009a;Bitchener & Knoch, 2009b;Ellis, 2008;Ellis et al., 2008;Ellis, 2009;Truscott, 2007;Truscott & Hsu, 2008;Sheen, 2007). In each of these studies, the focus is on the effectiveness of WCF to improve learners' writing abilities and the types of CF that teachers would provide. For example, in the study by Bitchener (2008), he attempted to investigate and gather evidence in support of WCF. After the participants received different types of CF, the results showed that those who received CF had their accuracy outperformed the controlled group which did not receive any CF. Furthermore, Bitchener and Knoch (2008) conducted another study on the value of a focused approach to WCF. Once again, those students who received WCF performed better than those who did not. This result gave evidence to support the claim (Bitchener, 2008) that WCF can be effective in improving learners' writing accuracy. As mentioned before, the studies from 2007 to 2009 have mostly investigated the types of WCF. For instance, such studies by Ellis et al. (2008), Sheen (2007), Bitchener & Knoch (2009a), and Bitchener and Knoch (2009b), these studies have either focused on direct and focused and unfocused WCF or the contribution of WCF to language development. The results of these studies showed a positive attitude towards WCF as a means to be used for writing accuracy among learners.
On the other hand, two important studies were conducted by Ellis (2008;2009); one was on the typology of WCF, and the latter one was on the relationship between CF and teacher development. In his first study Ellis (2008) (5) what is the best timing for CF (immediate or delayed)" (p.1) and some pedagogical usages of CF in the classroom. Interestingly, the last study that I am reviewing here is by Truscott (2007) and Truscott and Hsu (2008). Although all the above-mentioned studies have confirmed that CF has benefits for learners' accuracy improvement, the study by Truscott, (2007) indicated that "We can be 95% confident that if it has any actual benefits, they are very small" (p.255). In addition to that, Truscott and Hsu (2008) conducted their study to investigate the benefits of CF on learning. The participants were divided into two groups; the first group wrote an in-class essay and revised it in the next class based on the provided feedback (i.e., underlining the errors), but the second group did not receive CF. After the collected data was analyzed, the results showed accuracy improvement. Later, the participants were asked to write a new narrative after one week. The results revealed no significant difference between the two groups; rather, they were identical. Thus, before moving to review further studies, further investigation is necessary to dive more into the two-fold of CF: as being effective or not effective.
In an extensive body of research written and published in 2010, twelve of the most cited studies are reviewed in the current literature. Amrhein and Nassaji (2010) investigated the preferences of students and teachers towards WCF. Both qualitative and quantitative data were collected. The most popular choice (93.9%) among the students and the teachers was Mark all errors. The results show that there was a significant difference in the amount of CF between the students and the teachers. Most of the teachers thought it is better to mark only the errors that interfere with communication. While the students preferred that teachers mark all major errors.
Regarding the types of feedback, the students preferred explicit CF; that is; teachers correct their students' errors. On the whole, both teachers and students believed that WCF is a learning tool.
Not all the studies regarding CF in 2010 were practical ones. For example, Ferris (2010) published an article reviewing L2 writing research and CF in second language acquisition (SLA), and its Intersections and practical applications. In her article, Ferris (2010) reviewed previous studies on CF with a section on direct and indirect CF. Furthermore; she tackled how CF can be reformed and provided based on a variety of methods.
One of the important studies that were published in 2010 was by Bitchener and Knoch (2010). In their study, the researchers attempted to investigate if written corrective feedback (WCF) can raise the accuracy level of advanced L2 writers. The participants received three types of WCF: linguistic explanation, the indirect circling of errors, written meta-linguistic feedback, and oral form-focused instruction. The results showed significant differences between the treatment groups. They improved their accuracy level after they received WCF. Furthermore, Santos et al. (2010) investigated two types of WCF: Reformulation vs. Error Correction. The conclusions showed that there was a positive effect of WCF on both noticing and uptake. In a similar study by Storch and Wigglesworth (2010), the researchers attempted to investigate learners' processing, uptake, and retention of CF in writing. The participants were asked to work in pairs and compose a text based on a given graph. Two types of feedback were provided; either "the form of reformulations (direct feedback) or editing symbols (indirect feedback)" (p.303). The findings suggested that learners' attitudes, beliefs, and goals alongside CF played an important role in improving writing accuracy.
Although much research had been published from the 1990s to 2010, more research was published and questioned the effectiveness of CF. For instance, Storch (2010) published an article entitled "Critical Feedback on Written Corrective Feedback Research". In her article, the researcher raised some questions such as "are researchers and L2 writing teachers now any wiser about the efficacy of WCF?" (p.29); the researcher reviewed some articles at that time and concluded that there is a necessity for more robust research. For example, for CF to be more effective, WCF has to be provided in more authentic classrooms where CF can incorporate it with the instructional program.
Observations from previous studies indicate that researchers have been researching CF (either written corrective feedback or oral corrective feedback) to provide more effective approaches for giving WCF and enhancing learners' writing abilities or oral communication skills. In three studies (e.g., Evans, Hartshorn, & Allen Tuioti, 2010;, the researchers examined the nature of WCF. For example, Evans, Hartshorn, and Allen Tuioti (2010) attempted to investigate practitioners' perspectives on WCF. The results presented that WCF is commonly practised by experienced and well-educated practitioners. Furthermore,  investigated the effects of dynamic CF on learners' writing accuracy; opposite the results of some previous studies, the results showed that writing fluency and writing complexity were not affected by WCF, but the writing accuracy of the participants was significantly improved. Finally, in the study by Evans, Hartshorn, McCollum, et al. (2010), the researchers provided L2 writing teachers with a paradigm for understanding the WCF. The researchers have attempted to contextualize WCF and illustrated three variables to gain a better understanding of WCF, and they are learner variables (i.e., motivation, learning style, goal, and L1), situational variables (i.e., teacher, physical environment, and socioeconomic conditions), and methodological variables (i.e., instructional design, what is taught, and how it is taught).
Although much research in 2010 was conducted to investigate written corrective feedback and its types, three studies on oral corrective feedback (OCF) were carried out (e.g., Ellis, 2010;Sheen, 2010a;Sheen, 2010b). Ellis (2010) did not investigate the effectiveness of OCF; instead, he proposed a framework based on the previous studies on OCF. Whereas, Sheen (2010a-2010b) investigated whether there are any differences between oral corrective feedback and written corrective feedback on learners' accurate use of English articles. The results showed that all the groups which received CF significantly performed better than the controlled groups. In her study (Sheen, 2010b); the researcher examined the role of oral and written corrective feedback in SLA. She mentioned several theories such as cognitive, sociocultural and psychological theories of SLA support CF. Furthermore, she presented those studies which support the effectiveness of oral CF in L2 learning. Finally, she dedicated a section to talk about the types of CF which are the most effective. She reviewed several articles in that regard with their results and conclusions.

Corrective feedback: From 2011 to 2015
Although much research was conducted from 1990 to 2010, from 2011 to 2015 a massive body of research on corrective feedback was published, and I am reviewing some of the most cited ones in this review of the literature. Chu (2011) examined teachers' CF on college students' oral accuracy, and the results showed that CF had positive effects on learners' oral accuracy improvement. In another study by Ferris et al. (2013) on WCF for individual L2 writers, the participants were asked to write four essays and revise them after receiving WCF (i.e., they were given focused WCF, revision, and one-to-one discussion about errors). The conclusions suggested that teachers have to use more fined-tuned approaches to give WCF. Furthermore, Ferris (2012) published an important study on WCF in L2 acquisition. In her study, she started with the definition of WCF and provided a timeline of a historical overview of WCF. She, furthermore, provided a table of references regarding students' errors and teachers' feedback from 1930 to 2012. Afterwards, more studies were conducted on CF and its types. For example, Farrokhi and Sattarpour (2012) attempted to examine the effects of direct WCF on learners' grammatical accuracy. The results revealed that focused WCF was more effective than unfocused WCF. Similarly, Marzban and Arabahmadi (2013) investigated dynamic WCF on learners' accuracy, fluency, and complexity. The results showed that there was a significant difference between the treatment groups and controlled groups, but regarding the effect of WCF on writing fluency, the treatment group appeared not to be affected by WCF. Regarding the learners' writing complexity, the results showed that their writing complexity was affected but not significantly.
Research on CF is circulating mostly around certain topics such as its effects on learners' writing accuracy, and the effects of certain types of CF on written or oral skills. Several studies (e.g., (Alimohammadi & Nejadansari, 2014;Ebadi, 2014;Fazilatfar et al., 2014;Hosseiny, 2014;Jokar & Soyoof, 2014;Shintani & Ellis, 2013) investigated focused and unfocused (i.e., direct or indirect) on learners' writing accuracy. The results showed that metalinguistic explanation improved explicit knowledge, and the participants' syntactic and lexical complexity were improved. The conclusions also revealed that explicit CF was more effective to improvement in grammatical levels. In parallel, Stefanou and Révész (2015) investigated the effectiveness of direct CF in acquiring articles for generic and specific plural references. The results showed that there is an advantage to receiving direct CF. Furthermore, Frear and Chiu (2015) examined the effect of focused and unfocused CF on writing accuracy. The participants were EFL University Chinese learners of English in the Taiwanese context. The results showed that a single episode of indirect feedback is insignificant. In a similar study, Kang and Han (2015) attempted to examine the effects of WCF on L2 writing accuracy. The conclusions indicated that WCF directs learners to gain greater grammatical accuracy, but some other variables can affect this outcome such as learners' proficiency, the setting, and the genre of the writing task.
Much research has been written dealing with either written corrective feedback or oral corrective feedback, but the study by Sobhani and Tayebipour (2015) investigated the effects of both oral and WCF on learners' essay writing. The results presented that the three types of CF (i.e., oral: focused and unfocused, WCF: focused) were significantly effective; whereas, unfocused written CF was not that effective. In another study, Han and Hyland (2015) examined the effects of WCF and learners' engagement with WCF. The findings suggested that teachers have to understand learners' backgrounds at a deep level in terms of their beliefs and use WCF strategies carefully. Providing corrective feedback mostly has been investigated to find out if it is effective on learners' writing accuracy such as in paragraphs and essays, but Nguyen et al. (2015) examined CF to teach email requests. Data was collected from Vietnamese EFL learners. The results indicated that the treatment group outperformed the controlled group. Furthermore, those participants who received meta-pragmatic feedback significantly improved compared with the group which received direct feedback. In addition to that, Rummel and Bitchener (2015) investigated the effectiveness of WCF on Lao learners' beliefs and uptake. The study took seven weeks with advanced Vietnamese learners. The target grammatical feature was the simple past tense. The results indicated that the participants who received WCF improved in using the target structure significantly. Also, the results showed that learners' beliefs impacted the participants' linguistic accuracy because those who received their preferred WCF types were more successful than those who did not. So far, most previous studies on CF have been conducted with learners in classrooms, but an important study by Shintani (2015) dealt with the effects of computer-mediated synchronous and asynchronous direct corrective feedback on writing. The results showed that synchronous corrective feedback created an interactive writing process as the oral corrective feedback. Both synchronous corrective feedback and asynchronous corrective feedback assisted the learners to notice the errors, but self-correction was more effective and successful.
As can be observed, an abundance of research on corrective feedback (CF) in L2 writing has been published by prominent researchers in the field. Still, the fundamental question of to what extent WCF, oral CF, and types of CF enhance learners' writing fluency and accuracy has not been answered. Thus, more research has been conducted in different contexts. Therefore, it is necessary to review more studies on that controversial topic.

Corrective feedback: From 2016 to 2022
As observed, corrective feedback has been one of the controversial topics in L2 teaching. More research is published each year to investigate CF. Although CF has a very long history (i.e., more than five decades), more research has been published. Still, the conclusions and results from previous research indicate that more investigations are needed to investigate CF from different theoretical lenses. Atmaca, (2016) investigated teachers' and students' perceptions towards WCF in an EFL context. The results showed no significant differences between the groups, but the results from the open-ended questions revealed some differences. Furthermore, the conclusions suggested that teachers should explain their expectations from the beginning of the classes not to make misunderstandings between the teachers and the students. Most published studies on WCF are about the effectiveness of WCF and its types. For example, Aghajanloo et al. (2016) conducted their study to examine how much teachers' CF is effective in focusing on some types (i.e., focused direct CF, unfocused direct CF, focused indirect CF, and unfocused indirect CF). The results showed that those participants who received the CF outperformed those who did not. Regarding the effective types of CF, the results illustrated that unfocused directive CF has to be the focused type and be used as an important learning tool.
Although CF research is more focused on WCF, there are studies on oral corrective feedback (OCF). Yang (2016) investigated learners' preferences for WCF concerning their cultural backgrounds and their proficiency levels. The study was carried out in a Chinese EFL context. The learners were given OCF on "phonological, lexical, grammatical, and pragmatic errors" (p.75). The results suggested that metalinguistic, explicit CF and recast were the most preferred types of CF among the learners. Teachers try to utilize effective tools to help their learners to master their language skills. Communication is not apart from the important language skills. Therefore, corrective feedback can be given on any aspect of learning the target language. For instance, Tavakoli and Zarrinabadi (2016) attempted to investigate explicit and implicit CF on learners' willingness to communicate. The participants were Iranian EFL low-intermediate groups. The results indicated that explicit CF did not influence the participants'' willingness to communicate, but explicit CF increased the effects. On the whole, explicit CF increased learners' willingness to communicate and confidence.
Much research investigated learners' preferences for CF in different contexts. Researchers and L2 teachers believe that understanding learners' preferences can facilitate teachers in choosing the more effective types of CF upon providing it. Chen et al. (2016) published their study to investigate learners' preferences for WCF in China. Quantitative and qualitative data were collected from participants whose proficiency levels were intermediate, advanced-intermediate, and advanced. The results showed that error correction was favourable among the participants, and they wanted comments on content and grammar. Overall, the results suggested that WCF is also seen as a valuable learning tool in EFL contexts.
Although much research was published on the benefits of CF and its types, researchers attempt to investigate CF in different countries and contexts to have valid data to give more evidence either against providing CF or in favour of it. For example, Park et al. (2016) examined the benefits of indirect CF among Korean EFL learners. The participants' proficiency levels were beginner and intermediate, and the results were compared to learners' prior language exposure. The results showed that learners could self-correct more than a third of their errors. The results also suggested that language teachers have to consider individual learner differences. Similarly, Tangkiengsirisin and Kalra (2016) studied the perfections of Thai learners towards direct and indirect WCF. The results showed that the group which received direct CF improved significantly compared to the group which received indirect CF.
From the beginning of writing on CF, the focus has been on the utilization of CF to improve learners' L2 skills (i.e., more on writing and grammatical accuracy), very little has been written on how CF can be used for teacher development. In this regard, Lee et al. (2015) conducted a study on teachers' attempts at feedback innovations in the writing classroom. The study was carried out with two secondary teachers in Hong Kong. They participated in a writing teacher education course. The results showed that the teachers could not implement the CF fully due to some environmental factors from the school such as inadequate time, and they were not supported by the school.
Many factors affect language learning such as internal factors and external factors (Mahmoudi, 2015), but not much research was conducted on how factors such as writing anxiety and motivation affect learners' self-evaluation judgment of corrective feedback. In a study by Tsao et al. (2017), the researchers investigated how anxiety and motivation predict learners' judgment of corrective feedback. The results showed that learners were motivated to reach a higher level of their proficiency, and it also showed that (37%) of the participants showed anxiety when trying to learn how to write in English.
It is probably true that a bulk of research has been written on CF; therefore, reviewing published articles might make the reviewer see similar studies with the same topic but in different contexts. Sermsook et al. (2017) attempted to investigate how teachers' CF affects learners' grammatical improvement in an EFL context. They concluded that, based on previous studies, both direct and indirect CF either written or oral is beneficial for learners. Similarly, Kurzer (2017) explored dynamic written corrective feedback (DWCF) in improving learners' writing skills in a multilevel class. The results indicated that DWCF can be used as an effective pedagogical tool in writing classes to improve learners' linguistic accuracy. On the other hand, the results of the study by Tan and Manochphinyo (2017) indicated that for subject-verb agreement indirect CF was more effective than direct CF. As mentioned above, it is important to consider learners' preferences for CF and their beliefs before providing CF. This is also confirmed by the finding in the study by Han (2017). The conclusions suggested that when providing feedback, teachers should consider learners' beliefs because learners' engagement with CF is dependent on three main factors, namely, "person-related beliefs, task-related beliefs, and strategy-related beliefs" (p.9).
Not much research investigated the interaction between teachers' feedback and students' preferences, but the study by Irwin (2018) attempted to examine the types of relations between the teachers' CF and learners' preferences. The results showed that the teacher was the main source of addressing the learners' preferences. It also indicated that upon providing CF, the teacher was mostly the centre of the class, not the learners. The conclusions suggested that teachers must consider their learners' preferences when giving CF.
Researchers and L2 teachers have been investigating to find effective strategies to provide feedback; direct/focused and indirect/unfocused, comments, metalinguistic feedback, using codes, and many more. This way of research has been going on for more than four decades. Still, more and more research is published each month with either similar results/conclusions or vice versa. Several studies (e.g., Karim & Nassaji, 2018;Tang and Liu (2018;Benson & DeKeyser, 2018) attempted to investigate different types of feedback (I.e., indirect coded CF, direct and indirect comprehensive CF, metalinguistic CF) with L2 learners to check their writing accuracy. All in all, the results and conclusions revealed that learners' writing accuracy was improved significantly. Furthermore, Zheng and Yu (2018) examined WCF with lower-level Chinese learners, and how much they engage with the process. The results showed a positive engagement, but it does not help the learners to improve their writing accuracy. In a similar study, Han (2019) investigated some factors that help learners to engage with WCF. The analyzed data were collected from students' writing, verbal reports, interviews, field notes, and class documents. The results showed that the students perceived CF as an opportunity for the learning process and to enhance learners' engagement with WCF.
With circumstances changing, L2 teaching methodology might also change to be more suitable for learners and teaching atmospheres. In 2019 due to the spreading of COVID-19, most of the teaching classes were offered online or in a blended way. It was the time that Sarré et al. (2019) conducted a study to investigate the impact of different types of CF to enhance learners' writing accuracy in an experimental blended learning EFL course. The study was carried out in France. The participants were asked to write several pieces of writing with a controlled group with no given feedback. The participants received unfocused indirect feedback with metalinguistic comments. Computer assistance was also used. Based on analyzed data, the results showed that the groups which received the CF performed better than the group with no feedback.
Throughout the history of L2 teaching, much has been written on CF (i.e., written or oral). Researchers have been investigating the issues, effectiveness, benefits, the disadvantages of WCF in different countries and contexts. L2 teachers wonder how much CF has to be provided: more is better or less. Recently, Lee (2019) published a study entitled "Teacher written corrective feedback: Less is more". In her paper, the researcher argues that "more written corrective feedback is not better, but instead less is more" (p.1). The researcher argues that comprehensive written corrective feedback (CWCF) is problematic for both teachers and students. She believes that it takes too much time and hinders the teachers to focus on other important issues in writing, such as context, organization, and genre. Furthermore, providing feedback to a pile of students' writing papers affects the teacher emotionally and psychologically, and sometimes, teachers have to be rash to give feedback. The teacher might give illegible and inaccurate WCF. Then, what is worse than that when the students cannot understand teachers' feedback? Furthermore, giving too much feedback and selecting students' errors with red ink is overwhelming, and it confuses and discourages students to improve and utilize the given feedback. Therefore, based on the illustrated reasons, Lee (2019) believes that less feedback is more effective.
One of the main factors for providing CF is to increase learners' oral or written accuracy. Researchers are interested in investigating CF for the improvement of grammatical accuracy. Boggs (2019) and Kim and Emeliyanova (2019) attempted to investigate the role of CF in increasing learners' grammatical accuracy. The data was collected (i.e., survey and interviews, timed writing essays) among Korean EFL university learners in academic English writing classes, and intermediate-level English learners enrolled in an intensive English program in the USA. The findings showed that the groups which received the feedback increased in grammatical accuracy compared to the controlled group, but through giving metalinguistic reflections, the results revealed no significant grammatical accuracy among the treatment groups.
Research (e.g., Lee, 2019) showed that CF might affect teachers' and learners'' psychological and emotional aspects. Thus, it is crucial to understand how CF works on other parts of the mind such as the working memory of learners. In this regard, Li and Roshan (2019) conducted a study on the relationship between working memory and the effects of four types of WCF (i.e., direct corrective feedback, direct corrective feedback plus revision, metalinguistic explanation, and metalinguistic explanation plus revision). The participants were asked to write three writing tasks and took two working memory tests. The results showed that there is a relation between the complexities of working memory with the effects of metalinguistic corrective feedback, whereas; there was a negative predictor of the effect of direct feedback on short-term memory. In another study by Mao and Crosthwaite (2019), the researchers examined (mis)beliefs of teachers in practising WCF. The researchers collected data from five Chinese teachers to investigate how much their CF practice aligns with their beliefs. The results indicated that there was an alignment between giving WCF and teachers' beliefs. The participants also commented that they believed that they had given more indirect feedback in the class compared to the amount of feedback on learners' writing papers. They also expressed some other issues such as time constraints and the amount of workload they had misaligned with their beliefs towards WCF practice. However, the study by Eckstein et al. (2020) was conducted on the effects of dynamic WCF and feedback timing on graduate students. The researchers investigated how the given feedback impacts grammatical accuracy and lexical complexity. The results showed that neither providing timing feedback nor late feedback influences students'' writing accuracy, but timely feedback influenced the participants' writing fluency and writing complexity.
The controversy of providing CF is still a hot topic among researchers and L2 teachers. For that reason, conducting more research is a way to find more evidence on CF. Generally, learners receive feedback either too much or less, but how much it enhances their writing skills and accuracy. The crucial point is how much they use the feedback they receive from their teachers. Does asking them to work and use the given CF help them to reach improvement? In this regard, Ekanayaka and Ellis (2020) conducted a study to investigate the add-on effect of asking learners to revise upon receiving CF. The participants were EFL students in Sri Lanka. They were asked to compose three writing tasks. The results showed that the group which received CF improved their writing accuracy. On the whole, whether having a chance to revise the writing task or not, receiving feedback assists learners in improving their writing skills. An interesting study (e.g., Kartchava & Mohamed, 2020) was conducted to investigate the use of gestures in corrective feedback. The data was collected by observation and interview, and the participants were two English for academic purposes (EAP). After the data collection, the teachers watched short extracts from their lessons. They were asked about the reasons and motivations behind the use of their gestures in class. The results showed that EAP teachers actively used gestures while providing CF. Furthermore, the justification for using gestures was facilitating the role of CF in learning.
To review the published research from 2021 to 2022, I have tried to review the studies which have the most citations. For this purpose, a selective group of research has been reviewed. As observed above, CF has been researched extensively. Yet, more research investigating the same or little difference is available. Cao (2021) investigated the effects of WCF on young learners' writing accuracy. Once again, the participants were Chinese EFL learners. The participants received as they were asked to describe a picture and blank-filling task. WCF types were provided (i.e., direct corrective feedback and metalinguistic CF). The result showed a positive effect of CF on the learners, but the result was not significant in the delayed post-test. Regarding, the direct and metalinguistic CF, the results showed that they were statistically effective and significant in both tests. In a similar study, Bozorgian and Yazdani (2021) conducted their study on Iranian EFL learners. They received different types of WCF (i.e., Direct only, and direct with metalinguistic explanation). Overall, the results showed a positive effect of CF, and learners' writing accuracy was improved (see Cheng & Zhang, 2021). Furthermore, Mahmood (2021) investigated the effects of WCF and its types on university Kurdish EFL learners. They received explicit and implicit WCF. The results revealed that the participants were not fully aware of the benefits of WCF, and regarding the preferred types of WCF, the results showed that the participants were in favour of both types (i.e., explicit and implicit). On the other hand, Zhang and Hyland (2022) investigated the effects of three types of WCF, namely, automated, peer, and teacher feedback. The researchers wanted to examine the level of engagement. The results showed that the students actively engaged with the three types of CF; the provided CF encouraged the learners to be more motivated on revising their writing tasks.

Discussion, limitations, and future directions
As read from the existing literature throughout decades, corrective feedback has always occupied a corner in the area of L2 research and teaching. To answer the first inquiry; where have we reached? I believe that we as researchers and L2 instructors are in a loop, that is, like a pendulum; we are coming back and forth. Research has investigated the effectiveness of CF and its types available in the 1990s, 2000s, and up to 2020s. Results showed that CF can be beneficial or provide no significant effect on L2 learners, no matter what or where the context is. Therefore, I firmly believe that not much has changed or reformed from the beginning of researching CF to the current time.
When I say limitations, I do not mean limitations for not being able to find previous articles or sources. I mean by "limitations" those factors that affect researching CF and the factors which influence the results and conclusions of CF. Several factors can be mentioned such as time constraints, teachers' lack of knowledge on the usages of CF, focusing CF from one angle and not looking into it from different perspectives such as looking at a diamond from all sides, the school or institute authority on teachers to demand them completing the selected curriculum and class materials, a large number of students/learners in classrooms, psychological and sociological factors on both teachers and learners, and lack of teaching facilities in some contexts, especially, EFL contexts. I believe that considering the above factors while conducting and investigating CF can provide more accurate, reliable, and beneficial outcomes for researchers and L2 teachers on a variety of more effective strategies upon giving CF.
Conducting research on CF following previous titles might not take CF scope into any different circumstances. I believe CF has to be incorporated into teaching material alongside textbooks and supplemental materials. What I mean is to train L2 teachers and learners to understand the hidden power of CF and how CF can be used as a "game-changer" in teaching L2 because games such as chess have been the main analogy for human language (Aitchison, 2004). Therefore, future research could be on examining CF from the perspective of a variety of theories such as sociocultural theory, activity theory, cognitive theory, or to carve out a new model for providing corrective feedback (i.e., currently, I am working on one). Furthermore, CF is not bounded to one or two language skills (such as writing and grammatical accuracy). All the macro and micro-skills (i.e., pronunciation, listening, reading, even teacher education and teacher development) of the target language can be incorporated with CF and investigate how CF works on them.