Abstract
Previous research on oral corrective feedback (OCF) has primarily focused on teachers’ cognitions and practices in pre-service and private-sector contexts, with limited attention to in-service teachers, particularly in state-sector settings. This study investigates Iranian in-service English as a foreign language (EFL) teachers’ cognitions and reported practices concerning OCF in state-sector schools. Drawing on Borg’s teacher cognition framework and Lyster and Ranta’s typology of feedback types, the study focuses on five key aspects of OCF: source, provider, linguistic target, timing, and feedback type. An explanatory sequential mixed-methods design was employed: 204 teachers completed a questionnaire, followed by semi-structured interviews with 10 participants. The findings indicate that teaching experience, particularly through critical incidents, is the most influential source of OCF cognitions and behaviors. Crucially, affective factors, such as student anxiety and motivation, were found to override cognitive and contextual factors in shaping teachers’ feedback decisions. The study highlights the interplay of cognition, emotion, and context in OCF decision-making and calls for teacher education programs to address not only pedagogical knowledge but also the emotional dimensions of teaching. The article concludes by proposing a model of teachers’ OCF cognitions and reported practices, foregrounding affective factors as an influential source of teachers’ OCF cognitions and practice.
I Introduction
Over the past two decades, corrective feedback (CF) has garnered increasing attention in second language acquisition (SLA) research (for reviews, see Li & Vuono, 2019; Nassaji & Kartchava, 2017). CF is defined as ‘the feedback that learners receive on the linguistic errors they make in their oral or written production in a second language (L2)’ (Sheen & Ellis, 2011, p. 593). It is widely regarded as a crucial aspect of language pedagogy both for SLA researchers and teachers (Ellis, 2017). As an integral component of classroom interaction (Ellis, 2017; Nassaji, 2015; Nassaji & Kartchava, 2020), CF helps learners notice the gaps between their output and target language norms, prompting them to re-evaluate and refine their linguistic hypotheses (Doughty & Varela, 1998). Nonetheless, the effectiveness of oral CF (OCF) – the focus of this study – varies according to several mediating factors, including feedback type, the linguistic target, individual learner differences, and contextual variables (Nassaji & Kartchava, 2020).
Among these factors, the teacher – often the ‘principal CF provider’ (Gurzynski-Weiss, 2016, p. 258) – plays a central role in the effectiveness of feedback. One key influence on teachers’ OCF decisions is their cognition (Li, 2017; Nassaji et al., 2023). Teacher cognition is defined as ‘the complex, practically-oriented, personalized, and context-sensitive network of knowledge, thoughts, and beliefs language teachers draw on in their work’ (Borg, 2015, p. 321). It informs many classroom decisions, including whether, when, and how to correct students’ errors. Studying language teacher cognition is thus important as it ‘creates a space to show how language teachers process various knowledge, input, and contextual conditions cognitively to make pedagogical decisions’ (Guo et al., 2019, p. 136).
A strand of previous research has compared teachers’ and learners’ views on OCF (e.g., Roothooft & Breeze, 2016; Schulz, 1996, 2001; Yoshida, 2008, 2010), while other studies have focused on the OCF cognitions and practices of pre-service teachers (Agudo, 2014; Kartchava et al., 2020), university instructors (Gurzynski-Weiss, 2016; Mori, 2002; Yüksel et al., 2021), and English as a second language teachers in intensive programs (e.g., Kamiya, 2016). These studies have significantly contributed to the field, as one of the main aims of teacher cognition research is to describe unobservable dimensions of teachers’ work (Borg, 2019) in various contexts (Kubanyiova & Feryok, 2015).
More recent research has underscored the complex and dynamic nature of teachers’ OCF cognitions and practices, shaped by factors such as context (Ha & Murray, 2023), professional training (Ha & Murray, 2021), and learner expectations (Ha, 2023a). For instance, Ha and Nguyen (2021) identified a mismatch between teachers’ and students’ beliefs about CF providers, attributing this to contextual influences and student expectations. Ha and Murray (2023) showed that teaching contexts affect teachers’ prioritization of error types, with Vietnamese EFL teachers giving greater importance to pronunciation feedback. Similarly, Ha (2023b) found that lesson focus significantly influences CF choices, reinforcing the situated and adaptive nature of OCF decisions.
Despite these valuable contributions, notable gaps remain, particularly in underrepresented contexts such as Iran. Existing research in the Iranian context predominantly focuses on private language institutes (e.g., Gholami, 2021; Nassaji et al., 2023; Rahimi & Zhang, 2015), overlooking the distinct challenges and practices of state-sector teachers. This gap is notable for three reasons:
Public schools are the primary providers of English education for the majority of Iranian students, especially in rural and economically disadvantaged areas where private language institutes are either unavailable or unaffordable.
Significant differences exist between private- and state-sector teachers in terms of teacher qualifications, teaching methodologies, and teaching materials (Nejadghanbar, 2021; Sadeghi & Richards, 2021; Zarrinabadi & Mahmoudi-Gahrouei, 2018). For example, state-sector teachers in Iran undergo comprehensive four-year training programs, whereas private-sector teachers often rely on shorter, diverse training pathways (Ganji et al., 2018).
Iranian state schools face unique challenges, such as large class sizes, rigid curricula, and limited resources (Aliakbari & Saeedi, 2022), which further underscore the need for research in this underexplored setting.
While prior research has largely examined pre-service or private-sector teachers, in-service English as a foreign language (EFL) teachers in public-sector settings remain understudied. Moreover, much of our current understanding of in-service teachers’ cognitions and practices regarding OCF is drawn from studies conducted with public-sector teachers in the Vietnamese EFL context. Although these studies have offered valuable insights, further research is needed to develop a clearer and more contextually grounded understanding of teachers’ OCF cognitions and practices. This study addresses this gap by investigating Iranian in-service EFL teachers’ cognitions and reported practices concerning OCF in state-sector schools. Drawing on Borg’s (2015) framework of teacher cognition and Lyster and Ranta’s (1997) classification of CF types, the study focuses on five key aspects of OCF: (1) source (the origins of teachers’ OCF cognitions), (2) provider (who should deliver the feedback), (3) linguistic target (which language components should be corrected), (4) timing (when feedback should be given), and (5) feedback type (the strategies used to provide correction). By examining OCF in this context, the study furthers our understanding of teachers’ cognitions and practices and informs the design of teacher education programs in Iran and similar settings.
II Literature review
1 Research on key aspects of teachers’ OCF cognitions and practices
This section reviews previous research on key aspects of OCF, including the feedback provider, linguistic target, timing, and feedback types.
a Feedback provider
Research on OCF provider – whether feedback is delivered by teachers, peers, or through learner self-correction – shows that teachers generally view themselves as the most effective provider. Prior research (e.g., Ha & Nguyen, 2021; Yüksel et al., 2021) has indicated that teachers offer significantly more feedback than peers, or learners (i.e., self-correction), often considering OCF a core instructional responsibility (Ha & Murray, 2023; Ha & Nguyen, 2021; Rahimi & Zhang, 2015; Sepehrinia & Mehdizadeh, 2018). Teachers’ understanding of learners’ expectations (Ha, 2023a) and their awareness of contextual constraints further influence their OCF choices. For example, while Ha and Nguyen (2021) found that teachers viewed all three feedback sources (teacher, peer, and self-correction) as potentially effective, students in the same study expected their teachers to correct errors.
Although teachers note that self-correction can motivate learners, they also express concerns about learners’ limited proficiency and ability to self-correct. Despite being seen as time-consuming (Fallah & Nazari, 2019), self-correction is often preferred over peer correction, which teachers tend to avoid due to concerns about potential embarrassment or the risk of inaccurate feedback (Kaivanpanah et al. , 2015; Ha & Nguyen, 2021). While learners may favor peer correction (Sato, 2013), it remains the least preferred option among teachers (Yüksel et al., 2021).
b Linguistic target
Research on the linguistic targets of OCF – whether addressing pronunciation, vocabulary, or grammar – has only recently gained attention. In their study involving six in-service EFL teachers in Vietnam, Ha and Murray (2023) found that all participants regarded pronunciation errors as the most serious, citing the difficulty of correcting mispronunciations and the importance of fluency in oral communication. Vocabulary and grammar ranked second and third, respectively, as targets of OCF.
In another study, Ha (2023b) examined the OCF cognitions and practices of ten experienced Vietnamese EFL teachers. Regardless of the linguistic category, teachers emphasized correcting errors they considered serious and urgent, that is, those that related to the lesson focus, disrupted communication, or were frequently made by several students. Five teachers rated pronunciation as the most important target of OCF, while three chose grammar and two selected vocabulary. In practice, however, lesson focus strongly influenced OCF distribution. For instance, in grammar lessons, 89% of grammar errors were corrected.
c Timing
The timing of OCF – whether feedback is immediate or delayed – is a complex and context-sensitive issue. Studies report mixed findings, often moderated by teaching experience, learner needs, and task focus. Rahimi and Zhang (2015) found that experienced teachers in Iranian private institutes favored immediate feedback, whereas novice teachers preferred delayed correction. In contrast, Fallah and Nazari (2019) reported that novice teachers preferred immediate feedback, while experienced teachers favored a delayed approach.
Ha and Nguyen (2021) found that teachers tended to provide immediate feedback for errors that hinder communication, while opting for delayed feedback in the case of recurring errors made by multiple students. Similarly, Ha (2023b) observed that timing decisions were guided by lesson goals: immediacy was prioritized for communicative clarity, while delayed feedback was preferred for broader pedagogical aims. In line with these findings, teachers in Ha, Nguyen, and Hung’s (2021) study reported adjusting OCF timing based on error type – favoring immediate feedback for accuracy-focused tasks and delayed correction for fluency-based activities. All ten teachers in Ha (2023b) supported delayed feedback. However, in Yüksel et al. (2021), while teachers’ cognitions reflected a near-even split between immediate (48%) and delayed (52%) feedback, classroom observations revealed a strong preference for immediate correction in practice (79.6% vs. 20.4%).
d Types of OCF
Research on OCF types shows variation, ranging from balanced approaches (Dilāns, 2016) to clear preferences for specific strategies. Recasts have emerged as the most frequently used feedback type across numerous studies (e.g., Bao, 2019; Lyster & Ranta, 1997; Yüksel et al., 2021). For example, although teachers in Ha (2023b) considered prompts and explicit correction to be more effective, they primarily used recasts in practice. Similarly, Ha and Murray (2023) reported that recasts, despite being the second least favored OCF type, accounted for the majority of feedback (51.4%), followed by prompts (32.2%) and explicit correction (16.4%).
Teachers often choose recasts because they help maintain classroom flow and provide emotional support (Ha & Murray, 2023; Kartchava et al., 2020; Kamiya, 2016; Mori, 2011). However, their effectiveness in promoting learner uptake and retention has been questioned (Ha & Nguyen, 2021). Prompts, in contrast, are increasingly valued for fostering deeper cognitive engagement (Ha & Murray, 2023; Nassaji et al., 2023). Although teachers often regard prompts as more pedagogically effective, they tend to use recasts more frequently due to practical classroom constraints. This mismatch between teachers’ cognitions and practices has also been highlighted by Nassaji et al. (2023). In Ha et al.’s (2021) study, seven (of 24) teachers favored explicit correction, a preference attributed to the exam-oriented nature of Vietnamese classrooms and teachers’ perceptions of their roles as knowledge authorities responsible for providing correct answers.
2 Factors influencing teachers’ cognitions and practices concerning OCF
In the concluding chapter of his influential work, Borg (2015) identifies three major factors (or ‘sources’ as used in this study) that influence language teacher cognitions: schooling experiences, professional coursework (including pre- and in-service teacher education), and classroom practice shaped by contextual factors (see Figure 1). While research on teachers’ cognitions and practices regarding OCF has addressed aspects such as provider, timing, and feedback type, few empirical studies have investigated how teachers themselves perceive the influence of these three sources on their OCF cognitions and practices, and what reasons teachers cite for the possible impact of these factors on their cognitions and practices. Agudo’s (2014) study, though focused on pre-service teachers, remains one of the few that examines sources of OCF cognitions. In this study, language learning experiences emerged as the primary influence (60%), followed by teacher training courses (42%) and exposure to books and articles on language teaching methodology (41%).
Elements and processes in language teacher cognition.
Compared with the limited focus on schooling experiences, teaching experience and contextual factors have received more attention in studies on OCF cognitions and practices. Research indicates that experienced teachers are more adept at recognizing contextual and learner-specific factors (e.g., proficiency level) when delivering feedback. For example, they are more likely to employ delayed or peer feedback, while novice teachers often prefer immediate correction (Fallah & Nazari, 2019). Gatbonton (2008) found that experienced teachers draw on more developed pedagogical knowledge, enabling them to adapt feedback strategies more effectively.
Cross-cultural studies also demonstrate how teaching experience mediates in-serve teachers’ OCF cognitions. Mahalingappa et al. (2021), in a comparative study of teachers in China and the U.S., showed that both cultural context and teaching experience shape cognitions about effective feedback. Similarly, Ha and Murray (2023) found that experienced EFL teachers in Vietnam considered both contextual and learner-related factors in their OCF practices. Findings from higher education contexts (Soruç et al., 2024; Sun & Zhang, 2022) further underscore the crucial role of experience in developing effective OCF strategies.
As Borg (2011) notes, most studies on the relationship between teacher education and cognition have focused on pre-service teachers (e.g., Busch, 2010; Kartchava et al., 2020), leaving the experiences of in-service teachers underexplored. This gap is particularly relevant in countries like Iran, where formal teacher education programs in the public sector differ significantly from those in the private sector. For example, public-school teachers in Iran complete rigorous 4-year training programs, while private-sector teachers often receive ad hoc training of varying quality (Iranmehr & Davari, 2018). Understanding how such formal training influences in-service teachers’ OCF cognitions and practices is essential for assessing the broader impact of teacher education in similar contexts.
Recent research in the Vietnamese EFL context provides valuable insights into this issue. Studies by Ha and Nguyen (2021), Ha (2023b), and Ha, Murray, and Reynolds (2024) demonstrate that professional development significantly shapes in-service teachers’ OCF cognitions and practices, particularly in terms of feedback timing and type. Ha (2023b) also revealed inconsistencies between teachers’ stated beliefs and observed practices, influenced by contextual constraints and lesson objectives. These findings emphasize the importance of examining the interplay between formal education and real-world classroom practices, especially in contexts where CF is a critical component of language instruction. Building on this literature, the present study investigates how Iranian in-service EFL teachers perceive the influence of various factors – including their language learning experiences, pre- and in-service education programs, as well as teaching experiences and contextual realities – on their OCF cognitions and reported practices. By addressing this gap, the study contributes to a more comprehensive understanding of the factors shaping teacher cognition and practice in public-sector EFL settings.
The present study is guided by the following research questions:
Research question 1: What are Iranian in-service EFL teachers’ cognitions and reported practices concerning OCF?
Research question 2: What are Iranian in-service EFL teachers’ justifications for their OCF cognitions and reported practices?
III Method
1 Design
The present study employed an explanatory sequential mixed-methods design, combining a survey with semi-structured interviews. The aim was to first obtain a general view of teachers’ cognitions and reported practices concerning OCF and then to explore teachers’ justifications for their views. Accordingly, the study began with a survey to elicit teachers’ cognitions and reported practices, followed by semi-structured interviews to explore their justifications.
2 Participants
A total of 204 in-service EFL teachers from Iranian lower and upper secondary public schools participated in the study. Respondents had learned English as a foreign language and represented diverse geographic backgrounds. To ensure the inclusion of participants from both urban and less-developed regions, the survey link was shared through provincial head teachers across the country. The sample consisted of 47.5% male, 50% female, and 2.5% who preferred not to state their gender. The majority of respondents were over 35 years old (69.6%; see Table 1) and held bachelor’s (51%) or master’s (43.1%) degrees (see Table 2).
Age groups of respondents.
Academic degree of respondents.
Note. AA = Associate of Arts (a two-year degree).
Respondents taught at different school types (Table 3) with a large portion teaching at state lower (58.3%) and upper secondary (41.7%) schools. To meet the study criteria, respondents were required to teach primarily in state schools; thus, those employed exclusively in private schools or institutes were excluded from the dataset. Participants also varied in their teaching experience, with most having over 10 years of experience (55.4%; see Table 4).
School type of respondents.
Note. *These teachers were those who were teaching primarily at state schools but some of them also taught at private institutes.
Teaching experience of respondents.
Of the whole sample, those interested in participating in the qualitative phases of the study provided their contact information in the allocated space at the end of the questionnaire. Since more teachers volunteered than needed, 10 were purposively selected based on gender, academic degree, teaching experience, and school type. All were qualified EFL teachers and spoke Persian as their first language. None of them had previously participated in a study on OCF, nor had they attended any course on this topic. Table 5 represents the demographic information of this group.
Demographic information of the interviewees.
Note. Names are pseudonyms.
3 Instruments
a Questionnaire
The questionnaire used in this study (see Appendix A in supplemental material) was a substantially revised version of the questionnaire used by Zare et al. (2022). Several modifications were made to better align it with the goals of the present study and the context of Iranian EFL classrooms. The demographic information section (A) was updated to include age, gender, university degree and teaching experience. A new section (Section B) was added to capture the sources of teachers’ OCF cognitions and practices. Additionally, the wording of the instructions and the examples of learner errors were revised to enhance clarity and contextual relevance, using error types more representative of Iranian EFL learners. The questionnaire items and sections (see Table 6) were developed following a careful review of the literature on teachers’ CF beliefs (e.g., Li, 2017; Lyster et al., 2013; Nassaji, 2016; Nassaji & Kartchava, 2021; Nassaji & Kartchava, 2017). The main part of the questionnaire consisted of five sections: source of cognitions (3 items), provider (3 items), target (3 items), timing (3 items), and types of OCF (7 items).
Sections and the corresponding items of the questionnaire.
The questionnaire was initially written in English and then translated into Persian using a back-translation procedure to facilitate teachers’ understanding. It was subsequently piloted with three teachers, after which revisions were made to enhance clarity, organization, and usability. The questionnaire items, along with the related open-ended responses, captured both cognitions and reported practices across all aspects – except for the linguistic target and types sections, which measured only reported practices and cognitions, respectively. This was a deliberate design decision, based on piloting and recommendations from survey design literature (Dörnyei & Dewaele, 2023), to balance questionnaire length and response rate. The questionnaire results are presented in accordance with this clarification. The completion time of the questionnaire ranged from 10 to 15 minutes depending on the length of the information teachers provided in the open-ended sections. The reliability of the questionnaire (Dörnyei & Csizer, 2012) was measured through Cronbach’s coefficient (.79). Finally, the electronic version of the questionnaire was created on SurveyXact Platform (www.surveyxact.com), and the link to the questionnaire was shared with the provincial head teachers.
b Semi-structured interviews
Semi-structured interviews were conducted with 10 teachers. These interviews provided an opportunity for teachers to explain and clarify their cognitions and practices across all aspects of OCF, thereby complementing their questionnaire responses. Informed by research on constructing interviews (Galletta, 2012; Canh & Maley, 2012; Gubrium et al., 2012; Merriam, 2015; Patton, 2015; Silverman, 2006), I developed the interview guide (see Appendix B in supplemental material). The interview focused on the same topics as the questionnaire and gave teachers opportunities to elaborate on the answers they had provided in the questionnaire. I drew on the interviewees’ questionnaire responses to guide and extend the interviews. The interview guide was piloted with three teachers, and teachers found the questions clear and were comfortable answering them. The interviews were conducted synchronously online via Google Meet, in time slots convenient for teachers, following established guidelines for online interviewing (e.g., Salmons, 2015). Interviews lasted from 39 to 61 minutes (47 minutes on average), and were audio-recorded with teachers’ consent for further analysis.
4 Data analysis
Data in this study were collected from multiple sources. The 204 completed questionnaires were exported from SurveyXact Platform to SPSS for analysis. Since the study aimed to describe teachers’ cognitions and reported practices regarding predefined aspects of OCF in an underexplored context – rather than to extract latent factors or compare groups – descriptive statistics were deemed the most appropriate analytical approach. Thus, frequencies and percentages were calculated and reported for each item. Additionally, responses from the open-ended sections of the questionnaire (6,395 words) were compiled into a separate Word document for content analysis. Interviews were also transcribed verbatim for qualitative analysis. The study employed descriptive statistics for questionnaire data, and reflexive thematic analysis (Braun & Clarke, 2019, 2022) for the interview and open-ended questionnaire responses to address the research questions. Following transcription, I conducted multiple readings of the interview data to familiarize myself with emerging themes. I then color-coded topics to create categories; for instance, teachers’ comparisons between gifted and regular students were coded as learner factors. The coding process was iterative, ensuring accuracy and coherence across categories. The extraction of final themes was largely inductive. However, in interpreting the data, I drew on my insider position to contextualize teachers’ comments in light of my own teaching experiences in similar educational settings.
IV Results
1 Cognitions and practices
Before presenting the results, it should be noted that the term ‘practices’ in this study refers specifically to teachers’ reported practices elicited through questionnaires and interviews, not to observed classroom practices. This distinction is important for interpreting the findings. The second section (B) of the questionnaire asked about the source of teachers’ OCF cognitions and practices. As shown in Table 7, the highest level of agreement and strong agreement was on teaching experiences (89.2%), followed by teacher education programs (54.4%) and language learning experiences (45.1%). The closer examination of the individual items revealed that, while relatively high levels of uncertainty were reported for language learning experiences (22.1%) and pre-/in-service teacher education programs (23%), very few teachers (6.9%) expressed uncertainty (and chose the midpoint ‘neither agree nor disagree’) about the role of teaching experience in shaping their cognitions and practices. Similarly, high levels of disagreement were stated about the role of language learning experiences (32.8% disagree and strongly disagree) and teacher education programs (22.5% disagree and strongly disagree) compared to 4% for teaching experiences.
Source of teachers’ cognitions and reported practices: frequency (percentage).
Note. SD = strongly disagree; D = disagree; NA = neither agree nor disagree; A = agree; SA = strongly agree.
The third section of the questionnaire (C) examined teachers’ cognitions and practices regarding the provider of OCF (Table 8). The majority of teachers (68.7%) agreed and strongly agreed that self-correction is the most effective form of OCF. In the open-ended section of the questionnaire, this choice was justified with reference to reasons such as deeper and longer-lasting learning and the value of self-discovery and active involvement in the correction and learning process for the effectiveness of self-correction. Nonetheless, some teachers raised concerns about the wait-time and the time-consuming process of self-correction. Teacher was viewed, in terms of the respondents’ level of agreement, as the second most effective OCF provider (54.8% agree and strongly agree), because students viewed teachers as ‘knowledgeable, trustworthy, role model, and supportive authority’. Teachers were thus seen as a safer choice (than peers) especially in classes where the overall atmosphere was not friendly and supportive. Teachers agreed the least with peer OCF as the provider of OCF (36.3%) and reported in the open-ended sections to use it minimally. This is apparently related to the contextual nuances of each class where teachers seem to be having difficulty deciding on whether to allow peer feedback or not. For example, a significant percentage of teachers (37.3%) neither agreed nor disagreed with the statement that peer feedback was the most effective approach. In the open-ended section of the questionnaire, teachers wrote that, due to the potential ‘humiliation, resentment, negative feelings, decreasing self-confidence, and damaging each other’s (classmates) egos’ that are associated with the use of peer feedback, they prefer not to use it, and if they do, they exercise caution to prevent its possible side effects.
Teachers’ cognitions and practices concerning oral corrective feedback (OCF) provider.
Section 4 of the questionnaire (D) asked about the reported frequency of providing OCF for different linguistic targets (Table 9). Feedback was most frequently provided for vocabulary errors (54.4% often and always), followed by pronunciation (49.5%) and grammar (45.4%). These results suggest a slight prioritization of vocabulary in classroom feedback routines, though the overall distribution is relatively balanced.
Teachers’ reported practices concerning the linguistic target of oral corrective feedback (OCF).
Teachers’ views on the timing of OCF (immediate vs. delayed) showed notable variation (Table 10). Just over half (51.4%) agreed with immediate correction of pronunciation errors, justifying this choice based on:
Teachers’ cognitions and practices concerning the timing of oral corrective feedback (OCF).
Their personal views about the importance of pronunciation and language learning.
Preventing the fossilization of errors for the speaker and other students.
Simply the possibility of forgetting the error if not corrected immediately.
Another group of teachers disagreed and strongly disagreed (37.3%) with the immediate correction of pronunciation errors as they believed that it threatens students’ self-confidence and their tendency to participate in class activities. There was a third group, who chose immediate correction depending on, for instance, the focus of the lesson (they correct immediately if the focus of the lesson/task is on pronunciation) and the saliency of the error (they correct immediately if the error significantly deviates from the correct form).
The disagreement among teachers about the timing of OCF was more pronounced in vocabulary (46.1% agree and strongly agree vs. 34.8% disagree and strongly disagree) and grammar (38.7% agree and strongly agree vs. 43.1% disagree and strongly disagree) errors. While some teachers generally opposed immediate OCF because of its negative consequences for learners’ willingness to speak up in class and for embarrassing them, other teachers advocated the immediate treatment of vocabulary errors as it might lead to miscommunication. Regarding grammar errors, while most teachers disagreed and strongly disagreed (43.1%) with immediate OCF, 38.7% of the respondents agreed and strongly agreed with it.
The final section (F) of the questionnaire examined teachers’ cognitions regarding the types of OCF and their effectiveness (Table 11). Recasts were most frequently rated as very effective and extremely effective (71.6%) OCF strategy, followed by three subtypes of prompts, that is, clarification request (66.6%), elicitation (61.8%), and repetition (60.3%). Explicit correction ranked fifth with 44.6%, followed by metalinguistic feedback (44.3%). Ignoring was rated as very effective and extremely effective by 17.7% of the respondents.
Teachers’ cognitions concerning oral corrective feedback (OCF) types.
2 Teachers’ justifications for their cognitions and practices
Research question 2 examined teachers’ justifications for their OCF cognitions and reported practices. Analysis of the interview data revealed three overarching themes: affective, cognitive, and contextual factors (see Figure 2). These themes reflect the complex interplay of emotional, pedagogical, and situational considerations in teachers’ decision-making processes. Affective factors included learner-related factors such as anxiety and motivation that teachers take into account when providing feedback. Cognitive factors concerned teachers’ general knowledge of language learning and teaching, while contextual factors related to largely inside-the-class (but also curriculum) factors.
Factors shaping teachers’ oral corrective feedback (OCF) cognitions and reported practices.
Although affective, cognitive, and contextual factors could all feed directly into teachers’ cognitions and practices, affective factors appeared to override cognitive and contextual factors in teachers’ decisions about proving OCF. For example, a teacher who believed in the importance of pronunciation (a cognitive factor) and often corrected pronunciation errors might withhold feedback if an anxious student (an affective factor) made a pronunciation error – illustrating how emotional sensitivity often guided instructional choices. Thus, based on the distinction made between core and peripheral aspects of beliefs in Phipps and Borg’s (2009) study, affective factors can be perceived as the core component of teachers’ cognitions and practices regarding OCF, whereas cognitive and contextual factors are viewed as peripheral.
As for the source of OCF cognitions and practices, and consistent with the questionnaire results, teaching experience was a powerful determinant of teachers’ OCF cognitions and practices. Teachers’ ideas, however, seem to be largely influenced by critical incidents that happened to them during schooling time (as a learner), pre-service teacher education programs, and inside the class as a practicing teacher. Critical incidents, according to Brookfield (1990) refer to a ‘vividly remembered event which is unplanned and unanticipated’ (p. 84). Reflection on critical incidents may lead to shaping new understanding of learning and teaching process for teachers (Richards & Farrell, 2005). The critical incidents had important affective (e.g., feeling pity for the student who was humiliated by his friends after being corrected) or cognitive (learning about OCF from a teacher educator) ramifications for teachers’ current cognitions. Four such incidents are cited in Table 12.
Examples of critical incidents.
Figure 3 represents the mediating role of critical incidents in the source of teachers’ OCF cognitions and practices. Affective support was a major concern in teachers’ justification of their choices regarding feedback provider. For example, Rezi noted: Depending on the class atmosphere, peer feedback can be used, but, in my experience, it often leads to anxiety and humiliation. So, I am extra careful with it and in cases where self-correction fails often correct the errors myself.
The mediating role of critical incidents in the source of teachers’ oral corrective feedback (OCF) cognitions and practices.
Echoing Rezi’s concerns for the affective support of the learners, Rahil noted that ‘self-correction is the best choice, but in case there is a friendly and supportive atmosphere in the class, peer feedback can also work well’. She continued: But peer feedback doesn’t work well in special schools with gifted students. Though I know that students are proficient enough to provide each other with feedback, I am against it because the atmosphere in this school (gifted students’ school) is so competitive that it will lead to embarrassment and humiliation for the receiver of feedback.
Kari acknowledged the ‘longer lasting impact of self-correction in language learning’, while Safa, a first-year teacher, was against peer feedback but supported self-correction and teacher feedback. He explained: Peer feedback in my class is a waste of time. Students either give the wrong feedback (another error) or give the correct one but with some sort of humiliation. In both cases I have to intervene and give the correct form and also allocate time and energy to manage the class.
The choice of the frequency of giving feedback in the linguistic targets appeared to be largely influenced by teachers’ knowledge of language learning and teaching and the personal importance they attached to these targets. For example, Safa commented: I have watched a lot of Hollywood, and I see that in these movies, even native speakers don’t follow the grammatical rules strictly, but the pronunciation is standard. So, I personally think that a good English speaker should have a good pronunciation. That’s why I give more feedback to pronunciation errors than to vocabulary and grammar errors.
This idea resonated with Menay’s comment ‘I have worked a lot on pronunciation and, due to its importance in language learning, expect my students to pronounce well too.’ Rahil and Rezi, however, considered contextual factors such as ‘the focus of the lesson’ an important factor determining the frequency of feedback they give in different linguistic targets. Rezi stated: I correct pronunciation if the focus of the lesson is on pronunciation, for example, when teaching new words and sentences. But if we are working on grammar, and the student is reading her sentences out loud with pronunciation errors, I don’t correct them.
She was the only interviewee who supported more feedback on grammar as she believed that ‘grammar is the most important element of language learning’. Highlighting another contextual factor, that is, the aim of the curriculum, Monji maintained that ‘I correct pronunciation errors because the curriculum focuses on communication, and pronunciation is an important element of successful communication.’
The timing of OCF clearly revealed teachers’ concerns for the affective support of their students while providing feedback. Almost half of the interviewees were generally against immediate OCF because of its potential negative effects on their students. For instance, Zari and Rahil indicated their disagreement with immediate feedback, noting that ‘immediate correction decreases self-confidence and learners’ tendency to talk again in class’ (Zari), and that ‘what matters is to keep students motivated by letting them finish their turns and not disrupt the flow of communication’ (Rahil). In contrast, Kari, Monji, and Rekizad advocated the immediate correction of pronunciation errors, emphasizing the importance of pronunciation in spoken language (a cognitive factor). Meanwhile, Rohi, Rekizad, and Safa linked the timing of OCF to learners’ proficiency levels (a contextual factor), but still prioritized affective considerations when making feedback decisions. For instance, Rohi and Rekizad reported that they correct pronunciation errors immediately for higher-proficiency learners to prevent fossilization, but they also took into account students’ emotional readiness (e.g., their motivation) before giving feedback.
All interviewed teachers but Rahil viewed ignoring errors as not at all effective. Rezi, Rahil, and Safa stressed the teacher’s responsibility to correct errors. However, Safa was also concerned about the students’ distrust of him and losing his authority in class as a result of ignoring errors. He stated, ‘if gifted students in class realize that there are errors in other students’ productions, and the teacher doesn’t correct them, they take it as teacher’s lack of knowledge in that area and consequently lose their trust in teacher.’ Another idea about ignoring the error shared by Safa, Rahil, Rezi, and Rekizad was that they ignored errors that were not central to the focus of the lesson. In this regard, Rahil believed that ‘ignoring, by maintaining the flow of speech, can have a moderate effect for errors in uncovered parts of the lesson’.
Recasts were widely favored for being indirect, supportive, and minimally disruptive. Comments made by Rohi, Rahil, and Safa further support the idea that affective factors were powerful determinants of teachers’ choices regarding OCF types. For instance, Rohi stated ‘this type (i.e., recasts) is very effective, especially for female students, as they don’t like to be stopped in an explicit way as in other types (e.g., explicit correction).’ Similarly, Safa remarked, ‘I use this type frequently as I don’t like to directly tell them (students) that what they say is wrong.’ Menay and Rezi were the only teachers who expressed concerns about the noticeability of recasts and, for this reason, rated them as moderately effective.
Contrary to recasts, explicit correction – apparently owing to its direct and explicit nature and potential to threaten learners’ feelings – was not considered highly effective. For instance, Rohi noted ‘I don’t use explicit correction much as it embarrasses students.’ However, he also acknowledged a contextual justification for its use: ‘sometimes I am pressed for time and use explicit correction to correct errors quickly.’ Nevertheless, Kari supported the use of explicit correction and believed that it was very effective because ‘through it (explicit correction), not only the students who made the error but also the whole class notices the error and the probably not repeat it.’
Prompts – including clarification request, elicitation, repetition, and metalinguistic feedback – were generally rated as very and extremely effective types of OCF. The main reason teachers cited for their effectiveness was their potential to lead to self-correction, which, from the teacher’s perspectives, was the most effective form of OCF (as noted by Menay. Rahil, and Rekizad). The effectiveness of elicitation was also seen as dependent on learners’ proficiency levels (Monji, Kari, Menay). For instance, Menay explained ‘elicitation works better for higher proficiency learners, who know the correct form but only due to the affective factors (such as anxiety) make a mistake. If hinted, they can retrieve the correct form themselves.’ Although all but one interviewee rated metalinguistic feedback as very and extremely effective, it was less preferred than other types of prompts due to its time-consuming nature (Menay, Rohi, Monji).
V Discussion
This study investigated Iranian state-school EFL teachers’ OCF cognitions and reported practices, as well as their justifications for these cognitions and practices. The findings indicate that affective factors play a central role in shaping teachers’ OCF decisions, while cognitive and contextual factors serve as peripheral influences (Phipps & Borg, 2009). Furthermore, teaching experience emerged as the most influential source of teachers’ OCF cognitions and practices, supporting previous research (e.g., Kamiya, 2016; Mahalingappa et al., 2021; Rahimi & Zhang, 2015). Teachers’ reflections on critical incidents, such as instances of learner embarrassment during peer feedback, illustrate how specific experiences shaped their long-term OCF preferences and classroom behaviors (Brookfield, 1990; Farrell, 2008).
A key finding of this study is the complex interplay between teachers’ cognitions and their practices. While teachers’ cognitions often guided their choices, practical constraints frequently necessitated adaptations. For example, the belief that self-correction fosters longer-lasting learning and learner autonomy led many teachers to prioritize prompts and delay explicit correction. However, in practice, time limitations and learners’ proficiency levels often compelled teachers to provide direct feedback instead, aligning with findings by Ha et al. (2021) and Ha and Murray (2023). This interplay reflects the dynamic nature of classroom decision-making, where teachers navigate between pedagogical goals and contextual constraints (Kubanyiova & Feryok, 2015). While most teachers in this study endorsed self-correction, they frequently resorted to teacher feedback due to concerns about peer feedback’s accuracy and classroom management issues. Similar contradictions were observed in Ha (2023a), where lesson focus and classroom dynamics significantly shaped the alignment, or misalignment, of teachers’ stated beliefs and actual classroom practices.
Consistent with previous research (e.g., Kamiya, 2016; Rahimi & Zhang, 2015), teaching experience was identified as the most influential factor shaping teachers’ OCF cognitions and practices. Teachers often referred to critical incidents from their careers – such as embarrassment during peer feedback or students’ emotional reactions to correction – as defining moments that shaped their feedback preferences. This finding aligns with Brookfield (1990) and Farrell (2008), who emphasize the formative impact of critical incidents on teacher development. Furthermore, Ha and Murray (2021) emphasize that professional development plays a key role in refining teachers’ OCF strategies. While teachers in this study acknowledged the importance of professional learning, they did not explicitly mention receiving formal training in OCF. This highlights a gap in teacher education programs, where practical, context-sensitive training on OCF remains limited.
Teachers’ cognitions and practices regarding OCF provider largely echoed previous research (Sepehrinia & Mehdizadeh, 2018; Ha & Murray, 2023; Ha & Nguyen, 2021; Rahimi & Zhang, 2015). However, a notable difference emerged when comparing these findings to Yüksel et al. (2021). In their study, teachers were rated as the most effective OCF provider, whereas in this study, self-correction received the highest rating. This discrepancy may stem from differences in educational contexts: Yuksel et al.’s university-level participants dealt with complex academic content, making self-correction more challenging. In contrast, secondary school teachers in this study viewed self-correction as both feasible and beneficial for fostering learner autonomy. Despite this difference, both studies found peer feedback to be the least preferred OCF provider. Teachers’ reluctance to use peer feedback was primarily driven by affective concerns, such as the risk of learner humiliation. This finding aligns with Kaivanpanah et al. (2015) and Ha and Nguyen (2021), who emphasize the critical role of classroom atmosphere in shaping peer correction practices. Interview data reinforced this point: for example, Safa expressed that peer feedback led to classroom management issues and potential emotional distress for students, despite his recognition of its potential benefits. Ultimately, he relied more on teacher feedback and self-correction, as affective constraints made peer feedback impractical in his context. Furthermore, although self-correction was rated as the most effective, practical challenges – including learners’ proficiency levels and classroom time constraints – often led teachers to favor teacher-provided feedback. This supports Ha and Nguyen (2021), who found that teachers in EFL contexts frequently resort to teacher feedback due to its reliability and alignment with learners’ expectations. Taken together, these findings underscore the mediating role of contextual and affective factors in shaping the relationship between teachers’ OCF cognitions and classroom practices.
Findings regarding the linguistic target of OCF largely aligned with Yüksel et al. (2021), where teachers reported correcting vocabulary errors most frequently. However, the interview data revealed that, in practice, pronunciation errors were addressed more often, in line with studies such as Ha and Murray (2023) and Bao (2019). This discrepancy can be partly attributed to how error examples were framed in the questionnaire. For instance, a vocabulary error such as went instead of stepped on my foot may have been perceived as more critical than a pronunciation mistake such as draver instead of drawer. Interviews further revealed that OCF decisions were context-dependent, with teachers adjusting their focus based on lesson objectives and communicative significance. For instance, grammar received less feedback unless explicitly emphasized in the curriculum, a pattern also reported in Ha (2023a).
Teachers’ cognitions and practices regarding OCF timing reflect an ongoing tension between correcting errors and protecting students’ emotions (Nassaji et al., 2023; Roothooft & Breeze, 2016; Yüksel et al., 2021). Consistent with Ha and Nguyen (2021), pronunciation errors that impeded communication were corrected immediately, whereas grammar and vocabulary errors were more often addressed delayed to maintain learner confidence. Teachers’ open-ended responses and interviews further highlighted the emotional and cognitive tensions involved in OCF timing. Several participants, such as Safa, Monji, and Rohi, reported that while they personally favored immediate pronunciation correction, they often withheld feedback if a student was visibly struggling to articulate an idea. For instance, Safa, who, due to personal preferences, believed in the immediate correction of pronunciation errors, noted that in practice he withheld OCF when he noticed that ‘a student was effortfully trying to say something.’ Likewise, Rahil, who viewed the focus of the lesson, in line with Ha (2023b), and emotional factors, as important contextual factors in her OCF provision, ignored the error and corrected it later when the OCF was likely to trigger negative feeling for the student. These findings underscore the interconnected nature of cognition, affect, and classroom realities in shaping OCF decisions (Barcelos & Ruohotie-Lythy, 2018; Swain, 2013).
The above pattern was also evident in teachers’ cognitions and practices regarding OCF types. Recasts were the most favored OCF type due to their implicit and supportive nature, corroborating findings from previous research (Bao, 2019; Ha & Murray, 2023; Nassaji et al., 2023; Roothooft, 2014; Sepehrinia & Mehdizadeh, 2018; Yoshida, 2008, 2010; Yuksel et al., 2021). However, concerns about learners’ uptake of recasts, as raised in Ha and Nguyen (2021), suggest a need for further exploration of how implicit feedback can be made more effective. Prompts, particularly clarification requests and elicitation, were valued for their potential to foster self-correction, which teachers viewed as the most effective form of feedback. These findings align with Ha (2023b), who emphasized the role of prompts in promoting learner autonomy. Explicit correction, although less favored, was used strategically in time-constrained situations which reflects teachers’ pragmatic adaptation to classroom realities.
These findings contribute to ongoing calls in the literature to reconsider the boundaries of teacher cognition and to reach a unifying conceptual framework for researching outstanding issues in the field (Borg, 2003; Golombek & Doran, 2014; Kubanyiova & Feryok, 2015). One of these outstanding issues is the relationship between emotions and cognitions as documented in the recent surge of scholarly interest in this area (e.g., Amory & Johnson, 2023; Barcelos & Ruohotie-Lythy, 2018; Benesch, 2012, 2017). The emotional turn in applied linguistics (White, 2018) has demonstrated that emotions are not peripheral but central to teachers’ professional decision-making, including their feedback practices (Richards, 2022). Prior CF research has also highlighted the role of affective factors in teachers’ CF cognitions and practices (Gurzynski-Weiss, 2016; Ha & Murray, 2023; Mori, 2011; Roothooft, 2014; Sepehrinia & Mehdizadeh, 2018; Soruç, et al., 2024; Yoshida, 2008, 2010). The findings of this study add to this strand of research and suggest that affective factors not only play an important role in shaping teachers’ OCF cognitions and reported practices but can also override cognitive and contextual factors.
While Borg’s (2015) general framework of language teacher cognition is supported by the findings of this study, the results point to the need for a refined model specific to OCF. Based on this study and supported by prior research on the role of affective factors in teachers’ CF cognitions and practices (e.g., Gurzynski-Weiss, 2016; Sepehrinia & Mehdizadeh, 2018; Soruç et al., 2024; Yoshida, 2008, 2010), an expanded model of teachers’ OCF cognitions and practices is proposed in Figure 4. This model introduces two key modifications: (1) the inclusion of emotions as a central component of teachers’ OCF cognitions and practices, and (2) the addition of affective factors as an influential source – alongside existing sources, that is, schooling, professional coursework, and classroom practice – of teachers’ OCF cognitions and practices.
The proposed model of language teachers’ oral corrective feedback (OCF) cognitions and practices.
Although affective factors in this model refer to learners’ emotions, the findings show that these have a substantial impact on teachers’ OCF cognitions and practices. As Dewaele (2021, p. 2) notes, ‘emotions of learners are connected to those of teachers and vice versa.’ Similarly, Borg (2006, p. 23) highlights that effective language teachers ‘radiate positive feeling’ by accommodating learners’ affective factors. The model proposed here aims to redeem the imbalanced cognition-emotion relationship, emphasizing that teacher decision-making, particularly in OCF, is best understood as a product of interacting emotional, cognitive, and contextual influences.
VI Conclusions
The present study contributes to the growing body of research on teachers’ OCF cognitions and practices by examining in-service EFL teachers in the underexplored context of Iranian public schools. The findings indicate that teaching experience plays a central role in shaping teachers’ OCF cognitions and practices, with critical incidents serving as key mediators in their development. Importantly, this study expands our understanding of the core and peripheral components of teachers’ OCF decisions, highlighting that affective factors constitute the core influence, while cognitive and contextual factors play a supporting role.
Despite these contributions, the study has certain limitations. First, teachers’ reported practices regarding OCF may not fully reflect their actual classroom behavior. Previous research (e.g., Ha & Murray, 2023) has shown that while teachers can report on their feedback tendencies, there is often a discrepancy between reported and observed practices. In particular, teachers may underestimate the frequency of their corrections or misjudge the timing. Thus, the findings regarding teachers’ reported practices, especially in areas such as CF timing, should be interpreted with caution. Future research should incorporate classroom observations to provide a more comprehensive picture of how cognitions translate into practice. Second, this study relied on one-off questionnaire and interview data; a longitudinal approach could offer deeper insights into how teachers’ OCF cognitions and practices evolve over time.
Nevertheless, the study carries important implications for teachers, teacher educators, and policymakers in Iran and similar EFL contexts. By drawing attention to teachers’ affective reasoning in feedback decisions, the study underscores the need to view teaching not solely as a cognitive activity but also as a deeply affective practice. Teacher education programs should therefore explicitly address the role of emotions in teaching and provide pre-service teachers with opportunities to reflect on both their own affective experiences and those of their learners. For example, teacher educators might begin with discussions of pre-service teachers’ emotionally charged classroom memories and use these as the starting point for reflective dialogue and pedagogical planning. Practicum experiences can be enhanced by training future teachers in recognizing and responding to learners’ emotional cues during feedback exchanges. Policymakers may also strengthen teachers’ awareness of the role of critical incidents in the development of their cognitions and practices by designing professional development courses wherein such incidents are used as the content for reflection and discussion.
Finally, further research is warranted to enlighten other aspects of affective factors in teachers’ OCF decisions. For instance, investigating the role of teacher-related affective factors in developing OCF cognitions and practices can be a timely consideration. Future research can also examine the interdependence and the possible reciprocal relationship between teachers’ OCF cognitions and practices and the emotional aspects of teaching and learning. By deepening our understanding of the emotional dimension of OCF, future research can help refine teacher training programs and pedagogical practices to ensure that OCF is not only effective but also emotionally responsive to learners’ needs.
Supplemental Material
sj-pdf-1-ltr-10.1177_13621688251391502 – Supplemental material for Oral corrective feedback: Iranian in-service EFL teachers’ cognitions and reported practices
Supplemental material, sj-pdf-1-ltr-10.1177_13621688251391502 for Oral corrective feedback: Iranian in-service EFL teachers’ cognitions and reported practices by Mostafa Zare in Language Teaching Research
Footnotes
Acknowledgements
I would like to sincerely thank all the teachers who participated in this study. I am also grateful to the editors and anonymous reviewers of Language Teaching Research for their insightful feedback on earlier versions of this article.
Data Availability
In line with participants’ informed consent, the data collected for this study were to be used exclusively for the stated research purposes. Therefore, it is ethically not possible to share the data that support the findings of this study.
Funding
The author received no financial support for the research, authorship, and/or publication of this article.
Ethical approval and informed consent
This study protocol was reviewed and approved by the Norwegian Agency for Shared Services in Education and Research under reference number 249097.
Supplemental material
Supplemental material for this article is available online.
References
Supplementary Material
Please find the following supplemental material available below.
For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.
For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.
