Impacts of implicit, explicit, and emergent feedback strategies on EFL learners’ motivation, attitude and perception

Abstract The aim of this study was to develop an understanding of the possible various influences of teachers’ oral corrective feedback (CF) strategies on learners’ learning motivation and further to investigate the learners’ attitudes and perceptions toward their teachers’ feedback strategies. The participants were 54 pre-intermediate Iranian EFL learners comprising three experimental groups (Implicit, Explicit, and Emergent). The Implicit and Explicit groups received just recast and explicit correction for their erroneous oral production, respectively, while the Emergent group took CF from implicit to explicit. A motivation questionnaire and semi-structured interviews were used to gather the quantitative and qualitative data for the study. The results of inferential statistics indicated meaningful differences among the feedback types with respect to their effectiveness on learning motivation. Moreover, emergent feedback was more effective compared with the other feedback types in raising the learners’ learning motivation. The results of the qualitative data analysis also confirmed the effectiveness of the emergent feedback in boosting the positive affective characteristics in learners. This study could specifically increase teachers’ and teacher trainers’ awareness of the interrelationship between CF types and language learning motivation, attitude and perception.

Abstract: The aim of this study was to develop an understanding of the possible various influences of teachers' oral corrective feedback (CF) strategies on learners' learning motivation and further to investigate the learners' attitudes and perceptions toward their teachers' feedback strategies. The participants were 54 preintermediate Iranian EFL learners comprising three experimental groups (Implicit, Explicit, and Emergent). The Implicit and Explicit groups received just recast and explicit correction for their erroneous oral production, respectively, while the Emergent group took CF from implicit to explicit. A motivation questionnaire and semi-structured interviews were used to gather the quantitative and qualitative data for the study. The results of inferential statistics indicated meaningful differences among the feedback types with respect to their effectiveness on learning motivation. Moreover, emergent feedback was more effective compared with the other feedback types in raising the learners' learning motivation. The results of the qualitative data analysis also confirmed the effectiveness of the emergent feedback ABOUT THE AUTHORS Touran Ahour is an assistant professor in TEFL at Islamic Azad University, Tabriz Branch, Iran. She has authored several books and published many articles in scholarly journals and presented papers in national and international conferences. Her research interests include materials evaluation, reading-writing connection, assessment, teaching skills, and other ELT issues.
Maryam Zarei is currently a Ph.D. candidate in TEFL at Islamic Azad University, Tabriz Branch, Iran. She is also an English instructor at Islamic Azad University, Astara Branch, Iran. She has presented and published papers in national and international conferences and journals. Her research interests are individual differences, CF strategies, lesson planning, and task-based research.
Zohreh Seifoori is an associate professor in TEFL at Islamic Azad University, Tabriz Branch, Iran. She has published several papers in international and national journals and presented a number of papers at national and international conferences. Her major research interests include teacher training, learner strategies, learner autonomy, reflective teaching, and critical thinking.

PUBLIC INTEREST STATEMENT
The way a teacher provides CF might influence students' motivation to learn a language. On the other hand, differences between students' attitudes of which instructional practices can result in learning and strategies used by their teachers might have a negative impact on motivation. This study, therefore, set out to investigate whether implicit, explicit, and emergent CF types could affect pre-intermediate Iranian EFL learners' learning motivation, attitudes and perceptions differently. The results revealed that there were meaningful differences among the three feedback types with regard to motivation enhancement. Moreover, emergent feedback significantly raised learners' positive attributes including their motivation and attitudes. Hopefully, this study can help teachers and teacher trainers in finding the right path towards successful teaching and motivating students in their learning. It can also help future teachers of English in developing their own feedback strategies and techniques.
in boosting the positive affective characteristics in learners. This study could specifically increase teachers' and teacher trainers' awareness of the interrelationship between CF types and language learning motivation, attitude and perception.

Introduction
In EFL contexts, the classroom setting represents the original sociocultural milieu since it is a place where language is used interactively, and students encounter the target culture (Ananda, Febriyanti, Yamin, & Muin, 2017). As noted by Ananda et al. (2017), the role of the teachers is very significant here because they can help students develop positive attributes by meticulously tailoring their methodology to the students' needs and interests and engaging them in the learning process. Moreover, teachers are more capable partners whom students can rely on for assistance. A common way to assist learners during classroom interaction is monitoring their performance and providing CF (Ananda et al., 2017). Li (2013) defined CF as teacher and peer responses to second or foreign language learners' inappropriate products, by reformulating the forms or giving clues for corrections. According to Lantolf and Thorne (2007), a fundamental challenge to research based on educational interventions is the issue of how to put the quantity and quality of assistance into operation. A general weakness of current accounts of CF is that they have considered narrowly the cognitive aspects of correction and acquisition, whereas a fuller understanding requires a focus on both the social context of CF and the psychological aspects of individual learners (Ellis, 2010).
As Jin (2012) pointed out, exploring CF within Vygotsky's Sociocultural Theory (SCT) offers this important opportunity to consider the learner as an active rather than a passive receiver who actively participates in a cooperative problem-solving situation and eventually achieves the competence to independently solve the problem. Hence, from the SCT view, learning emerges in social interaction and optimally takes place in the Zone of Proximal Development (ZPD), which is defined by Vygotsky (1978, p. 86) as "the distance between the actual developmental level as determined by independent problem solving and the level of potential development as determined through problem-solving under adult guidance or in collaboration with more capable peers".
Drawing on the premises of SCT, Ellis (2007) referred to the important role of learners in the process of language learning and particularly in the process of CF. He noted that it is teachers' responsibility to understand their learners' attitudes and perceptions since learners determine the appropriateness of teacher's feedback. Moreover, motivation is critical for language learners and their success. The more we understand the factors that develop motivation, the more teachers will be able to help motivate language learners in classroom. Accordingly, this study attempted to examine whether different techniques of oral CF can affect learners' learning motivation differently. Furthermore, it explored the learners' attitudes and perceptions towards different CF types used in the classroom.

Literature review
The following subsections will introduce the theoretical literature with regard to the variables of the current study, including CF strategies, motivation, attitude, and perception. In addition, the empirical studies concerning the relations between CF and motivation, and CF and attitude and perception will be reviewed.

Oral corrective feedback strategies
In any classroom, there are numerous forms of interaction between students and teachers. Oral CF is one of many communication forms which focuses CF on students' speech. In oral CF, students receive information about the success or more likely lack of success of their utterances from their teacher who either corrects them implicitly or explicitly (Lyster, Saito, & Sato, 2013).
Implicit CF is a method where the teacher implies rather than states the existence of an error and might draw the learners' attention to their non-target-like utterance. Therefore, with implicit CF, the correction is elicited without a clear linguistic signal (Loewen & Philp, 2006). On the other hand, explicit CF is indicated by an overt hint of the existence of an error and the provision of the target-like form (Long, 2006). Lyster and Ranta (1997) proposed six different oral CF types, among which recast, clarification request, repetition, and elicitation are classified as implicit feedback, whereas metalinguistic feedback and explicit correction fall under the category of explicit CF.
Recast refers to "the teacher's reformulation of all or part of a student's utterance, minus the error" (Lyster & Ranta, 1997, p. 46). Clarification request is defined as a signal to learners either that the teacher has misinterpreted the utterance or that it was ill-formed; consequently, it requires a repetition or reformulation. In a repetition, the teacher usually with a change in intonation repeats the ill-formed part of the student's utterance to enable him/her to see that the error has been made. In an Elicitation, the teacher elicits the correct form by either asking questions (e.g., How do we say that in English?), pausing to allow the student to complete the teacher's utterance (e.g., It's a … .), or by getting him/her to reformulate the utterance (e.g., Say that again.) (Lyster & Ranta, 1997). Metalinguistic feedback, without explicitly presenting the correct answer, uses either comments or questions regarding the well-formedness of the student utterance (e.g., What do we put at the end of the verbs when we talk about the past?). Finally, explicit correction provides the language learners directly with the correct forms of their ill-formed utterances (Lyster & Ranta, 1997). Aljaafreh and Lantolf (1994), drawing on Vygotskian SCT, introduced a series of mechanisms for effective negative feedback while the teacher and learner are engaged in oral communication and the teacher takes the initial step to adjust the error correction based on the learner's development until the learner reaches the state of self-correction. They proposed that the teacher's intervention as a more experienced member in the oral communication activity should be graduated. That is, it should discover the learner's ZPD in order to offer the appropriate level of help and to encourage the learner to function at his or her potential level of ability. To achieve this goal, the teacher should concentrate on emergent feedback, meaning that he/she should start his/her assistance at a highly implicit level and gradually move to more explicit and concrete feedback until the appropriate level is reached. Moreover, the teacher's help should be contingent, meaning that it should be removed as soon as the learner shows signs of self-correction and ability to perform independently.

Learning motivation
Motivation is a highly complex term which is used not only in everyday life but also in many areas of social sciences, for instance, in various branches of psychology, educational studies, and applied linguistics (Dornyei, 2001). Learning motivation, as one of the most important psychological concepts in education, can be conceptualized as students' energy and desire to engage in learning and plays a large part in students' interest to be engaged in school and study (Martin, 2003). Richards and Schmidt (2002) defined L2 learning motivation as the combination of the learner's attitudes and willingness to spend a lot of effort in order to learn the second language.
Student motivation can be influenced by both internal and external factors. Internal factors include students' individual characteristics, interests, responsibility for learning, values, and perceived ability. And, external factors include the types of schooling experiences that promote or inhibit motivation (Grombczewska, 2011;Jalongo, 2007). Margolis (2010) added that during a lesson, the way a teacher corrects students' errors can influence their motivation to learn a language, and good CF strategies can boost student motivation.
According to Turner (2001), contexts which are helpful to L2 learners turn into affordances. For example, a thoughtful person who is fluent in the target language and presents appropriate CF in daily conversations can provide an optimal learning situation for the L2 learner. On the contrary, a teacher-centered L2 classroom where there are few opportunities for meaningful L2 interactions may conflict with the learning goal of an L2 learner who dreams of developing L2 communication skills (Lantolf & Genung, 2002). Based on SCT, the L2 learning in the former environment serves the L2 learner's self-image within ZPD through appropriate linguistic scaffolds from more capable other, and the L2 learner can maintain or even boost L2 learning motivation. On the contrary, in the latter situation, the L2 learner's self-image does not function within the ZPD, and the learner's motivation may gradually decrease (Kim, 2009).
A number of experimental and non-experimental studies have reported positive results for the relations between CF and motivation (e.g., Abdollahifam, 2014;Guilloteaux & Dornyei, 2008;Hamidun, Hashim, & Othman, 2012). Guilloteaux and Dornyei (2008) research of EFL students in South Korea investigated the effect of language teachers' teaching strategies on students' language learning motivation. The results demonstrated a highly significant positive correlation between teachers' practices which comprised CF in the form of prompts and encouraged students to self-correct and learners' motivated performances. Hamidun et al. (2012) employed an action research to examine the impact of teacher's feedback in fostering EFL students' motivation in Thailand. Through classroom observation, it had been noticed that students had little motivation to participate in language production. Hence, direct immediate feedback was used to boost up their motivation. It was found that the participants responded well to the direct feedback and the feedback increased their level of motivation. Abdollahifam (2014) attempted to discover the usefulness of the interactional CF on Iranian EFL students' learning motivation. The control group received only direct feedback on form and organizational structure, whereas the experimental group also received additional comments from the teacher on their personal ideas about the topic in their production, and in some cases request for further elaboration. The study indicated the positive effect of interactional feedback on learners' learning motivation.

Attitude and perception
Attitude refers to a positive or negative evaluative reaction to the object, person, class, language, or event inferred on the basis of the individuals' beliefs or opinions about them (Gardner, 1985). As Prosper (2002) mentioned, in the process of second or foreign language learning, a learner can develop different attitudes and perceptions towards the learning situation. These attitudes and perceptions may be helpful or detrimental to successful language learning. According to Martinez (2008), a teacher's CF may also seriously influence learners' attitudes and feelings when the teacher's feedback is inconsistent, and they are unable to fully understand it. Brown (2009) asserted that to avoid students' unrealistic expectations and the sense of failure, teachers should devote some time to investigate and understand their students' attitudes and perceptions on teaching and learning and discuss the differences in their expectations. This does not mean that teaching strategies should get students' approval, but, at the same time, the gap between teachers' and students' perceptions should be removed. Consequently, there have been some attempts to investigate learners' preferences in relation to oral CF strategies (e.g., Abedi, 2015;Amador, 2008;Ananda et al., 2017;Smith, 2010;Yoshida, 2008). Amador (2008) conducted a survey among English students in the University of Costa Rica's School of Modern Languages in order to find their preferences for different error correction techniques. The results showed that the students asked for those techniques in which they were explicitly told what their mistake was. In Japanese EFL classrooms, Yoshida (2008) explored teachers' and learners' views for CF types. The findings revealed that recasts were the most favored CF type by the teachers. On the contrary, the learners had a strong preference to have an opportunity to think about their errors in order to find the correct forms before they receive CF from their teachers.
Smith (2010) examined the affective impact of oral error correction and learners' preferences regarding when and how their errors should be corrected. The participants were adult ESL in a small rural town in Central Florida. Data analysis showed that the learners preferred to be corrected individually and immediately after their errors. Ninety percent of the participants reported feeling positive or neutral emotions when being corrected. Hence, the results of his study showed that CF did not create a negative experience for learners.
In an Iranian EFL context, Abedi (2015) investigated learners' opinions about their teachers' oral error correction practices. The results indicated that learners' opinions were not in line with their teachers' actual error correction practices. That is, they wanted to receive CF indirectly, but their teachers used direct CF strategies. Ananda et al. (2017) attempted a qualitative study among EFL students in Indonesia to find out their preferences toward oral error correction given by their lecturer. The results showed that repetition was the most wanted kind of oral CF which students preferred.

Statement of the problem
If learners are not satisfied with teachers' CF strategies which are provided as one of the most decisive techniques in language classes, they might get disappointed and fail to continue their learning activities during language instruction (Zhao, 2015). Unfortunately, some researchers such as Williams, Abraham, and Negueruela-Azarola (2013), Sultana (2015), and Van Compernolle and Henery (2015) have reported that most of the teachers do not know different approaches of CF and how to use them. They are particularly unfamiliar with sociocultural pedagogies that emphasize dialogic mediation. Therefore, they are following just one technique and are not taking any attempt to apply other approaches.
Some studies have attempted to find the relations between teachers' CF and learners' motivation, attitude and perception (e.g., Amador, 2008;Ananda et al., 2017;Guilloteaux & Dornyei, 2008;Hamidun et al., 2012;Yoshida, 2008). However, the existing literature has not specified whether various CF types, including implicit, explicit and emergent feedback, can have differential impacts on learners' learning motivation. Further, they have not considered in detail the learners' attitudes and perceptions toward those specific CF strategies. Zarei, Ahour, and Seifoori (2018) compared the effectiveness of the implicit, explicit and emergent feedback strategies on Iranian EFL learners' oral accuracy, fluency and attitude, but their study did not involve the learners' learning motivation. Also, the dimensions of the attitude focused in their study were not the same with the current study. Thus, the introduction of the concept of the emergent feedback needs further scrutiny into the extent to which this form of feedback might prove more effective compared to the other feedback types. In this regard, the present study aimed to fill the gap in the related literature by examining the following research questions: (1) Do the implicit, explicit, and emergent feedback types have significantly differential effects on learners' learning motivation?
(2) How do the learners feel toward the implicit, explicit, and emergent feedback types?
(3) What are the learners' perceptions regarding the advantages and difficulties of implicit, explicit, and emergent feedback types?

Design of the study
The present study was a mixed-methods research, employing both a quasi-experimental quantitative study, consisting of a pretest, treatment sessions and a posttest, and a qualitative study conducted via interviews.

Participants
Fifty-four-female Iranian EFL learners whose age ranged from 15 to 25 years participated in this study. They were studying the book "American English File 2" in Middle East Language Teaching Center and Taktazan Institute in Tabriz, a large city in the north west of Iran. They were considered as pre-intermediate level learners based on the institutes' placement test which they had already taken in order to be assigned into their classes by the officials. However, the researchers also administered a Preliminary English Test (PET) among the three classes, each including 18 learners, to ensure their proficiency level homogeneity at the beginning of the study. Those learners whose scores were within one Standard Deviation (SD) minus and one SD plus the mean were going to be chosen and the extreme cases be ignored. After the researchers administered the test among the three classes and obtained the results, they found out that there were not significant differences among the classes, and the learners' scores in each class were within the determined criterion; hence, no one was excluded. Consequently, the researchers assigned the classes randomly into three experimental groups (i.e., Implicit, Explicit, and Emergent). Their mother tongue was either Azari Turkish or Persian.

Preliminary English test (PET)
The researchers adopted a version of PET in order to check the initial homogeneity of the classes in terms of their proficiency level. PET, developed by Cambridge English, tested four language skills, and the results for each participant were out of 100.

Target structures
Ellis (2009) mentioned that highly focused CF, which focuses on just one error type, or somewhat less focused CF, which is directed at a limited number of error types are potentially more effective than unfocused CF, which will target all or most of the errors learners commit. Following Ellis (2009), the researchers just focused on the use of simple past, past continuous, and be going to or present progressive for expressing future prior plans, which, according to the language institutes' teachers, were problematic areas for the learners of the study especially in their oral production.

Materials
The teaching materials which were centered on focused oral production tasks were chosen from "American English File 2", written by Latham-Koenig, Oxenden, and Seligson (2014). American English File 2 is designed, generally, to give pre-intermediate learners full skills coverage, and specifically, to elicit specific target forms in their speaking.

Learning motivation questionnaire
To determine the learners' motivational level in learning a foreign language, the researchers employed the adapted version of Taguchi, Magid, and Papi (2009) questionnaire (see Appendix). The original version of the questionnaire consists of ten main components (i.e., criterion measures, ideal L2 self, ought-to L2 self, family influence instrumentality-promotion, instrumentality-prevention, attitudes to learning English, attitudes to L2 community, cultural interest, and integrativeness). The researchers, however, chose the items from those dimensions (i.e., criterion measures, ideal L2 self and attitudes to learning English) that were more relevant for the purpose of this study. Besides, some minor changes were applied in the wording of the questionnaire in order to make it more suitable for the purpose of the study.
According to Dornyei (2005), criterion measures (items 1-6) assess the learners' intended efforts toward learning English, ideal L2 self (items 7-12) refers to L2-specific facet of one's ideal self, and attitudes to learning English (items 13-18) measure situation-specific motives related to the immediate learning environment and experience. The questionnaire had 18 items and was rated by 6-point rating scales ranging from "strongly disagree" (receiving the value 1) to "strongly agree" (receiving the value 6). The content validity of the questionnaire was initially checked by two experienced teachers in this area. The reliability of the whole questionnaire had already been investigated by Taguchi et al. (2009) and some other researchers such as Haji Mohammadi (2017), and a high index of reliability had been reported; however, we also checked the reliability of the selected parts in a pilot study with 30 learners who were representative of the learners in the main study. The Cronbach's alpha reliability of the motivation questionnaire was calculated to be .87, which was acceptable for the purpose of the study.

Semi-structured interviews
The researchers applied semi-structured interviews to a total of 21 learners from the experimental groups to find out their attitudes and perceptions towards the oral CF types in their language classes. The content validity of the designed interview guide for the purpose of the study was checked by getting two experienced teachers' and some EFL learners' opinions in a pilot study. All of the interviews were recorded in their entirety for later analysis.

Procedure
Checking the learners' homogeneity in terms of general language proficiency at the outset of the study, the researchers randomly assigned the three classes into three experimental groups: the Implicit group, the Explicit group, and the Emergent group. Then, they asked them to complete the motivation questionnaire, as a pre-testing instrument, in their own class time in about 10 min. Following the pretest, the teacher who was one of the researchers started their treatments. Each class received the treatment for 10 sessions each about 75 min within a time span of 5 weeks. Every treatment session provided opportunities for all learners to be engaged in one or two monolingual focused oral tasks that elicited the target form. For instance, they were asked to retell a story to the class about a set of pictures which they read or listened about. Or, they were asked to describe some related pictures while the initial sentence was constructed by the teacher.
Learners in the Implicit group received just recast for their erroneous oral production. That is, whenever they made errors in the use of target forms, the teacher reformulated the erroneous form in her echoing the learners' statement. The excerpt below clarifies the teacher's attempt in providing recast for the Implicit group.

Excerpt 1
L: Tomorrow Jack visits his rich uncle. T: Yes, Jack is going to visit his rich uncle.
In the Explicit group, if a learner made an error in the target areas, the teacher produced the correct form directly while giving some information to specify that their utterance was ill-formed. Explicit correction provided by the teacher can be seen in the following excerpt.

Excerpt 2
L: Yesterday they go the zoo and … T: They went to the zoo. You should use went instead of go.
The Emergent group took CF from implicit to explicit which was withdrawn by the teacher when the learner showed signs of self-correction for their erroneous production. This means whenever they made errors in using the target forms, the teacher began with the most implicit type of CF, that is, recast. If this strategy failed to produce a response, the teacher asked the learner for more clarification. If this also failed to prompt a response, she adopted repetition strategy through repeating the erroneous part. If this strategy also failed, she offered elicitation strategy. If the learner was still unable to recognize the error, the teacher gradually went to the most explicit CF types, including a metalinguistic explanation and finally an explicit correction. The excerpt below illustrates learning in the emergent CF-focused classroom: In the last session, the researchers conducted the same language learning motivation questionnaire in order to determine the possible effect of program on learners' learning motivation. When the treatment sessions were finished, seven learners from each group were randomly selected and invited to have an individual semi-structured interview with the one of the researchers rather than the teacher in order to encourage them to answer the questions honestly. During these interviews, she asked questions with regard to the learners' attitudes toward the CF technique that they had received in their classes and their perceptions toward different oral CF types.

Data analysis
The researchers applied a one-way analysis of variance (ANOVA) to check the initial homogeneity of the classes. In order to answer the first research question, which looked into the impact of three different CF types on the learners' learning motivation, the researchers employed a one-way analysis of covariance (ANCOVA).
To answer the second and third questions of the study, which considered the learners' attitudes and perceptions toward the CF types, two of the researchers conducted the analyses of the transcribed recorded interviews separately; then, compared their analyses and reached an agreement. Following Roothooft and Breeze (2016), they described and compared the answers of the interviewees for each question, so that they could identify the themes that emerged repetitively. Answers were also coded according to whether they showed positive, negative or mixed attitudes.

Results of the preliminary English test (PET)
A PET was applied to test the initial homogeneity of the three classes in terms of their general proficiency level. Table 1 demonstrates the descriptive statistics of the classes. To compare the mean scores of the classes, the researchers conducted a one-way ANOVA. The results revealed that there was not a significant difference among the classes regarding their general language proficiency (F = .192, p = .82 > .01). Table 2 shows the results of one-way ANOVA.

The differential effects of CF types on motivation
The first research question asked whether the implicit, explicit, and emergent feedback types have significantly differential effects on the learners' learning motivation. Table 3 shows the distribution of the pre-test and post-test motivation scores of the three groups.
A one-way ANCOVA was conducted to find any significant difference among the groups regarding their learning motivation. Initially, the related assumptions, including normality and homogeneity of variances, were checked, and then, ANCOVA was run. The results of the test of normality are demonstrated in Table 4. Since the significance values for all the three groups in the pre-test and post-test were more than alpha level (.01), it could be concluded that the distributions of the data were normal.  Based on Table 5, the assumption of the homogeneity of variances in motivation scores (F = 5.53, p > .01) was met. This means that the variances of the motivation scores among the three groups were equal.
The results of ANCOVA for motivation scores are shown in Table 6. In this analysis, the effects of pre-test scores were removed from the post-test scores and then, the three groups were compared based on the remained variance. The results demonstrated a significant difference among the three groups regarding their motivation, F (2, 50) = 7.39, p = .002 < .01, partial eta squared = .22. It means about 22% of the improvement in motivation scores was related to the treatment, which compared to the Cohen's (1988) criteria (.01 = small, .06 = moderate, .14 = large), it indicated a large effect size.
According to Table 7, the adjusted means of the Implicit, Explicit, and Emergent groups in the post-test after controlling their differences in the pre-test with regard to motivation were 85.18, 92.99, and 94.09, respectively, implying that the Emergent group outperformed the other groups with respect to motivation. After the Emergent feedback, the Explicit and Implicit feedback were in the second and third positions, respectively, in terms of their effectiveness on motivation.

The learners' attitudes towards CF types
The purpose of the second research question was to find out how the learners felt toward the implicit, explicit, and emergent feedback types in their language classes. In the Implicit group, five out of seven learners emphasized that they preferred the teacher obtain the correct form from the own learner. In fact, they had negative attitudes toward their teacher's strategy of correction (i.e., recast). However, the other two learners expressed that they liked to be corrected by the teacher indirectly. These two learners enjoyed the correction they received in their class and had positive attitude toward recast.
In the Explicit group, all learners reported that they preferred the teacher initially extract the correct form from the learner. They added that in the case they were not able to correct themselves, the teacher could present more examples with explanations. The analysis of their attitudes showed that they had negative feelings toward their teacher's strategy of correction (i.e., explicit correction). The learners in the Emergent group all reported their high preference for their teacher's CF strategy which indicated the priority of the emergent feedback. They mentioned that they preferred the way their teacher corrected them since they were gradually corrected. Thus, the attitudes of the Emergent group were completely positive toward their teacher's strategy of correction.

The learners' perceptions of the advantages and difficulties of CF types
The third research question focused on the advantages and difficulties experienced by the learners regarding their teacher's CF strategy. One main advantage expressed by the majority of learners in the Implicit group was related to the positive affection this type of strategy had on the learners. That is, the teacher's correction did not lower their value in front of the other learners. Two excerpts of their views are presented below: L3: The way I was corrected did not make me embarrassed in front of the others. L14: I did not lose my face in the class.
The Explicit group reported that although they did not prefer their teacher's strategy of correction, it was better compared with some classes where teachers did not provide feedback to learners' errors. In other words, according to them, their errors were important to the teacher. Due to similarity in their responses to this question, one excerpt is mentioned here: L16: There are some teachers whose learners' errors are not important to them, but our teacher corrects us.
Unlike the Implicit and Explicit group, the Emergent group referred to a variety of benefits they experienced with their teacher's CF strategy. They stressed that their teacher's strategy reinforced their mind and memory. Another benefit was that it motivated them to learn more. They also noted that they learnt how to say something and how to correct it. L12 and L14 expressed: L12: The way the teacher gave feedback to us made us not expect the direct correction from her, and we also worked for extra learning. L14: We learnt how to make and repair our speech.
The other interview question was related to the difficulties faced by the learners while receiving feedback in their classes. Unlike some of the learners in the Implicit group who highlighted just the already mentioned advantage of recast, others mentioned to one apparent difficulty. It concerned the difficulty of recognizing where their problem was. Two excerpts below highlight this difficulty: L11: It was somehow difficult for me to understand what the teacher was doing.
L3: It took some sessions for me to find that the teacher was correcting me.
For the Explicit feedback, two noticeable difficulties were identified. The first was it did not have the learner find the solution and understand it. The other was forgetting the point in which the error was committed. In fact, they believed that since the teacher did not ask them to participate in the process of error correction, they easily forgot what was corrected. L5 reported: L5: Because we did not look for our errors too much, we forgot the point quickly.
The Emergent group did not report any difficulty for the feedback that their teacher provided to their errors. L7 put it in this way: L7: I did not have any difficulty. The teacher helped me to correct myself, but if I could not correct my error, she corrected me and I found it out.

Discussion
This study initially set out with the aim of determining the possible differential effects of implicit, explicit, and emergent feedback types on learners' learning motivation. The results of the study indicated that there were statistically significant differences among the three CF types regarding their effectiveness on motivation. Moreover, emergent feedback had the most positive influence on learners' learning motivation.
These findings are in agreement with Lee (2016), whose study revealed that teachers' oral CF had positive impacts on the students' affective variables. The findings further support Guilloteaux and Dornyei (2008), which showed a highly significant positive correlation between teachers' CF practices and learners' motivated performances. The obtained results also corroborate with the assertion of Lewis (2002) and Brophy (2004). Lewis (2002) suggested that CF is a form of motivation since it can encourage learners to study and do their best. Brophy (2004) mentioned that the degree to which a particular motivation develops, as well as the qualitative nuances it takes on in the individual person, are influenced by the socialization provided by significant others including teachers in the person's social environment. The socially oriented treatment of learner errors which was rooted in SCT was emphasized in terms of the emergent feedback in the current study.
With the Emergent group, this study, drawing on Aljaafreh and Lantolf (1994), emphasized the negotiated error correction process and individual differences, paying particular attention to the relevance of CF types determined on the basis of learner's responses. That is, the teacher focused on those learners' abilities which were on the edge of emergence and assisted them to extend their current competence by equipping them with the assistance which was in tune with their ZPDs. Hence, the learning context providing appropriate CF might have afforded an optimal language learning environment for the learners. Consequently, the language learning motivation among the learners enhanced considerably.
The learners' responses to the interview questions also revealed that the majority of the learners in the Implicit group and all of the learners in the Explicit group preferred to participate in the process of error correction. Besides, the Emergent group reported their high preference for their teacher's CF strategy which prioritized the emergent feedback. The results obtained are in line with Yoshida (2008), who found that the learners preferred to have an opportunity to think about their errors before receiving direct CF from their teachers. The positive attitude for the emergent feedback can be explained by the fact that in the emergent feedback, learners' abilities were considered as emergent which were increased through interaction in the social context particularly within learner-teacher interactions in the classroom. Accordingly, as Aljaafreh and Lantolf (1994) proposed, the teacher focused on those abilities which were on the edge of emergence and assisted learners to extend their current competence by equipping them with the assistance which was in tune with their ZPDs.
In response to the advantages experienced by the learners regarding their teacher's CF strategy, the Implicit group reported that the recast used by their teacher maintained their face and value in front of the others in the classroom. This is in agreement with Yoshida (2008), who suggested that implicit feedback is less abrupt and carries less risk of intimidating or embarrassing the student. Although the Explicit group was not too much satisfied with the explicit correction, they emphasized on the importance of the correctness of their utterances for their teacher. This represented that the learners generally had positive attitudes toward receiving CF from their teacher. The advantages cited by the Emergent group for the emergent feedback were reinforcement of mind and memory, encouragement for additional learning, and involvement in the process of learning and correction. These advantages can be attributed to the fact that in SCT, from which emergent feedback is proposed, education is concerned with making learning experiences meaningful and relevant to the individuals, and with assisting them to develop and grow as a whole person (Williams & Burden, 1997). Kim (2009) supported that from the SCT perspective, appropriate linguistic scaffolds within the L2 learner's ZPD and from more capable other(s) can help the learner maintain (or even enhance) L2 learning motivation.
The Implicit group's responses to the difficulties facing in the classroom with regard to their teacher's CF strategy revealed that they could not recognize where their errors were. This finding is consistent with the idea of Panova and Lyster (2002), who argued that the recasts are inherently ambiguous, such that their reformulating corrective function is not transparent. The Explicit group noted that after getting feedback, they did not understand their errors sufficiently and forgot the corrected errors easily. Unlike recast and explicit correction, no difficulties were identified with respect to emergent feedback. A possible explanation for the difficulty experienced with the explicit correction is that teachers' explicit correction, due to its total isolation from meaningful context, might serve to develop learned linguistic knowledge but not true competence (Ellis & Sheen, 2006); hence, it cannot produce satisfactory results during communicative activities (Brown, 2007;Richards & Rodgers, 2001). This also corroborates with Ryan (2012), who noted that since in the explicit correction, the teacher gives the corrected form, there is no chance for the student to work out why their form was incorrect. This lack of understanding can lead to the noninternalisation of the corrected form because if the student does not understand the reason why their utterance was incorrect; then, they have no grounded way of modifying their hypothesis.

Conclusion
Based on the findings, the researchers came to conclude that learners' learning motivation can be influenced differently by teachers' various CF strategies. Furthermore, the emergent feedback can have the most positive influence on learners' learning motivation. Although recast, due to its implicit nature, does not cause the learners to be ashamed while they are corrected, the learners have somehow difficulties in detecting their errors which are corrected indirectly in the form of recast by the teacher. Moreover, teacher's explicit correction alone is not preferable by the learners though their opinions confirm that CF should be done in their language classroom context. The learners' ideas reveal that they have positive attitudes toward the emergent feedback since it provides opportunities for them to participate in the process of error correction.
The findings can make noteworthy contributions particularly to teachers and teacher trainers. Teachers should recognize that in their teaching sequence, as Aljaafreh and Lantolf (1994) pointed, different learners might have different ZPDs in using the same structure, and the same learner might have different ZPDs in using different structures. Therefore, learners would need different levels of help provided by the teacher to obtain enough motivation to continue the process of learning. The results of this study can also present suggestions to teacher trainers since teacher training programs can help student teachers know about how to give appropriate approaches of feedback in their classrooms. By being sensitive to student teachers' individual differences and applying pedagogies that emphasize dialogic mediation for promoting students' motivation in their training courses, teacher trainers can partly familiarize the student teachers with these types of pedagogies.
The participants of the current study were pre-intermediate level learners in an Iranian EFL context. Therefore, other studies can explore the impact of these CF types on learners' motivation in other levels of proficiency (e.g., beginners). Furthermore, future studies can investigate how other contexts benefit from different CF types in terms of their motivation development. Finally, it can be useful for future research to examine whether the role of oral feedback and different ways of correcting especially the emergent feedback are included in teacher training programs, and to what extent these programs are able to influence feedback practices.

Learning Motivation Questionnaire
Dear Participant, The following questionnaire is designed to investigate your motivation in learning a foreign language. The results of this survey will be used only for research purpose so please give your answers sincerely. Thank you very much for your cooperation! For each item, indicate how true it is for you, using the following scales:

Disagree
Slightly disagree Slightly agree Agree Strongly agree 1 2 3 4 5 6 1 Teacher correction in the classroom motivates me to take an English course which is offered in the future. The way teacher corrects me in the classroom makes me be prepared to expend a lot of effort in learning English. Teacher correction in the classroom makes me concentrate on studying English more than any other topic. The way teacher corrects me motivates me to volunteer to do an optional assignment which is given in the classroom.