1 introduction

As one of the main means of modern teaching, teacher-student dialogue teaching is widely used in college English teaching, and this practice is supported by corresponding theories, such as sociocultural theory and cognition-interaction theory. The teacher-student interaction practice is the prerequisite of students' interactive practice. Relevant experimental data also show that teacher-student interaction is of great help to the improvement of students' English proficiency.

Experts have conducted similar studies before, which are helpful for us to understand the interaction between teachers and students, and it also provides relevant theoretical support and practical experience for classroom practice teaching. However, the previous research focus of experts was mainly from the cognitive perspective, such as the social, behavioral, emotional, self and other perspectives were not considered. As far as learning itself is concerned, its learning effect is affected by many factors, such as self, social factors, emotional factors, etc. Of course, these factors will interact with each other to have a comprehensive impact on the learning effect. Therefore, students should not only learn from the perspective of language cognition, but also put more effort into the social, emotional and behavioral aspects to maximize the learning effect. Social, emotional, and behavioral factors are playing an increasingly important role in classroom teaching as experts continue to study them. Therefore, in order to have a comprehensive and clear understanding of teacher-student interaction, we must comprehensively consider various factors and study the performance of teachers and students in the interaction from multiple perspectives at the same time, rather than being limited to a certain factor.

2. Literature review

Some experts have used familiarity as a dependent variable in dialogue teaching before. In1993, Plough and Gass conducted a relevant study, which showed that compared with conversational participants with low familiarity, participants with high familiarity would be more efficient in terms of communication efficiency, semantic supplement and meaning discussion. The more familiar the conversation participants are, the more willing they are to engage in deep meaningful communication, and they are more likely to regard the conversation as a formal communication. After that, there are few studies that put familiarity into conversation teaching, and experts have shifted their attention to other factors, but the few studies that do offer some reference for later studies.

3. Research Methods

3.1 Research object

The50 students are all first-year students majoring in accounting and finance from a university in Anhui. Their average age is18 years old, and there are23 male students and17 female students. The two teachers, one female and one male, are also from the same university's Public English Teaching Department. In the course of teaching, the same textbook is used, and the teaching plan is designed exactly the same as the teaching time. The average score of the two classes (25 students respectively) is about the same, so there is not much difference in English proficiency. Before the beginning of this study, teachers and students are familiar with this activity because they have practiced the classroom teaching that mainly based on dialogue teaching for many times and recorded the conversations between teachers and students. The50 students and2 teachers are free to form a teaching class, with one teacher for each class and25 students to form a "familiar class". Then, the50 students were disrupted, and the researchers randomly divided them into two classes, and the two teachers were also randomly assigned to form an "unfamiliar class".

3.2 Task materials

Dialogue teaching mode is one of the main teaching methods used in college English classes. That is to say, the teacher gives a discussion topic, and the students and teachers discuss on this topic and express their own opinions, so as to promote the absorption of knowledge in the discussion. Therefore, in order to meet the requirements of dialogue teaching, the mode of free discussion between teachers and students is adopted, and the topics of discussion can be selected from the textbook. Topics can be environmental protection, modern technology, artificial intelligence and so on. Considering the teaching time of each class, teachers and students can complete a topic discussion between each other.

4、discussions

In terms of cognitive engagement, the familiar class was higher than the unfamiliar class in language acquisition, conversational language and meaning negotiation (see Figure1). According to the multivariate variance test, the familiarity of teachers and students is an important factor affecting the learning effect, η2=0.124, P =0.027, Lambda=0.745, F(3,56)=2.475. Further study on the data showed that the familiarity of teachers and students had a certain influence on the meaningful negotiation.

η2=0.122, P =0.003<0.015,F(1,58)=8.20. In teacher-student interaction language, η2=0.006, P =0.407,F(1,58)=0.356 and related language acquisition η2=0.024,F(1,58)=1.623, The effect of P =0.178 has not reached a certain extent.

		N	M	SD	F	Sig	Partial Eta Squared
Language Acquisition	Familiarity Class	26	1.74	1.18	1.601	0.173	0.024
	Unfamiliarity Class	26	1.23	1.24
Conversational Language	Familiarity Class	26	6.18	2.74	0.342	0.467	0.006
	Unfamiliarity Class	26	5.94	3.61
Meaning Consultation	Familiarity Class	26	0.64	0.79	8.979	0.003	0.129
	Unfamiliarity Class	26	0.29	0.29

As can be seen from the above figure, familiarity between teachers and students has a great influence on meaning negotiation. When teachers and students who are familiar with each other encounter communication difficulties, they will constantly use communication strategies such as repeating language and clarifying opinions to eliminate communication difficulties. The above situation was also confirmed in the interview. The two teachers agreed that students would feel embarrassed because they could not answer questions, so teachers would try their best to start from basic questions and explain their problems step by step. As far as students are concerned, in order to save their own face, when the teacher raises some questions that they do not understand, they will repeat the words to shift the teacher's focus from the question itself to others, which seems to help them out.

But there are also some exceptions, such as in the familiar class, there will be a teacher to save the student's face. A teacher said in the interview like that: in most cases all accept it, but there are exceptions, teacher asks a student to answer question, students seem to not understand what teacher said, but he still pretend to understand what teacher mean, and goes and answered teacher’s question, because the teacher has close relations with the student, so the teacher also can't let him explain again. Sliman-Rolls has said in previous research that if teachers and students who are close to each other often work together, some problems can appear over time. Aston also said that the long-time use of meaning negotiation teaching method may have some impact on the teacher-student relationship. According to the above experimental results and experts' opinions, social factors are an important factor affecting dialogue teaching. If we ignore the influence of this factor in dialogue teaching, we cannot accurately understand the investment behavior of teachers and students in dialogue teaching.

In this study, in terms of language acquisition, the gap between the familiar class and the unfamiliar class is not obvious, which means that the familiarity of teachers and students does not have a great impact on language acquisition. This is different from the research of Mozaffari. From the experimental data, Mozaffari found that compared with the unfamiliar teacher-student relationship, the language acquisition effect between familiar teachers and students is poor, but the output of topics unrelated to teaching is relatively large. Mozaffari believes that the familiarity between teachers and students will have a significant impact on the effectiveness of teacher-student dialogue teaching. However, in this experimental study, we did not find any topics unrelated to classroom teaching. This also verifies that not only the familiarity of teachers and students will have an impact on language acquisition from another aspect , but also other factors will have an impact on it.

In terms of social involvement, the familiar class had an advantage over the unfamiliar class in terms of cooperative sentence completion. However, in terms of feedback, the unfamiliar class had an advantage over the familiar class (see Figure2). According to the multivariate data analysis of variance analysis, it can be seen that the familiarity of teachers and students has a great impact on social investment, Lambda=0.623,F(2.16)=12.654,P=0.001, skew η2=0.311. Through further analysis of the experimental data, it can be seen that the familiarity of teachers and students has a significant effect on both the cooperative completion sentences and the feedback sentences. The cooperative completion sentences are F(1,16)=12.546, P =0.002, η2=0.186, and the feedback sentences are F(1,16)=8.125, P =0.001, η2=0.125. In terms of familiarity between teachers and students, when teachers and students are more familiar with each other, there will be more complementary dialogues, which is conducive to the improvement of teaching effect. When the teacher-student relationship is relatively unfamiliar, teachers will use more feedback to encourage students to complete the dialogue.

		N	M	SD	F	Sig	Partial Eta Squared
Feedback language	Familiarity Class	26	1.21	1.01	12.765	0.001	0.173
	Unfamiliarity Class	26	0.39	0.49
Combine complete sentences	Familiarity Class	26	2.14	1.81	8.234	0.003	0.129
	Unfamiliarity Class	26	4.56	3.47

In terms of cooperative sentence production, the familiar class has more advantages, which indicates that the more familiar the teachers and students are, the stronger their cooperative desire will be. Among them, the social relationship between teachers and students is an important factor. The familiar teacher-student relationship generally presents a state of friendship, so in the dialogue teaching, it is easier for teachers and students to understand each other's inner meaning, and they can't help complementing each other's words. Therefore, the teachers and students said that because they were familiar with the students or teachers, they would timely supplement each other's words. For unfamiliar teachers or students, when they encounter ineffective communication, they will find it difficult to explain their own ideas. The results of this study are consistent with those of Plough and Gass. In the process of research, with the constant contact between teachers and students, their familiarity is becoming deeper and deeper, and both sides will gradually supplement each other's words.

In addition, the data also showed that there was more feedback between teachers and students who were unfamiliar with the class than those who were familiar with the class. This may be associated with the idea of cooperation between teachers and students who are not familiar with the class. Unfamiliarity between teachers and students will produce more obstacles in dialogue teaching. In order to eliminate these obstacles, teachers and students may use more feedback to show that they have responded to each other very actively. In the interview, some students or teachers described it like this: they want to leave a good impression on the other side. However, some teachers and students said that in dialogue teaching, when the other party can not understand the meaning of the feedback, but just out of respect for each other. Therefore, the teacher-student cooperative dialogue teaching may only be a surface phenomenon, because teachers who are not familiar with dialogue teaching or students with weak dialogue expression ability will use feedback to indicate the other. It also proves that unfamiliar classes rarely produce meaning negotiation in the cognitive dimension.

In terms of emotional investment, both familiar and unfamiliar classes had higher emotional investment (see chart3). In terms of the five sub-dimensions of total emotional immersion, process teaching perception, teacher-student interaction attitude, attitude towards each other and task evaluation, the familiar class has advantages. The results of multivariate analysis of variance show that the familiarity of teachers and students has a great influence on affective investment. Aiaa Lambda =7.436, F (4) =7.436, p =0.001, partial eta2 =0.209. The further experimental data show that the familiarity of teachers and students has a great influence on the interaction attitude of teachers and students, F(1,32)=11.785,P=0.002, η2=0.084. The analysis of the familiarity of teachers and students on the task evaluation data shows that F(1,32)=0.199,P=0.594, η2=0.001. F (1,32) =2.379, p =0.106, partial eta2 =0.011, and the attitude to treat each other F = (1,32) =4.199, p =0.018, the influence of partial eta2 =0.037 are not obvious. Interviews with teachers and students also confirmed the above conclusions. Both teacher and fifty students said that the interactive in the dialogue teaching can effectively stimulate the learning enthusiasm of all teachers, teacher or students also not completely reject the other to complete the dialogue task, but if there are familiar between the teacher and students can be chosen, they will choose familiar team so as to avoid embarrass. This is consistent with previous studies. In a previous study, after10 weeks of teacher-student interaction, fixed teachers and students became more and more satisfied with conversational teaching, while non-fixed teachers and students gradually lost interest in conversational teaching. Experts believe that the familiarity of both parties is an important factor affecting the teaching of dialogue.

		N	M	SD	F	Sig	Partial Eta Squared
Total emotional involvement	Familiarity Class	26	3.99	0.35	3.467	0.054	0.024
	Unfamiliarity Class	26	3.87	0.32
Teacher-student interaction attitude	Familiarity Class	26	3.67	0.39	0.195	0.585	0.002
	Unfamiliarity Class	26	3.70	0.45
Process teaching perception	Familiarity Class	26	4.03	0.45	2.285	0.106	0.019
	Unfamiliarity Class	26	3.86	0.41
Attitude towards each other	Familiarity Class	26	4.13	0.39	11.985	0.001	0.086
	Unfamiliarity Class	26	3.87	0.40
The task to judge	Familiarity Class	26	3.89	0.46	4.954	0.019	0.038
	Unfamiliarity Class	26	3.69	0.42

5. Findings

The experimental study shows that the familiarity of teachers and students is an important factor affecting the learning effect of dialogue teaching. The influence is mainly reflected in the following three aspects: First, In the cognitive dimension of meaning negotiation, the familiar class is more effective than the unfamiliar class. Second, in the social dimension of the cooperation to complete the sentence, familiar class is more positive than unfamiliar class. Third, in the emotional dimension of the teacher-student interaction attitude and attitude towards each other, familiar classes are more positive than unfamiliar classes. However, in terms of feedback, the unfamiliar class used more feedback than the familiar class. In terms of task completion time, word output and turn, the familiar class was higher than the unfamiliar class, but the difference was not very obvious.

参考文献：

1. [1] Review of Peer Interaction Research Methods [J]. Journal of PLA Foreign Languages Institute.2019(06)
2. [2] The Influence of Task Types on Learner Engagement in Peer Interaction [J]. Journal of PLA Foreign Languages Institute.2019(06)
3. [3] Thirty Years of International Second Language Whisper Research [J]. Xu Jinfen, Fu Hua. Foreign Language Journal.2019(05)
4. [4] An Empirical Study on the Influence of Different Pair Patterns on Students' Interaction in College English Classroom [J]. Xu Jinfen, Cao Zhongkai. Foreign Languages in China.2012(05)
5. [5] Decemption in Intercultural Communication -- The Use of Back-channeling Strategies in Foreign Language Learners' Communication and Its Implications [J]. Huang Qinggui. Journal of Fujian Normal University (Philosophy and Social Sciences Edition).2008(03)
6. [6] Engagement with language: interrogating a construct [J]. Agneta Marie-Louise Svalberg. Language Awareness .2009 (3-4)