Dialogue through the eyes : exploring teachers’ focus of attention during educational dialogue

This is a self-archived version of an original article. This version may differ from the original in pagination and typographic details. Author(s): Title: Year: Version: Copyright: Rights: Rights url: Please cite the original version: CC BY-NC-ND 4.0 https://creativecommons.org/licenses/by-nc-nd/4.0/ Dialogue through the eyes : exploring teachers’ focus of attention during educational dialogue © 2020 Elsevier Ltd Accepted version (Final draft) Muhonen, Heli; Pakarinen, Eija; Rasku-Puttonen, Helena; Lerkkanen, MarjaKristiina Muhonen, H., Pakarinen, E., Rasku-Puttonen, H., & Lerkkanen, M.-K. (2020). Dialogue through the eyes : exploring teachers’ focus of attention during educational dialogue. International journal of educational research, 102, Article 101607. https://doi.org/10.1016/j.ijer.2020.101607 2020 Running Head: DIALOGUE THROUGH THE EYES 1 Dialogue through the eyes: Exploring teachers’ focus of attention during educational dialogue Heli Muhonen, Eija Pakarinen, Helena Rasku-Puttonen, & Marja-Kristiina Lerkkanen Author notes Heli Muhonen ([email protected]); Eija Pakarinen ([email protected]); Helena Rasku-Puttonen (helena.r[email protected]); Marja-Kristiina Lerkkanen (marja- [email protected]), Department of Teacher Education, University of Jyväskylä, P.O. Box 35, 40014 Jyväskylä, Finland Corresponding author Heli Muhonen, Department of Teacher Education, University of Jyväskylä, P.O. Box 35, 40014 Jyväskylä, Finland. E-mail: [email protected], Phone: +358445508256 Running Head: DIALOGUE THROUGH THE EYES 2 ABSTRACT This study explored teachers’ focus of attention during educational dialogue. Teachers’ focus of attention was recorded in 54 Grade 1 classrooms using Tobii Pro Glasses 2 mobile eye-tracking device. From the video recordings, episodes of educational dialogue were identified and categorised by quality. Teacher’s focus of attention on students was examined during the dialogue episodes. Results showed that teachers allocated their attention relatively unevenly among the students. More students got visual attention during high-quality educational dialogue than during moderate-quality dialogue. This study provides insight into the quality of educational dialogue by combining assessment of the verbal dialogue with observations of the non-verbal focus of teacher’s attention in classroom. Keywords: Educational dialogue; Focus of attention; Eye tracking; Teacher; Student Running Head: DIALOGUE THROUGH THE EYES 3 1. Introduction People’s cognitions on the background of their gaze can be studied from interaction and speech that occur simultaneously (McNeill, 1985). Since gaze behaviour is central to social interaction and its regulation (Pfeiffer, Vogeley, & Schilbach, 2013), it is suggested that more research is needed regarding teachers’ focus of attention and its distribution during wholeclass dialogue, a busy and dynamic learning situation that requires both visual and verbal attention from the teacher. Recent research has highlighted the key role of dialogic interactions in students’ learning, development and reasoning (e.g. Littleton & Howe 2010; Mercer & Littleton, 2007). Therefore, there has been growing interest in studying the diverse qualities and forms of effective educational classroom dialogue. However, previous studies have predominantly focused on the verbal exchanges between the teacher and the students and mostly neglected the non-verbal aspects of educational dialogue. These non-verbal aspects have been studied in terms of gestures, facial expressions, gaze and other symbolic actions used as a way of interacting and conveying meaning (Bavelas, Coates, & Johnson, 2002). However, only a few studies have examined the teacher’s or whole-class verbal interaction in association with the teacher’s focus of attention. In the present study, teachers’ gaze during educational dialogue is not studied as a ‘typical’ non-verbal way of communication but as a way of discovering how teachers distribute their focus of visual attention among the students during classroom dialogue. In interaction among people, varying from two people to larger groups such as classrooms of teacher and students, gaze is an important way of showing attention and of distributing turns (Mason, 2012). A few innovative studies have explored teachers’ focus of attention in classroom situations with their students (e.g. van den Bogert, Bruggen, Kostons, & Jochems, 2014; Cortina, Miller, McKenzie, & Epstein, 2015; McIntyre, Jarodzka, & Klassen, 2019; McIntyre, Mainhard, & Klassen, 2017), but they have predominantly focused Running Head: DIALOGUE THROUGH THE EYES 4 on exploring teachers’ classroom management, communication style or teaching experience (Cortina et al., 2015; McIntyre et al., 2019; McIntyre et al., 2017). To date, hardly anything is known about teachers’ focus of attention during educational dialogue in the classroom, in which shared knowledge is built together between teacher and students. Research focusing on verbal interaction has shown that managing educational dialogue and scaffolding students’ participation and shared knowledge building can be challenging for the teacher (Muhonen, Rasku-Puttonen, Pakarinen, Poikkeus, & Lerkkanen, 2016). Good and effective teaching through dialogue requires teachers to be adaptive and sensitive to critical moments in the discussion, to be able to interpret situations, to have a repertoire of actions to draw on and to display judgment regarding instructional decisions (Lefstein & Snell, 2014). Especially in whole-class educational dialogue, the capabilities required of a teacher include clarity and coherence, adequate representation of content, equitable participation, time management (O’Connor, Michaels, Chapin, & Harbaugh, 2017) and standardised testing of the students’ learning (Segal, Snell, & Lefstein, 2017). Many teachers reportedly feel pressure to engage students equally in classroom interaction (O’Connor et al., 2017). This raises questions regarding how many students actually get the chance to participate vocally in the discussion in a whole-class setting and how many students receive the teacher’s attention during educational dialogue. The present study utilises a mixed-method approach of eyetracking analysis and classroom talk to examine how teachers distribute their visual attention among the students during educational dialogue and to discover whether the distribution of attention varies according to the quality of the dialogue. 1.1. Educational dialogue The characteristics, quality and patterns of effective educational classroom dialogue have been broadly studied and have yielded suggestions for characterising teachers’ and Running Head: DIALOGUE THROUGH THE EYES 5 students’ involvement in dialogue. Howe, Hennessy, Mercer, Vrikki and Wheatley (2019) recently presented some of the vital characteristics of verbal educational dialogue in the classroom, namely: 1) the use of open questions; 2) making extended contributions, elaborations and building on previous knowledge; 3) acknowledging, probing and critiquing different opinions; 4) seeking integrated lines of inquiry, and 5) adopting the metacognitive perspective on verbal interaction. This is largely in agreement with Alexander’s (2000, 2006, 2017) concept of dialogic teaching, which describes five principles for effective educational dialogue. Principles of collectivity, reciprocity and supportiveness represent the type of culture of relationships in which dialogue is likely to emerge and support students’ participation and sharing of their thoughts. The principle of cumulation highlights the dialectic nature of classroom discussion and the stepwise growth of shared understanding. The fifth principle, purposefulness, indicates that, although dialogue is of itself extremely important, the discussion should comprehend an educational goal or content if it is to be developmentally meaningful for the students. Lefstein (2006) has suggested adding two further features of dialogic teaching: criticalness (participants identify and investigate points and explore questions within the group) and meaningfulness (teachers and students relate to the topic and bring their own perspectives to the discussion). Nevertheless, even if the educational dialogue fulfils all the above-mentioned criteria, the quality and pattern of the verbal exchange between teacher and students is likely to vary, since not all dialogues follow the same pattern. Muhonen et al. (2016) characterised educational dialogue by differences in quality. In the coding undertaken by Muhonen et al., higher-quality educational dialogues included versatile and rich scaffolding strategies that were likely to support students’ conceptual thinking, joint understanding and synthesis of ideas and information. Moderate-quality educational dialogues included relatively unitary forms of questioning, less support for active participation and lower support for shared Running Head: DIALOGUE THROUGH THE EYES 6 understanding of content. These two qualities of educational dialogue have been verified in several studies, at different levels of schooling and in both Western and non-Western contexts (see Muhonen et al., 2016; Muhonen, Pakarinen, Poikkeus, Lerkkanen, & Rasku-Puttonen, 2018; Muhonen, Pakarinen, Lerkkanen, Barza, & von Suchodoletz, 2020). Although educational dialogue can also happen between students only, it is acknowledged that the teacher has an important role in orchestrating the classroom discussion and scaffolding students’ learning through it (Gillies, 2013; Rogoff, 2008; Rojas-Drummond, Torreblanca, Pedraza, Vélez, & Guzmán, 2013). However, teachers tend rarely to encourage their students to explain their thinking or broaden the discussion for the whole class to elaborate (Webb et al., 2009). One reason why teachers do not conduct educational dialogue may be lack of both time and the opportunity to involve all the students equally in the discussion (O’Connor et al., 2017). Students seldom take the initiative to explain their reasoning or justify their thoughts without the teacher’s encouragement (Myhill, 2006), which is why active teacher involvement is often needed (Muhonen et al., 2016). The expectation of effective educational dialogue is that all participants (teacher and students) should be highly involved (Howe et al., 2019). O’Connor et al. (2017) studied vocal and silent participants in classroom discussion and found that students in both categories learned the content equally well, suggesting that vocal participation is not a necessary condition for benefiting from such discussion. The present study aims to explore teachers’ focus of attention on both vocal and silent students during educational dialogue in the classroom. 1.2. The teacher’s focus of attention in classroom interaction Although interaction and educational dialogue are important in supporting students’ development and learning in the classroom (e.g. Howe et al., 2019; Muhonen et al., 2018), we are still far from understanding how teachers allocate their visual attention during classroom Running Head: DIALOGUE THROUGH THE EYES 7 interaction. In interaction, eye contact and gaze have been acknowledged to have several major functions, such as asking for feedback or a reply, listening, indicating that the space for discussion remains open or recognition of a social relationship (Argyle & Cook, 1976). Teachers’ visual attention, measured through gaze, represents their ability to process information present in the classroom environment (van den Bogert et al., 2014). In previous research, similar terms such as focus of attention (van den Bogert et al., 2014) and gaze pattern or gaze behaviour (McIntyre et al., 2017; McIntyre et al., 2019) have been used to describe teachers’ visual attention in the classroom. In the present study, the term ‘focus of attention’ is used when referring to teachers’ visual attention, as measured through their gaze behaviour. It is only during the past decade that the technology for mobile eye tracking has been developed and utilised to study focus of attention in authentic classroom interaction (see Cortina et al., 2015; McIntyre et al., 2017, 2019). Fixation data obtained in the form of time duration are the measure most often used in eye-tracking studies to determine the focus of visual attention on an object (Cortina et al., 2015; Yamamoto & Imai-Matsumura, 2012), in the present case on students. In addition, pupil diameter and its dilation can reflect person’s cognitive load and concentration (Beatty, 1982). The greater the concentration and mental effort, the greater the person’s pupil diameter typically is (Rosenbaum, 2010). As mentioned above, studies to date have predominantly investigated teachers’ focus of attention in relation to classroom management, communication style or teaching experience (see van den Bogert et al., 2014; Cortina et al., 2015; McIntyre et al., 2019; McIntyre et al., 2017). Cortina et al. (2015), for example, stated that, although traditionally it has been argued that an effective teacher should monitor his or her classroom by distributing attention evenly, this may not be correct. In fact, an effective teacher is likely to give more attention to students who either have difficulties in learning or actively demonstrate their knowledge. However, the gaze data in themselves are not enough to explain the teachers’ focus of Running Head: DIALOGUE THROUGH THE EYES 8 attention, and additional data such as verbal interaction that happens simultaneously with the gaze are needed to better understand the teachers’ gaze distribution and the rationale behind it (van den Bogert et al., 2014). McIntyre et al. (2017) analysed ‘communicative gaze’ (gaze during talking and sharing information) and ‘attentional gaze’ (gaze during questioning) among expert and novice teachers. They found that expert teachers focused their attention more on students and had greater gaze efficiency during both sharing information and asking questions. The expert teachers were also more flexible when asking questions but showed less flexibility during communicative gaze. Cortina et al. (2015) examined teachers’ focus of attention in association with the quality of classroom interaction. They found that the quality of teacher feedback, meaning teacher’s efforts to encourage students to elaborate on their responses, was associated with teachers focusing their attention more on certain individual students than on all equally. The findings are important, since providing high-quality feedback for the students requires the teacher to focus attention on those students who are receiving the feedback. On the other hand, Cortina et al. (2015) also suggested that experienced teachers are capable of monitoring their whole classroom through gaze while providing feedback or facilitating discussion with individual students. Dessus, Cosnefroy and Luengo (2016) investigated teachers’ classroom awareness during interaction, utilising eye tracking. They found that every student in the classroom received at least some attention from the teacher but very small groups of students received more. They also found very little variability across different pedagogical activities. 1.3. Aims of the study To date, educational dialogue has been predominantly studied by exploring the verbal aspect of the interaction. However, verbal participation, though very important and the most Running Head: DIALOGUE THROUGH THE EYES 15 Average fixation duration per student who participated verbally 12,105.09 16,531.09 533.33 73,352.20 Average fixation count per student who did not participate verbally 6.15 6.52 0.00 28.57 Average fixation duration per student who did not participate verbally 2,284.09 2,231.82 0.00 9,751.71 GC 0.56 0.15 0.33 0.86 Notes. a Number of students per classroom. GC = Gini coefficient 3.2. Variation in teachers’ focus of attention according to the quality of educational dialogue The second research question examined whether teachers’ focus of attention on students varied according to the quality of educational dialogue. The identified episodes of educational dialogue were divided into two groups based on the quality of the dialogue. Thirteen of the dialogic episodes represented educational dialogues of moderate quality and 18 of the dialogues represented high-quality dialogues. Table 2 presents descriptive statistics for teacher’s focus of attention and students’ vocal participation in relation to these two qualities. The Mann–Whitney U test showed a marginally significant difference between the two qualities of educational dialogue in terms of the number of students who received visual attention (U(1) = 38.0, p = 0.051). During high-quality educational dialogue, more students received visual attention from the teacher than during educational dialogue of moderate quality. However, there were no statistically significant differences between the two qualities in fixation count per student (U(1) = 60.0, p = 0.505 ), in total fixation duration per student (U(1) = 51.0, p = 0.235) or in GC (U(1) = 55.0, p = 0.339). In addition, the eye pupil diameter size for the two teachers conducting the moderate and high quality dialogues was calculated. There was no significant difference in the pupil size between the moderate and high quality dialogues (Table 2). Running Head: DIALOGUE THROUGH THE EYES 16 Table 2 Descriptive statistics for the two qualities of educational dialogue. Moderate quality High quality Mean Std. deviation Mean Std. deviation Episode length (in seconds) 181.64 190.29 178.31 96.36 Students in the classroom 15.64 5.14 19.00 3.06 Vocal students 6.64 (43.14%) 3.70 9.00 (47.40%) 4.89 Students who participated verbally in education dialogue a 12.55 (77.21%) 6.09 17.23 (90.28%) 3.79 Students who received visual attention from the teacher a 13.15 12.02 12.18 5.69 Average fixation duration per student (in milliseconds) 5,708.66 4,457.76 5,888.43 2,889.34 GC 0.60 0.18 0.53 0.12 Left eye pupil diameter (in millimetres) 4.09 0.34 4.10 0.22 Right eye pupil diameter (in millimetres) 4.21 0.39 4.10 0.23 Notes. a Number of students per classroom. GC = Gini coefficient Example 1 demonstrates a typical teacher-initiated dialogue of high quality in a science lesson. The verbal discussion in which both teacher and students participate actively is focused on small creatures that overwinter. In the dialogue, the teacher asks one main question (What kind of creature spends the winter under rocks, leaves and branches?), to which children suggest various alternative answers (voles, rats, spiders). The teacher is relatively open to suggestions and asks clarifying questions (Oh, you mean like voles or… You mean those?), even though he/she is looking for one specific answer (beetle). With the clarifying or extending questions, the teacher shows his/her interest in the child and the Running Head: DIALOGUE THROUGH THE EYES 17 knowledge that the child is sharing. Both teacher (Rats are big. Rats don’t hibernate, they don’t go into winter quarters or anything like that.) and students (Voles are scared of people.) share their knowledge and the children participate in the discussion even without the teacher’s continuous encouragement. Example 1. Educational dialogue of high quality T: What kind of creature spends the winter under rocks, leaves and branches? [C2] C2: Mmmm those …. Those little ones. I can’t remember their name but… T: What little ones? C2: I can’t remember the name. T: All right. C2: It’s brown and it looks like a guinea pig but it is… T: Oh you mean like voles or… You mean those? C2: Yeah. T: Well, actually no, but those voles and others spend their winter in their holes and may scurry around above ground; mice sometimes scamper around on the snow in winter. C5: And voles run around on roads. And voles on roads in winter. T: But these ones are really tiny, the ones that spend winter here. So tiny. And even a bit repulsive to me. You knew straight away. Not necessarily creatures that make me run out of the classroom, but there are other ones as well. [C14] C12: A spider. T: Well, yeah, that is a disgusting one. But there are other tiny ones. What could they be called? [C4] C1: Ants. T: [C4]. C4: Rats. T: Not rats, smaller. Very tiny ones. C3: Rats are big. T: Rats are big. Rats don’t hibernate, they don’t go into winter quarters or anything like that. T: [C14] wants to continue. C5: Voles are scared of people and that’s why they go underground. C14: Well… a beetle. T: Yes. All kinds of bugs, for example beetles. C3: Beetles are just fun. C5: And voles are scared of people. T: Well, yeah, in theory they can be fun, but… C5: Voles are really cute. T: … they’re a bit nasty too, I think. C5: Voles are scared of people. T: That’s right, many animals are scared. The same dialogic episode continues around the topic of bugs, and more children join in. Figure 1 (see also Appendix 1, gaze map of the same episode of educational dialogue) Running Head: DIALOGUE THROUGH THE EYES 18 demonstrates how the teacher distributes his/her focus of attention (fixations) among the students during the high-quality educational dialogue reproduced in Example 1. Seventeen out of 18 students (94%) received the teacher’s visual attention (at least one fixation), but his or her attention was not distributed evenly among the 18 students. The GC for the teacher’s visual attention distribution was 0.55, indicating relatively high variation in the teacher’s focus of attention per student (see also Figure 1). On the other hand, 11 out of 18 students (61%) in the class participated in the dialogue vocally by sharing their knowledge and views. Interestingly, the only student (C1) who did not receive the teacher’s visual attention still participated vocally in the discussion. There was no statistically significant difference in the teacher’s focus of attention (fixation duration) between vocal and silent students during the episode (U = 25.0, p = 0.221) (see Table 2). Figure 1. Teacher’s distribution of fixation duration per student during educational dialogue of high quality (Example 1). Example 2 demonstrates a typical educational dialogue of moderate quality. The subject was literacy. The verbal discussion focused on the famous Finnish author Aleksis 0 5000 10000 15000 20000 25000 30000 1* 2* 3* 4* 5* 6 7 8 9* 10* 11* 12* 13 14* 15 16 17* 18 Fixation duration (ms) Students Note. *Student participated verbally in the educational dialogue Running Head: DIALOGUE THROUGH THE EYES 19 Kivi and his flag day. As in Example 1, the teacher asked questions to which he/she was looking for a certain specific answer (What day is October 10th?; And who might Aleksis Kivi be?). Students are participating actively but often suggesting answers that are not what the teacher intended to elicit (Finland’s 100-year birthday; Finland’s president!). However, unlike in Example 1, the teacher is not very open to accepting new turns in the discussion and provides only short feedback to the suggested answers (No, it’s not the 100-year birthday yet. I just said so.; No, not Finland’s president.). There are no clarifying or extending questions from the teacher and the discussion is focused on factual knowledge, rather than the sharing of views or experiences. Example 2. Educational dialogue of moderate quality T: So, as I just mentioned, what special thing did you see this morning on your way to school? What difference did you notice in the school yard? [C6] C16: A swing. C6: Mmm, well, there was the Finnish flag on the pole. T: Yes. Why is the Finnish flag up on the pole today? What day is October 10th? [C2]. C2: Mmm, I’ve forgotten. T: There was a little hint about it earlier if you happened to notice. Finland … it’s not Finland’s 100-year birthday yet. [C1]. C1: 100-year birthday. T: Could you say that a bit louder? C1: 100-year birthday. T: No, it’s not yet the 100-year birthday. I just said so. So, [C14]. C14: Aleksis Kivi day. T: Right. And who might Aleksis Kivi be? C16: Finland’s president! T: No, not Finland’s president. C2: Sauli Niinistö! T: Yes, he is Finland’s president. C11: He is expecting a baby! T: But who is Aleksis Kivi? C16, C2: [shouting out their unclear suggestions] T: This is not a shouting match! We are not having a shouting match [C2]. [C16]. Look, I’ll have to put your name on the board if you can’t be quiet. The other students don’t like this either. So, Aleksis Kivi was a famous Finnish author. He was the first man to write a novel in Finnish. And a novel means a very large book, which includes stories. Running Head: DIALOGUE THROUGH THE EYES 20 Teacher and students continue discussing Aleksis Kivi. Figure 2 (see also Appendix 2, gaze map of the same episode of educational dialogue) demonstrates how the teacher distributes his/her focus of attention (fixations) among the students during the moderate-quality educational dialogue reproduced in Example 2. Fifteen out of 18 students (83%) received the teacher’s visual attention (at least one fixation). There were three students (C5, C7 and C17) who did not receive visual attention and one student (C11) who received notably more attention than the others (see Figure 3). The GC measuring total fixation duration per student was 0.53, indicating relatively high variation in the teacher’s visual attention distribution. Six out of 18 students (33.33%) in the class participated in the dialogue vocally, sharing their knowledge. All the vocally participating students received the teacher’s visual attention. There was no statistically significant difference in the teacher’s focus of attention (fixation duration) between vocal and silent students during the episode (U = 21.0, p = 0.159) (see Table 2). Figure 2. Teacher’s distribution of fixation duration per student during educational dialogue of moderate quality (Example 2). 0 10000 20000 30000 40000 50000 60000 1* 2 3 4* 5678910 11* 12 13* 14* 15 16* 17 18 Fixation duration (ms) Students Note. *Students participated verbally in the educational dialogue Running Head: DIALOGUE THROUGH THE EYES 21 4. Discussion The present study was one of the first attempts to investigate teachers’ focus of attention during educational dialogue. The results showed, first, that, in 31 episodes of educational dialogue, teachers allocated their visual attention relatively unevenly among the students and verbally participating students received higher amount of teachers’ attention compared to the silent students. Second, when taking into account the quality of the dialogues, the results showed that, during high-quality educational dialogue, more students got visual attention from the teacher than in educational dialogue of moderate quality. Also, the two example dialogues indicated quality differences in the verbal exchange and in teachers’ focus of attention: more student received attention in high-quality dialogue but there were no significant differences in the teacher’s focus of attention between the vocal and the silent students. The first research question investigated how teachers distribute their visual attention among students during educational dialogue in general. The results showed that, on average, fewer than half of the students (45%) present in the classroom participated in the dialogue vocally, but teachers focused their visual attention on 84% of the students on average (at least one fixation per student). However, the verbally participating students received more teacher’s attention than the students who did not participate verbally. This might indicate that teachers do monitor and pay attention to the majority of the students in the class, even though they are focused on verbal dialogue and particularly on certain students. McIntyre et al. (2019) reported that expert teachers prioritised students in their focus of attention in the classroom during both attentional and communicative parts of their teaching. This finding may indicate that it is not only the verbal questions but also the gaze that teachers use to invite students to participate in classroom discussion. Running Head: DIALOGUE THROUGH THE EYES 22 Studying teachers’ focus of attention during educational dialogue can provide insights into his/her orchestration of the dialogue, not only verbal but visual as well. Cortina et al. (2015) suggested that experienced teachers are able to monitor their whole classroom while providing verbal feedback for certain individual students. The results of the present study are in line with Dessus et al. (2016) in that, although the majority of the students received at least some visual attention, the amount of attention (total fixation duration per student) varied between the students, measured by GC. However, it should be kept in mind that a low GC number does not necessarily indicate effective teacher monitoring and a high number does not reflect low monitoring; rather, some students can participate differently in the classroom activities, requiring a different amount of teacher attention (Cortina et al., 2015). For instance, in educational dialogue it might be expected that the teacher focus more on the students who share their ideas with the class. The second research question compared teachers’ focus of attention between the moderateand high-quality educational dialogues. Overall, the number of fixations and the total fixation duration per student varied substantially in educational dialogues of both moderate and high quality. In high-quality educational dialogues, the teachers were found to pay attention to a higher number of students than in the moderate-quality educational dialogues. Cortina et al. (2015) found that the quality of teacher feedback, meaning teachers’ efforts to engage students in elaborating on their responses, was associated with teachers focusing their attention more on certain individual students, rather than distributing their attention evenly. Moreover, our exploratory study illustrated that the teachers indeed distributed their attention among the majority of the students but that some students received more attention than others. Previous research on teachers’ focus of attention in the classroom suggests that an effective teacher is likely to give attention especially to students who participate actively and display their knowledge (see Cortina et al., 2015). In educational Running Head: DIALOGUE THROUGH THE EYES 23 dialogue, students are expected to demonstrate their knowledge vocally, which would lead us to assume that vocally participating students would receive more attention during educational dialogue. However, this was not the case in the present study, since there was no statistically significant difference in teachers’ focus of attention between vocal and silent students in the educational dialogues of either quality. This may partly be due to the very small sample, indicating that more research needs to be done with a larger sample size, the present study having been one of the first attempts to combine two different approaches methodologically. On the other hand, during the high-quality dialogue, the teacher also invested time in verbal dialogue with the vocally participating students by asking clarifying questions or extending their knowledge. This finding seems reasonable, since investing time in discussion and providing individualised feedback require focus of attention. In educational dialogue, which includes active shared knowledge building between the teacher and the students, the teacher usually provides intentional scaffolding to encourage students to explain and justify their ideas (Alexander, 2006). According to Howe at al. (2019), important characteristics of educational classroom dialogue are that the participants make extended contributions and elaborations and build on previous knowledge and that they also acknowledge and critique different options and opinions. Through eye contact, the teacher can communicate these functions that support vocal discussion, such as asking for feedback or a reply, listening, indicating that the space for discussion remains open or recognition of a social relationship (Argyle & Cook, 1976). The example of high-quality dialogue showed that the teacher acknowledged answers other than those he or she was suggesting and extended them to others for the purpose of sharing views and opinions. Therefore, it may not be realistic to expect the teacher to allocate his or her attention evenly among students, when visual attention is linked with extended knowledge building. In the moderate-quality educational dialogue, the verbal discussion was mainly focused Running Head: DIALOGUE THROUGH THE EYES 24 on brief exchanges between the teacher and the students consisting essentially of facts. Example 2 also shows that the teacher’s gaze behaviour was scattered and that there were three students who did not receive visual attention at all. In conducting an educational dialogue, teachers may feel pressure to include most of the students in the discussion as equally as possible (O’Connor et al., 2017). This pressure could lead to a scenario in which the teacher scans the classroom, inviting all students into the discussion with short questions, but does not have time to invite them to engage in deeper elaboration and exploration. Conducting whole-class educational dialogue can undoubtedly be very demanding and require practice and expertise. McIntyre et al. (2017), for example, showed that expert teachers were more student-centred in their focus of attention than novices. Short questions may be an easier and faster way to involve more students in classroom interaction, whereas extending the discussion may require more experience, in both vocal and attentional aspects of classroom interaction. In the present study, also teachers’ pupil diameter was measured to examine their concentration and cognitive load during educational dialogue. Dilated pupil (48mm) can indicate person’s higher cognitive processing (Rosen, 2010), which is required especially when conducting high-quality educational dialogue. However, the findings of the present study did not indicate difference in the teachers’ pupil size between the moderate and high quality dialogue and did not indicate significant dilation. In the future, research with larger sample size is required to study this phenomenon in more details. 4.1. Implications, limitations and future directions The present study has both practical and theoretical implications. The findings may encourage and help teachers to become more aware of their focus of attention during wholeclass situations that include dialogue with students. Conducting and scaffolding verbal educational dialogue and shared knowledge building in a whole-class situation can be Running Head: DIALOGUE THROUGH THE EYES 31 Mercer, N., & Littleton, K. (2007). Dialogue and the development of children’s thinking: A sociocultural approach. London: Routledge. Michaels, S. & O'Connor, M. C. (2011). Coding guide for teacher talk moves [Coding manual]. Unpublished Instrument. Pittsburgh, PA: Pittsburgh Science of Learning Center. Muhonen, H., Pakarinen, E., Lerkkanen, M-K., Barza, L., & von Suchodoletz, A. (2020). Patterns of dialogic teaching in kindergarten classrooms of Finland and the United Arab Emirates. Learning, Culture and Social Interaction, 25, 100264. doi:10.1016/j.lcsi.2018.11.011 Muhonen, H., Pakarinen, E., Poikkeus, A.-M., Lerkkanen, M.-K., & Rasku-Puttonen, H. (2018). Quality of educational dialogue and association with students’ academic performance. Learning and Instruction, 55, 67–79. doi:10.1016/j.learninstruc.2017.09.007 Muhonen, H., Rasku-Puttonen, H., Pakarinen, E., Poikkeus, A-M., & Lerkkanen, M-K. (2016). Scaffolding through dialogic teaching in early school classrooms. Teaching and Teacher Education. 55, 143–154. doi:10.1016/j.tate.2016.01.007 Myhill, D. (2006). Talk, talk, talk: Teaching and learning in whole class discourse. Research Papers in Education, 21(1), 19–41. doi:10.1080/02671520500445425 O’Connor, C., Michaels, S., Chapin, S., & Harbaugh, A. (2017). The silent and the vocal: Participation and learning in whole-class discussion. Learning and Instruction, 48, 5– 13. doi:10.1016/j.learninstruc.2016.11.003 Olsen, A., Smolentzov, L., & Strandvall, T. (2010). Comparing different eye tracking cues when using the retrospective think aloud method in usability testing. Proceedings of the 24th BCS Interaction Specialist Group Conference (pp. 45–53). British Computer Society. Running Head: DIALOGUE THROUGH THE EYES 32 Pfeiffer, U. J., Vogeley, K., & Schilbach, L. (2013). From gaze cueing to dual eye-tracking: Novel approaches to investigate the neural correlates of gaze in social interaction. Neuroscience & Biobehavioral Reviews, 37, 2516–2528. doi:10.1016/j.neubiorev.2013.07.017 Resnick, L. B., Michaels, S., & O'Connor, C. (2010). How (well structured) talk builds the mind. In D. Preiss & R. Sternberg (Eds.), Innovations in educational psychology (pp. 163–194). New York, NY: Springer. Rogoff, B. (2008). Observing sociocultural activity on three planes: Participatory appropriation, guided participation, and apprenticeship. In P. Murphy, K. Hall & J. Soler (Eds.), Pedagogy and practice: Culture and identities (pp. 58–74). Los Angeles, CA: Sage. Rojas-Drummond, S., Torreblanca, O., Pedraza, H., Vélez, M., & Guzmán, K. (2013). Dialogic scaffolding: Enhancing learning and understanding in collaborative contexts. Learning, Culture and Social Interaction, 2(1), 11–21. doi:10.1016/j.lcsi.2012.12.003 Rosenbaum, D.A., 2010. Human motor control (2nd ed.). Academic Press, San Diego.Segal, A., Snell, J., & Lefstein, A. (2017). Dialogic teaching to the high-stakes standardised test? Research Papers in Education 32(5), 596—610. doi:10.1080/02671522.2016.1225803. van den Bogert, N., Bruggen, J. V., Kostons, D., & Jochems, W., (2014). First steps into understanding teachers' visual perception of classroom events. Teaching and Teacher Education, 37, 208–216. doi:10.1016/j.tate.2013.09.001 Webb, N. M., Franke, M. L., De, T., Chan, A. G., Freund, D., Shein, P., & Melkonia, D. (2009). ‘Explain to your partner’: Teachers’ instructional practices and students’ dialogue in small groups. Cambridge Journal of Education 39(1), 49–70. doi:10.1080/03057640802701986 Running Head: DIALOGUE THROUGH THE EYES 33 Wells, G. (1999). Dialogic inquiry: Towards a sociocultural approach to mediated action. Hemel Hempstead: Harvester-Wheatsheaf. Yamamoto, T., & Imai-Matsumura, K. (2012) Teachers’ gaze and awareness of students’ behavior: using an eye tracker. Innovative Teaching, 2, 1–7. doi:10.2466/01.IT.2.6 Running Head: DIALOGUE THROUGH THE EYES 34 Appendix 1. Teacher’s gaze map of Example 1 Notes. Circle numbers indicate the order of the fixations during the dialogue. The size of the circle indicates the duration of the fixation (bigger circle indicates longer fixation). The lines between the circles map the teacher’s gaze path that follows the order of the fixations (teacher’s focus of attention moved from one student to another). Running Head: DIALOGUE THROUGH THE EYES 35 Appendix 2. Teacher’s gaze map of Example 2 Notes. Circle numbers indicate the order of the fixations during the dialogue. The size of the circle indicates the duration of the fixation (bigger circle indicates longer fixation). The lines between the circles map the teacher’s gaze path that follows the order of the fixations (teacher’s focus of attention moved from one student to another).