Scaffolding children’s participation during teacher–child interaction in second language classrooms

ABSTRACT This study investigates the nature of teacher–child interaction and teacher scaffolding in Finnish second language (L2) classrooms. Although previous research on classroom interaction has highlighted the importance of scaffolding in language learning, much less is known about the importance of scaffolding for younger children's L2 learning. Utilising a sociocultural approach, this study aims to fill this gap. The data were collected from seven pre-primary/early primary L2 classrooms, with four teachers and 205 children through video recording (56 h) and observational field notes. The interaction analysis focused on the communicative functions of teacher–child interaction, teacher and student initiations, and teachers’ scaffolding children's participation during whole-class and small-group sessions. The results show how teacher–child interaction during small-group sessions supported the children’s active participation in L2 interaction and learning. The results also suggest how changing from teacher-centred modes of classroom interaction to child-centred and playful interaction is challenging and requires strong teacher professionalism.


Introduction
Previous research has shown that teachers have a significant role in creating opportunities for productive classroom interaction and in supporting children's active participation in second language (L2) education (Gardner, 2019;Muhonen et al., 2016).Both teachers' scaffolding and children's active participation in classroom interaction are essential for children's L2 learning (Eadie et al., 2019;Hu et al., 2018;Langeloo et al., 2019;Ribot et al., 2018).At the same time, research has shown how classroom interaction that promotes children's active participation and L2 learning requires strong teacher professionalism (Leite et al., 2022;Skarbø Solem & Skovholt, 2019).However, most of the previous studies focused on older age groups of children and youth, whereas younger children's L2 learning has been studied much less often.
In many countries, including Finland, there is an emphasis on L2 education in the early years that is child-centred and playful, and affords children with rich opportunities for active participation in L2 classroom practice (see e.g., FNBE, 2016a;FNBE, 2016b).It has been suggested that using playful pedagogies and materials such as games, create affordances for language learning and foster learner engagement and motivation; for instance, while learners engage with the game, they use vocabulary and language resources suggested by the game (Piirainen-Marsh & Tainio, 2009).Also, children are reported as liking to learn the target language when they can be actively involved (e.g.talking, singing, and playing), feel a sense of accomplishment and satisfaction after task completion, and express their foreign language knowledge (Brumen, 2011).However, little empirical research is available on teacher-child interaction in early childhood L2 education, although there is an increasing prominence in the early start of L2 education across the world (Alstad & Sopanen, 2021;Huhta & Leontijev, 2019;Inha, 2018;Mourão & Lourenço, 2015;Spada & Lightbown, 2009).Also, little is known about how teachers scaffold young children in their L2 learning during classroom interaction and the opportunities and/or challenges involved (Boutin-Charles, 2019;Pinter, 2014).It is hence important for more research attention to be directed at teacher-child interaction in early years L2 education to generate knowledge about teacher-child interaction in L2 classrooms based on children's active participation during playful and child-centred activities.
The present study responds to this research gap by investigating the nature of teacher-child interaction and teachers' scaffolding strategies in Finnish early childhood second language classrooms.In Finland, the national curriculum for pre-primary and basic education (FNBE, 2016a;FNBE, 2016b) encourages teachers to create L2 education that is open to joy and playfulness.Hence, our study tapped into the question of teacher-child interaction framed by efforts to create child-centred and playful learning opportunities for children's L2.Our study focuses on the communicative functions of teacher-child interaction, teacher and student initiations in classroom interaction, and teachers' scaffolding during whole-class and small-group sessions in L2 classrooms.The research questions for the study were as follows: 1) What characterises the communicative functions of teacher-child interaction in early childhood L2 classrooms during whole-class and small-group sessions?2) What kind of scaffolding do teachers use to create opportunities for children's participation in L2 practice during teacher-child interaction?
Therefore, this study will provide further understanding about classroom interaction research and practice in early childhood L2 education.

Conceptualising and researching teacher-child interaction in L2 classrooms
This study was grounded on the sociocultural approach that provides useful conceptual and analytical tools for researching and understanding the social and cultural dynamics of classroom interaction for children's L2 learning.Specifically, this study draws on the sociocultural framework by Vygotsky (1978), which underscores the role of language and social interaction in human development and learning.Zone of proximal development (ZPD) and scaffolding are two important concepts used in the sociocultural approach.ZPD is defined in the literature as the distance between a child's actual development level in independent problem solving and the potential level in problem solving in collaboration with others (Lantolf, 2006;Vygotsky, 1978).Scaffolding is the process that enables children to carry out a task which would be beyond their current level of understanding (Wood et al., 1976).
Scaffolding has been explained in the literature through three characteristic features: contingency, fading and transfer of responsibility, which describe the nature and processes of the support the teacher provides to the child in the educational process.Contingency refers to tailored, responsive and adaptable teacher support depending on the child's performance.Fading refers to withdrawal of the teacher's support gradually over time and transfer of responsibility means transferring the responsibility of performing the task to the child ( Van de Pol et al., 2010).To encourage children to participate in the interaction, teachers use some means, scaffolding strategies, which can be verbal (such as lowering or simplifying speech, using questions, elicitation, clarification, feedback), or procedural (such as using gestures, tools or waiting time strategies) (Masters & Yelland, 2002;Tharp & Gallimore, 1988).
In this study, we investigated the ways in which the teachers scaffolded children's participation in L2 learning practices and the nature of evolving classroom interaction between teachers and children.We were particularly interested to research how interactions in the early years L2 classrooms reflected the educational goals of child-centredness and playfulness.Our research focus is supported by earlier research that demonstrated how children's active participation in classroom interaction is essential for their L2 learning.In their study, Ribot et al. (2018) showed that children's pragmatic and expressive skills developed better when opportunities were created for active language use.In another study, Kovalainen and Kumpulainen (2005) found that encouraging children to make initiations by using various communicative functions supported their participation and learning across several subject areas.
Existing research on teacher-child interaction has shown that in many classrooms a common pattern is the Initiation-Response-Evaluation (IRE) sequence (Mehan, 1979;Sinclair & Coulthard, 1975).This interaction sequence includes a teacher asking a question (I = Initiation), a student answering (R = Response), and the teacher giving feedback to a student's response (E = evaluation).Despite being common, it has been reported that IRE interaction sequences can limit children's opportunities for active participation (Nystrand, 1997).Instead, extended exchanges between teachers and children, and scaffolding children's initiatives through probing, hints or various forms of feedback enhance children's participation in interaction (Nassaji & Wells, 2000;Tognini, 2008).Using extended exchanges like Initiation-Response-Prompt-Response-Prompt, IRPRP (Lemke, 1990;Scott et al., 2006), or using "tasks, recast or referential questions" to encourage children into genuine communication (where teacher-child interaction takes not the form of IRE, but a series of exchanges) increase children's participation in L2 practices and learning (Long, 2018;Touhill, 2012).Similarly, children's own initiations to classroom interaction by asking questions, sharing ideas through explanations, descriptions or evaluations extend traditional IRE exchanges and contribute to children's learning (Kovalainen & Kumpulainen, 2005;Rasku-Puttonen et al., 2012;Waring, 2011).
We also know from existing research that children's benefitting from classroom interaction depends on the teacher's scaffolding.The study by Papandreou and Yiallouros (2018) in early childhood education showed how teachers scaffolded children into doing more complex thinking by using the communicative functions of defining and/or exemplifying in whole-class activities.Research has shown how productive teacher scaffolding is needed to afford children sufficient time for thinking and responding (Cekaite, 2007;Lee, 2017;Tognini, 2008;Walsh, 2011).Research has pointed out that for L2 education in the early years to be effective, there needs to be a balance between teacher-initiated and child-initiated activities with recognition of children's playful participation in their language learning (Brumen, 2011;Siraj-Blatchford & Sylva, 2004).As creating and facilitating classroom interaction in L2 practice that recognises young children as active participants is not easy or straightforward, there is a need for further studies on L2 education in the early years.Our study contributes to this research knowledge by investigating teacher-child interactions and teacher's scaffolding strategies in Finnish early years L2 classrooms.

Participants
This study was undertaken in four pre-primary and three first grade classrooms in the capital area of Finland (see Table 1).At the time of the research, there was a larger pilot study on L2 education in the capital Helsinki area.An invitation to participate in the present study was disseminated among the pilot study early childhood education units and schools.Drawing on convenience sampling as a research strategy (Creswell, 2012), for this study we then selected those early childhood education units and schools who expressed an interest in participating in our research.In sum, this study was participated in by four teachers and 205 children; 85 of these children were from pre-primary classrooms and 120 children were from the first grade of the primary schools.The class size ranged from 19 to 25 students (Mean: 22).The children were six to seven years old.Most of the children were native users of the Finnish language and beginners in English.Fifteen children were using Finnish as their second language.None of the children were native users of English.The participating teachers were qualified: English language teacher, English and Swedish language teacher, English language and class teacher, and early childhood education teacher.Their work experience ranged from two to 11 years.
The study follows the ethical standards of the Finnish Advisory Board on Research (https://www.tenk.fi) and the Data Protection Act.Research Permission was provided by Helsinki City.Informed consent was obtained from all the participants and the children's guardians and the aims of the study were explained to the children to get their permission.All children and teachers participated in the study voluntarily and could withdraw from the study at any time.

Research setting
In the most recent Finnish national core curriculum, L2 education starts from the first grade (age 7) (FNBE, 2016a;FNBE, 2016b).The renewal of the curriculum was preceded by pilot studies in 2018 based on research evidence about the benefits of early L2 education for children's language learning (Huhta & Leontijev, 2019;Inha, 2018), and formally put into effect as of the spring term of 2020 (VOPS, 2019).Some pre-primary schools (six-year-old children) voluntarily included L2 education in their programs.The Finnish national curriculum for pre-primary and basic education (FNBE, 2016a;FNBE, 2016b) presents general and complementary aims and objectives for L2 education.These objectives include promoting children's plurilingual and multicultural identity and developing children's language skills by allowing adequate space for joy, playfulness and creativity.
In Finland, teachers have flexibility in the pedagogic implementation of the curriculum objectives and in their teaching practices (Alstad & Sopanen, 2021).The participant teachers followed several pedagogical approaches in their L2 teaching.These approaches included the communicative approach (Hymes, 1979), which emphasises opportunities for children to use L2; the task-based learning approach (Ellis, 2003), which provides children with opportunities to use the language in an authentic way via a variety of tasks; and the natural approach (Krashen & Terrell, 1983), in which L2 is used throughout the lessons while children are also allowed to use their first language (L1).The teachers actively used both whole-class and small-group sessions in both pre-primary and first grade classrooms.The teachers frequently used English as the language of instruction and used Finnish to make sure the children understood the instructions.The children were allowed to use Finnish freely to provide a safe and comfortable environment to communicate.
In the pre-primary classes, the children had L2 classes for 90 min per week, which included a 30minute whole-class session (with 20-25 children) and a 20-minute small-group session (with sixeight children) for each group (three groups) in turn.The whole-class sessions, occurring typically at the beginning of each lesson, included a classroom routine to revise days, months, date through simple sentences while the small-group sessions included the introduction of a new topic and games, role-plays, songs for practicing speaking and listening.Since children were not competent in reading or writing, practising writing and reading skills includes small tasks such as following the texts while listening and/or watching and writing letters or words.When the English language teacher worked with six-eight children during small-group sessions, the rest of the children in the class were doing pre-primary activities (from their own pre-primary program, not related to L2) with their teachers until their turns came.
In the primary schools, the children attended two 45-minute lessons (90 min in total) per week.The English language teachers in the primary schools also arranged whole-class and small-group sessions; the class teachers of children or an assistant teacher were present in the classes.At least two teachers were present in the class.The teachers did the first 45-minute lesson as a wholeclass to present new language structures with a revision of the previous lessons.They used the following 45 min to practise L2 in small-group sessions.The L2 teaching contexts in all classrooms were similar, but some differences resulted from teachers' practices/preferences.For example, the English language teacher of the pre-primary classes used small groups to teach new structures/vocabulary, whereas the teachers of first graders used whole-class sessions for the same purpose.Some other differences in teaching contexts resulted from the task, method or material choice among the teachers.

Data collection
The data were collected from seven L2 classrooms by means of video recordings, observation and field notes for triangulation by the researcher.Generating video data was necessary for our detailed interaction analysis procedures, allowing us to analyse the data multiple times.The data corpus consists of 56 h of video recordings from the English lessons, 28 h from the pre-primary classrooms and 28 h from first grade at the primary schools, over a 10-week period (October 2018-May 2019).Equal amounts of video data were taken from whole-class and small-group sessions by using one camera per class.The first author took responsibility in videoing all the sessions and taking observational field notes about classroom activities during the L2 session to support and enrich our interaction analysis.

Data analysis
Responding to our research questions required us to employ a combination of interaction analysis methods and procedures that allowed us to focus on the communicative functions of classroom interaction, teacher and child initiations and teachers' scaffolding.
We used a classroom interaction analysis framework, originally developed by Kovalainen and Kumpulainen (2005) to investigate the communicative functions of classroom interaction between teachers and students across classrooms and lessons.This analysis framework provided us with a systematic analysis method for detecting the nature of teacher-child interaction in L2 classrooms during whole-class and small-group sessions.Namely, it highlights the communicative functions and discursive roles of students and teachers, shedding light on their participation opportunities, rights and responsibilities (see e.g., Kovalainen & Kumpulainen, 2005;Kumpulainen & Wray, 2002).Table 2 illustrates the analysis framework and its communicative functions, namely; (1) Evidence negotiation (EVI) focuses on asking for and presenting evidence, justification or reasons, (2) Defining (DEF) includes asking for and providing definitions, elaboration or demonstration, (3) Experiential (EXP) consists of asking for and sharing personal experiences, feelings, (4) View sharing (VIEW) comprises asking for and expressing views, opinions, (5) Information exchange (INFO) relates to asking for and providing information, observations, (6) Orchestration (ORC) illustrates managing interaction and speaking turns, (7) Non-verbal communication (N-VERB) refers to non-verbal participation, (8) Neutral interaction (NEU) displays echoing and re-voicing ongoing interactions, (9) Confirming (CON) shows acceptance of the topic interaction and (10) Evaluation (EVA) focuses on judging, correcting and/or assessing contributions to classroom interaction.
In the analysis, the video recordings were transcribed, categorised as whole-class and smallgroup sessions, read carefully with the observational field notes, and episodes including L2 practice were chosen (in which children use communicative functions when either child or the teacher initiates the interaction).To address our first research question, we analysed the frequencies of teacher-initiated (TI) and child-initiated (CI) interactional acts from the data, and identified the frequencies (f) and percentages (%) of these acts with their communicative functions during wholeclass and small-group sessions.
Our analysis of the teachers' scaffolding was theory-driven, grounded in the earlier research and theories on teacher scaffolding (see e.g., Masters & Yelland, 2002;Tharp & Gallimore, 1988).Based on these frameworks on teacher scaffolding, the examples presented in the results demonstrate some of the most typical forms of teacher scaffolding such as hints, questions, probing, and wait-time in the data.Also, the IRPRP exchanges, in which the teacher extended IRE sequences and scaffolded the children's turns to increase their participation are highlighted in the excerpts presented.
The analysis of the data was first conducted by the first author and then checked by the coauthors.All differences in the analysis were negotiated to reach consensus.There was no disagreement among the authors regarding the analysis and the trustworthiness of the study.

a. The nature of teacher-children interaction
Our analysis reveals that the teacher-child interactions in the seven classrooms were both teacherinitiated (TI) and child-initiated (CI).The teachers initiated classroom interaction in all the classrooms and sessions more often than the children.During whole-class sessions (Figure 1), 62.7% (f: 342) of the interactions were teacher-initiated (TI), whereas 37.3% (f: 204) of the interactions were child-initiated (CI).During small-group sessions, teacher-and child-initiated interaction was more balanced: the teachers initiated 51.6% of the interactions (f: 320) and the children initiated 48.4% of the interactions (f: 300 ) (Figure 2).
Next, we have illustrated two extracts of teacher-child interaction.Table 3 describes typical teacher-initiated interactional sequence from a whole-class session while Table 4 displays a typical teacher-initiated interactional sequence from a small-group session.Whole-class sessions usually started with a routine in which the children practised the previously-learnt language structures as a whole group.The extract in Table 3 from a whole-class session shows a typical Initiation-Response-Evaluation (IRE) sequence between the teacher and children.During the ten-week data collection, it was observed that such IRE sequences made the Reprinted from Kovalainen and Kumpulainen (2005).
children passive; the children started not to focus on L2 practice, but on the completion of the task.
Table 4 displays a teacher-initiated interaction sequence during a small-group session when practising food vocabulary.Here, the children were sharing their likes and dislikes by using pictures and realia (the teacher brought fruit to the class).The teacher recorded the children's interactions  and made them hear their own voices in English with follow-up pair work.The children were eager to say what food they liked and to hear themselves speaking in L2.Only one child, not willing to participate in the follow-up task at the beginning, joined her friends soon after watching them for a while.During small-group sessions, the interaction was more balanced between the teacher and children in terms of initiations.Although the teacher's initiations also resulted in traditional IRE sequences during small-group sessions, they encouraged the children's active participation through communicative functions like asking for the children's experiences (What do you like?) and/or asking for definitions (What is this?).
A closer examination of the nature of the teachers' interaction in the seven classrooms (see two columns on the left of Table 5) reveals that the interaction was often characterised by asking for and providing definitions (DEF) and information (INFO) and evaluating (EVA) the children's responses.These communicative functions reflect typical IRE interaction sequences during which the teacher takes most responsibility for the cognitive work and children have a more passive and responsive role (Kovalainen & Kumpulainen, 2005).Our analysis also shows that the teacher's interaction changed during small-group sessions as here they more often supported the children's active engagement in L2 practice by echoing what the children were saying (NEU), confirming (CON) and by providing feedback (EVA).These communicative functions give evidence of the teachers' efforts to build their instructional interactions on the children's contributions.
Our analysis of the children's participation in classroom interaction (see the right-hand two columns of Table 5) further confirms our findings about the teachers' dominance during whole-class sessions and children's more active involvement in L2 learning practices during small-group sessions.For example, non-verbal acts (N-VERB: 23.5%) were frequently used by the children during whole-class sessions as the children raised their hands and/or used various forms of body language to request a speaking turn.The children were also often repeating the words or sentence structures (NEU 13.7%) in whole-class sessions whereas this function was lower in small-group sessions.In small-group sessions the children had more room and time to share opinions (VIEW 20%) and experiences (EXP 16.6%) and provide definitions (DEF 16.6%).

b. Teachers' scaffolding children's participation in L2 learning through interaction
Our analysis shows that the teachers actively used questions, hints, probing, and wait-time as their scaffolding strategies to support the children's active participation in L2 practice.These scaffolding strategies often extended the traditional IRE sequences and supported the children to use various communicative functions like informing, sharing experiences, defining and evaluating.
One of the common scaffolding strategies used by the teachers to enhance the children's participation in L2 interaction was questions.For example, the teachers used probing questions to encourage the children's active participation and thinking (e.g., "Do you think it's cloudy, what do you think?").They also used procedural strategies like wait-time, gestures, demonstrations, and audio-visual tools while scaffolding the children's participation in L2 interactions.In the episode (Table 6), which exemplifies the teacher's scaffolding strategies, the children looked at pictures on which the first letter of the word representing the picture was uppercase.The teacher-initiated the interaction by asking a question, then scaffolded the children through non-verbal acts (demonstrating, pointing, wait-time), questions (what, what kind of) and later scaffolded children prompts (saying the second letter and a melody of the song about the word) to engage them in L2 and to help them in their collective meaning-making process.As we can see in this IRE sequence, this also resonates well with Lemke's IRPRP framework.At the end, the children found the word dolphin in L2.Table 7 shows another example of teacher-child interaction from a whole-class session.In this episode, the children were practicing 'clothing' through a video song.The teacher stopped the video and presented the name of the clothes through questions and asked the children to name the clothes.Here, the teacher scaffolded the children's participation through questions (Do you think?), demonstration and clues (showing the photo, acting out) and explanation (his hands can be cold).These scaffolding strategies resulted in the children's responses characterised by informing and confirming functions and non-verbal acts.
From a small-group session, Table 8 presents an interaction episode, during which the children were building a house for their favourite toy character and presenting it to their friends.The episode shows that the children used a variety of communicative functions when they were scaffolded by the teacher through questions (What's that?), rising intonation, gestures and explanations (He needs a house) or hints (The teacher telling the first syllable of the word).Here, the children responded to the teachers' initiations via several communicative functions including defining, explaining, re-voicing, sharing experiences and giving evidence in addition to giving information and non-verbal acts.
The observational field notes show that in small-group sessions, when time and space were shared between fewer children, the children's participation was more active.Small-group sessions also presented an easier threshold for all the children to participate in the interaction.However, small-group sessions were not the only reason for the children to initiate or participate in interactional sequences in L2 practices.It was observed that the teachers created more child-centred and playful learning activities during small-group teaching.Playful L2 activities with a variety of materials (e.g., audio-visual & digital materials, real objects, toys & handicrafts) made the children active.Such tasks created an interaction-rich environment, thereby encouraging the children's active participation.For example, recording the children's voices when they were talking about their favourite food (Table 4); watching a cartoon to practise clothes vocabulary with songs (Table 7); building a house for a toy they like and presenting the house to their friends (Table Table 7.The communicative functions of TI and CI interactional patterns in a whole-class session/primary.

Discussion
This study sheds light on the nature of teacher-child interaction, teachers' scaffolding as well as children's participation in early childhood L2 interaction in Finnish pre-primary and primary education.The study shows that the interaction in L2 classrooms was largely based on teacher initiations, and it was characterised by the IRE patterns both in the traditional way (Initiationresponse-evaluation sequences) and in the extended way with the teachers' scaffolding children for further interactional exchanges.Our study shows that while the traditional IRE exchanges were beneficial in attracting the children's attention to the topic of the lesson and to provide L2 input, repeated IRE exchanges diminished the children's active participation (See also Nassaji & Wells, 2000).The results show that a smaller group size made the children more active in their participation in L2 practice and the children's interactions were richer in terms of communicative functions.These findings align with earlier classroom interaction research that have demonstrated how small-group activities increase children's participation (e.g., Papandreou & Yiallouros, 2018;Rasku-Puttonen et al., 2012).Another finding from whole-class and small-group sessions is that the teachers' scaffolding was different in terms of the communicative functions used.While the teachers mostly orchestrated L2 interaction through asking for information, definition, or evidence in whole-class sessions, they allowed more room for the children's participation in small-groups sessions.
Scaffolding strategies like procedural/verbal have been found to be important in the research literature promoting quality interaction between the teacher and children and contributing to children's language learning and skills (Eadie et al., 2019;Langeloo et al., 2019).Our study additionally shows how the teachers' scaffolding helped the children to participate in L2 practice by using various communicative functions.The teachers' non-verbal acts and procedural scaffolding strategies, such as questions, prompts and hints were commonly in the classrooms in our study.Previous research (e.g., Baraldi & Iervese, 2012;Lee, 2017) has shown that non-verbal acts (e.g., gesture, head nod, gaze) can help the organisation of classroom interaction and ongoing meaning making.In addition to these results, we found that the teachers used non-verbal acts to coordinate turntaking (pointing or gazing at a child) in the classroom and to demonstrate words (body language, drawing on air) while the children expressed their willingness to participate in the classroom (raising hand, pointing themselves to the teacher) and/or to indicate their understanding (nodding) or need for clarification (facial expression).
Playful learning opportunities and using a range of tasks and materials contributed to the children's participation in L2 practice.We discovered that the IRE sequences in whole-class sessions including several formulaic language examples which are rich sources for language play (Bell, 2012), gave the children the chance to practise L2 via their responses to the teacher initiatives.The small-group sessions provided the children with opportunities to enhance their participation in the interaction through their initiatives or responses during their playful L2 activities.The teachers used a range of tasks and methods to support the children's participation in L2 interaction.The methods and tasks included pair/group work, which encouraged the children's collaborative language learning while materials and games created a playful learning environment to support the children's engagement in learning of new words and expressions.For example, the children's participation was high during playful learning activities (see Tables 4 and 8) when the teacher recorded the children's voices and when the children built a house for their hero character (see also Brumen, 2011;Siraj-Blatchford & Sylva, 2004).Everyday themes related to the children's daily life and visual materials, such as pictures, toys or realia created a playful learning environment aroused the children's interest and participation in L2 practice.In line with Cekaite (2007) who observed how a silent child turns into a skilful student by participating in interactions during a variety of classroom activities, we found that the children's participation was frequent and varied during the playful activities.When considering the findings from these studies, it can be inferred that playfulness is a significant element in enhancing children's participation in early years of L2 education.

Conclusion
This study demonstrates the benefits of small-group activities for creating children with more opportunities for active participation in L2 interaction and learning.Our study highlights the benefits of small-group sessions and how children can be active in L2 classes when they are supported by teachers with a variety of scaffolding strategies in a playful context.The results also show how changing from traditional teacher-centred modes of classroom interaction to playful and child-centred interaction is challenging for the teachers and requires strong teacher professionalism.As stated by Gardner (2019), teachers have a significant role in making children active in the L2 learning process and this role is often actualised during classroom interaction and scaffolding.Considering that scaffolding is pivotal in L2 education, it is important to raise teacher awareness about scaffolding strategies such as using hints, probing questions, wait-time or modifying the speech in early L2 classrooms.Interactional patterns, teaching practices, tasks, material choice, pair/group work and a playful environment are also significant for increasing children's participation in L2 learning.Therefore, they are worthy of more research attention.
This study has some limitations and implications for further studies.Firstly, in this study the videorecording of classroom interaction was realised with one video camera.In the future, several video cameras could be used to capture children's multimodal interaction during their participation in L2 learning better, particularly during small-group activities.Secondly, in our study we did not focus on investigating similarities or differences across individual teachers.Neither did we investigate variations between L2 education in pre-school and primary schools, including teaching practices or teachers' pedagogical thinking behind their practices.These are important future research directions.Our study also points to the importance of more research on playful learning pedagogy, including the use of games in early years L2 contexts.An important future research area will be to generate knowledge on how classroom interaction around games contributes to children's L2 learning.There is also a need for more research on how teachers can successfully scaffold classroom interaction and children's language learning through games and playful activities.Then, as also indicated by this study but not adequately addressed, there is a need to investigate the differences between teachers and their pedagogical practices in and across L2 classrooms to address equity and quality of L2 education in the early years.

Figure 1 .
Figure 1.The distribution of interaction initiated by the teachers (N = 7) and children in whole-class sessions.

Figure 2 .
Figure 2. The distribution of interaction initiated by the teachers (N = 7) and children in small-group sessions.

Table 3 .
Teacher-child interaction sequences in a whole-class session/pre-primary classroom.All repeat the sentence Good morning, nice to see you.N-VERB, NEU Teacher: Let's see, what day is it today?INFO Some raise their hands, the teacher calls one by name.

Table 4 .
Teacher-child interaction sequences in a small-group session/pre-primary classroom.

Table 5 .
The communicative functions of teacher-initiated (TI) and child-initiated (CI) interaction sequences in whole-class and small-group sessions.

Table 6 .
Teacher-child interaction in a whole-class session/primary classroom.

Table 8 .
The communicative functions of TI and CI initiations in a small-group session/pre-primary.Yes, well done.It is a swimming pool CON, EVA, NEU building the house in groups with wooden blocks, teacher monitors and visits them Children: swimming pool, dining room, garden, he plays with the dog there and kitchen in native language