Journal of Linguistics and Language Teaching
Volume 17 (2026) Issue 1
Learning Possibilities in Spanish as a Foreign Language –
The Value of Peer Interaction Tasks
Fredrika Nyström (Uppsala University, Sweden)
Abstract (English)
The study explores peer interaction tasks in Spanish as a second foreign language in Swedish lower secondary school. Drawing on Sociocultural Theory (SCT), the research examines how language functions as a tool for collaboration and language learning in interactions among teenagers in the foreign language classroom. The data consist of audio and video recordings of peer interaction tasks carried out without teacher involvement, and the recordings were transcribed and analysed through thematic analysis and tools from the field of Conversation Analysis (CA). The findings reveal two main categories in the interaction, namely language functions that continuously complete the task and language functions that support solving the task. Moreover, two subcategories were identified within the functions that support solving the task: clarifying procedural problems and solving language-related problems. The study shows that the choice of either the target language, Spanish, or the majority language, Swedish, is functional when solving the tasks. These results underscore the value of peer interaction in the classroom and demonstrate that language is not only the aim for learning but also functions as a medium for learning. The study contributes to the body of research on language alternation and collaborative learning in second- and foreign-language educational settings.
Keywords: Spanish, classroom interaction, young learners
Resumen (Español)
El estudio explora las tareas de interacción entre pares en español como segunda lengua extranjera en la escuela secundaria inferior sueca. Basándose en la Teoría Sociocultural (TSC), la investigación examina cómo el lenguaje funciona como una herramienta para la colaboración y el aprendizaje de idiomas en las interacciones entre adolescentes en el aula de lengua extranjera. Los datos consisten en grabaciones de audio y video de tareas de interacción entre pares realizadas sin la participación del profesor, y las grabaciones fueron transcritas y analizadas mediante análisis temático y herramientas del campo del Análisis de la Conversación (AC). Los hallazgos revelan dos categorías principales en la interacción: funciones lingüísticas que completan la tarea de forma continua y funciones lingüísticas que apoyan la resolución de la tarea. Además, se identificaron dos subcategorías dentro de las funciones que apoyan la resolución de la tarea: aclarar problemas de procedimiento y resolver problemas relacionados con el lenguaje. El estudio muestra que la elección tanto de la lengua meta, el español, como de la lengua mayoritaria, el sueco, es funcional al resolver las tareas. Estos resultados subrayan el valor de la interacción entre pares en el aula y demuestran que el lenguaje no es solo el objetivo del aprendizaje, sino que también funciona como un medio para el aprendizaje. El estudio contribuye al conjunto de investigaciones sobre la alternancia de idiomas y el aprendizaje colaborativo en entornos educativos de segunda lengua y lengua extranjera.
Palabras clave: Español, interacción en el aula, estudiantes jóvenes
1 Introduction
Developing the ability to speak and communicate lies at the core of foreign language teaching and learning. For many students, oral proficiency is not only an essential aspect of language competence but also the most desired outcome of their studies (Finndahl, 2023). To achieve this goal, learners must be provided with sufficient opportunities to speak the target language. However, research from Scandinavian second foreign language classrooms has indicated that the amount of spoken target language is low, and that the majority language is used frequently in activities such as giving instructions, managing classroom routines or explaining grammatical content (Stoltz, 2011; Vold & Brkan, 2020).
Despite the recognised importance of oral language skills, there is still limited knowledge of how speaking is actually taught and practised in Swedish second foreign language classrooms (Granfeldt et al., 2023), which indicates a need for further research on the matter.
One strategy that appears to support students’ spoken target language is the use of spoken tasks in small groups or pairs rather than conducting whole-class activities. Compared to whole-class activities, where speaking in front of the entire class and the teacher can increase anxiety, peer interaction tends to promote a more relaxed and supportive environment (Nilsson, 2020; Nyström, Säfström, & Söderblom, 2019; Sato & Ballinger, 2016). Speaking in small groups also allows more students to make their voices heard than within whole-class settings. In pairs or small groups, students speak more freely, collaborate to complete tasks and develop a wide range of skills beyond the target language, including cooperation, creativity and socio-cognitive development (Barfod Lund, Jakobsen, Skovgaard Andersen, & Wacher Kjærgaard, 2023; DiCamilla & Antón, 2012; Källermark Haya, 2015; Marian & Kunitz, 2017; Sato & Viveros, 2016; Swain & Lapkin, 1998).
However, pair and small-group structures constitute a dilemma for language teachers, as they are unable to hear each student and provide support or evaluate them properly when everyone is speaking at the same time in a classroom. In addition, Brooks & Donato (1994), as well as Nyström (2025a), have noted that teachers may be concerned that students do not perform tasks correctly or that they speak ‘too much’ of the majority language instead of the target language when speaking outside the teacher’s earshot. The prevalence of the majority language in the foreign language classroom appears to be a challenge for Scandinavian teachers (Llovet Vilà, 2016; T. Nilsson, Harjanne, & Rosell Steuer, 2019). This raises the question of how the target language can be learned in a context where the majority language is the usual means of communication?
Given the tension between the potential benefits of peer interaction and the challenges it presents for teachers’ ability to support students’ language development, it is important to examine what actually occurs when students engage in peer interaction tasks. In particular, there is a need to investigate the specific functions that both the target language and the majority language fulfil in these interactions. Therefore, the present study seeks to answer the following research questions:
What functions does spoken language serve in solving peer interaction tasks in the second foreign language classroom?Which language is chosen for different functions when solving these tasks? |
2 Theoretical Framework
The present study, in which students carry out peer interaction tasks in the second foreign language classroom (1), is based on a socio-cultural framework outlined below.
2.1. Socio-Cultural Theory and Second Language Learning
Vygotskyan Socio-cultural Theory (SCT) is based upon the idea that learning and development occur in a social context and are mediated by the use of tools (Vygotsky, 1978). Among these tools, language is considered the most powerful mediating tool for thinking, learning and communication. Thus, SCT is essential in all education and particularly relevant in research on second- and foreign language learning, given that language and communicative interaction play a central role in human psychological functions (Lantolf, Swain, & Poehner, 2018; Van Compernolle, 2015).
A function of language can thus be understood as the way it is used to create meaning for social or psychological purposes, which is also an appropriate approach in the present study. Following Wertsch (1991) and Lantolf et al. (2018), language is a mediating tool for both social interaction and cognitive processes. Moreover, they consider meaning making as a process in which individuals are involved in purposeful actions using cultural tools, mainly language and symbols, to construct understanding within a specific context. When students engage in peer interaction tasks, their language serves different purposes at different moments.
Other relevant concepts within SCT that are relevant in the present study include the use of language as meaning making in private speech, when speakers talk to themselves rather than to others (Lantolf et al., 2018; Ohta, 2001), and the term internalisation, which defines the process by which learned knowledge becomes a part of the individual's cognitive structure. As for language learning, communicative interaction can be considered both the source and the outcome of internalisation, since speakers simultaneously use the language they have learned and develop new linguistic knowledge while speaking it (Ohta, 2001; Van Compernolle, 2015).
Given that learning is mediated by various means, SCT explains that the most effective learning occurs within the Zone of Proximal Development (ZPD). This hypothetical zone refers to the developmental state that humans can achieve with the help of other means, typically other persons, but not on their own (Lantolf et al., 2018; Xi & Lantolf, 2021). The ZPD has commonly been connected to models of scaffolding, which is a methodological strategy aimed at supporting learners on their way to proficiency. The concept stems from Vygotsky’s theory and is inspired by ZPD, but it was not used by Vygotsky himself. However, as it aligns closely with SCT principles, it has been integrated into later versions of the theory (Xi & Lantolf, 2021). Currently, the notion of scaffolding comprises a variety of methods for assisting learning. It is contextual in its nature and can best be defined as assisted performance adapted to a specific educational situation (Stetsenko, 1999; Van Lier, 2004).
Another activity that aligns well with the principles of ZPD is collaborative dialogue (Swain, 2000; Swain & Lapkin, 1998, 2000). A collaborative dialogue defines spoken interaction in which learners are specifically engaged in problem-solving and the construction of linguistic knowledge. When learners work together to completed communicative or linguistic tasks, they support each other and achieve more collectively than they would individually. Finally, and also aligned with, but not included in, SCT is the widely used notion of Language Related Episodes (LRE) in second and foreign language research. An
LRE is defined as any part of the dialogue where the students talk about the language they are producing, question their language use or correct themselves or others. (Swain & Lapkin, 1998: 326)
This notion is an essential element in the present study, given that students talk a lot about their own language in the peer interaction tasks. They also correct themselves and support each other in language issues.
2.2 Interaction Hypothesis and Interaction Approach
As explained above, the present study applies a socio-cultural perspective to spoken peer interaction in the second foreign language classroom. Traditionally, a more cognitive perspective has been influential in second and foreign language learning, as demonstrated, e.g. by the concept of negotiating for meaning within the Interaction Hypothesis (Long, 1996). This hypothesis argues that spoken interaction in the target language is essential for language learning, provided that interlocutors understand each other and communicative breakdowns are avoided. To prevent such breakdowns, speakers modify their language in various ways that differ from interaction in a first language (Lightbown & Spada, 2015; Long, 1996). The underlying rationale when negotiating for meaning in interaction is that language is learned when a more proficient speaker, typically a native speaker, provides a less proficient interlocutor, typically a non-native speaker, with the correct expression during conversation, thereby resolving communication problems. A socio-cultural perspective, on the other hand, favours an interaction approach (Gass & Mackey, 2014) rather than the interaction hypothesis, which views learning through interaction as a model rather than a theoretical hypothesis. The central interaction-approach idea is that language learning and the construction of new knowledge are mediated not primarily by the individuals’ respective proficiency levels but by the degree and the type of collaboration between learners (Gass & Mackey, 2014; Sato & Ballinger, 2016). Although the learners’ proficiency is of importance for how tasks are completed, the sociocultural oriented interaction approach emphasises not only language acquisition, but also collaboration in its own right and is thus appropriate for peer work in the classroom.
3 Literature Review
The following section describes relevant findings related to spoken interaction in second- and foreign-language classrooms, divided into three subsections: collaboration and language learning, the classroom context and language alternation.
3.1 Collaboration and Language Learning
As mentioned previously, research in second language learning has traditionally applied a cognitive approach, emphasising the individual’s mental learning processes, while a socio-cultural perspective focuses on learning through collaboration and meaning-making. In one study, Foster & Ohta (2005) combined the two perspectives, explaining that ‘traditional’ negotiating for meaning, in which an expert educates a novice, rarely occurs in the foreign language classroom. Instead, according to their findings, collaborative behaviour – and the subsequent learning it fosters – was central and described in terms of interactional moves. Interactions involved relatively few communicative breakdowns, and students actively assisted each other while maintaining a friendly discourse. They initiated self-repair or addressed their peers’ interactional moves, which were defined as: clarification requests, comprehension checks and confirmation checks (Foster & Ohta, 2005). Foster (1998) further explained that not only was evidence lacking from classroom research regarding traditional negotiation for meaning, but students also pretended that they understood their interlocutor or simply avoided speaking altogether to prevent communicative breakdowns.
In interaction, including in the context of school tasks, speakers assume different roles. In his study, Fernández Dobao (2016) demonstrated that students benefitted from small-group work and could learn language even when they assumed a more silent role in the interaction. Moreover, Sato & Viveros (2016) showed that low-proficiency students may benefit more from collaborative work in the foreign language classroom than their high-proficiency peers. Thus, although aptitude clearly affects language learning, a collaborative mindset, understood as learners’ approach to their partner or to the task, may constitute a stronger mediating factor for learning than aptitude alone (Sato & Viveros, 2016).
Similarly, Reichert & Liebscher (2012) found that learning opportunities arose when learners positioned themselves as either language experts or novices during peer interaction. They emphasised that these roles were changeable within a classroom and were thus distinct from simply being proficient, especially when students co-constructed knowledge without any teacher involvement. Similar findings were reported by Storch (2002), who characterised roles along a continuum in dyadic interaction, represented by the axes passive/dominant and novice/expert.
3.2. Spoken Classroom Tasks
Within the field of Conversation Analysis (CA), spoken interaction has been investigated broadly and thoroughly over many years. When it comes to foreign language interaction in classroom tasks, however, patterns of conversation partly differ from those observed in natural conversation (Filipi & Markee, 2018). The following section presents relevant research related to the function of spoken language in peer interaction tasks.
In carrying out the same interactive task in two different classrooms, Storch & Sato (2020) highlighted the importance of the educational context. The same task was performed in one class of English as a Second Language (ESL) and one class of English as a Foreign Language (EFL), and it was found that learning activities and interactional patterns differed significantly, despite the teachers’ instructions being the same. Group dynamics and the use of mediating tools, primarily the majority language, also differed between the two classes. While collaboration scaffolded language learning in both groups, the EFL class relied more heavily on their L1 and produced more Language Related Episodes (LREs).
Similarly, the educational context mattered in a comparison of spoken task types and how they were negotiated by language learners (Mashayekh, 2022). Learners adopted various positions in peer interaction, while their positions were more fixed when interacting with the teacher. Mashayekh (2022) claimed that responses are always constructed by the participants’ backgrounds and motives and therefore dependent on these contextual factors. Consequently, the same task will result in different and exclusive outcomes, given that the individuals are “agents rather than subjects” (Mashayekh, 2022: 1).
When it comes to specific language functions discussed in the present study, Hancock (1997) applied a similar system within frame-theory (originally from Goffman, 1974), labelling these functions as layers of discourse. These layers have different purposes in peer interaction, and language is spoken either in the literal or the non-literal frame. Language in the literal frame involves negotiation between learners and includes meta-language about the task, typically in the majority language. The non-literal frame, on the other hand, contains actions within the task – utterances that could be “ overheard by a referee” (Hancock, 1997: 217) or assessed by a teacher – and is generally performed in the target language. Another way of describing this distinction is that speakers act as their normal selves in the literal frame, whereas they are role-playing in the non-literal frame (Hancock, 1997: 229).
The literal and non-literal frames resemble the findings of Toth & Gil‐Berrio (2022), who distinguished LREs from so-called intersubjectivity negotiation episodes (INEs) in peer interaction tasks. The role of INEs was to collaboratively negotiate the content and form of what should be said within a task, which differs from the function of LREs, whose purpose is to solve language-related problems. Given that the amount of INEs outnumbered the LREs, Toth & Gil‐Berrio (2022) suggest that INEs play a vital role in interaction procedures, the co-construction of knowledge and, accordingly, in language learning.
As displayed above, beginners who engage in spoken communication are often supported by their peers. Other types of support are also common. For example, learners may use written language from the textbook and read aloud to support their spoken target language (e.g., Konzett-Firth, 2020; Nyström, 2025b).
3.3 Language Alternation
In foreign language classrooms where learners share a majority language, the use of that language constitutes a constant possibility or problem. The present study aims to nuance this discussion. Various studies have shown how the two languages interact and are used for different purposes, such as peers providing support or solving problems through the majority language (e.g. Antón & DiCamilla, 1998; DiCamilla & Antón, 2012; Swain & Lapkin, 1998, 2000). Anton & DiCamilla (1998) specifically examined the use of the L1 in L2 interaction and demonstrated its importance for both social and cognitive functions. Examples of these functions include maintaining a counterpart’s engagement in the task, keeping a focus on task objectives, emphasising central elements of the task or discussing possible solutions to problems.
Research in the field of translanguaging (García, Wei, & Palgrave, 2018) suggests that languages are not mentally switched on or off; rather, a person’s entire linguistic repertoire remains active simultaneously and as such is also useful when learning new languages. A recent literature review in this field indicates a growing tendency in curricula and educational practice towards the use and understanding of pedagogical translanguaging (Fuster & Bardel, 2024) as a medium for learning. Lämsä-Schmidt (2024) investigated these practices and found that German multilingual students engaged in translanguaging and drew on their whole linguistic repertoires during peer interaction tasks. In other words, by relying on their multilingual resources – not only the majority language German – they co-constructed knowledge in the target language, English.
In sum, this section has presented research findings relevant to the investigation of language functions in peer interaction. However, only one of these studies (Fuster & Bardel, 2024) has included research from the context of the present study, namely Spanish as a second foreign language in Swedish lower secondary school. The present study contributes new classroom-based knowledge, and more specifically, insights into how language is used to complete peer interaction tasks in this context.
4 Data
4.1 Collection and Selection
To investigate the different functions of language in peer interaction tasks, lessons were audio- and video-recorded in Spanish groups in school year nine (15–16 years old) in five different Swedish schools. The educational context consisted of second foreign language classrooms in compulsory schools, with students whose proficiency in Spanish corresponded to beginner or basic user level, approximately A2.1 in the CEFR (Council of Europe, 2020).The study adhered to the ethical guidelines of the Swedish Research Council (Vetenskapsrådet, 2024). No personal or sensitive data were collected, not even names or genders, as the focus was on class- and group-level interactions rather than individuals. Participants were informed orally and in writing about the study’s purpose, methods, voluntary nature, anonymity, and the right to withdraw at any time without risk. Written consent was obtained from teachers, school principals, students, and both parents or guardians, and non-participating students were ensured placement outside the reach of recording equipment.
In three of the classes, tasks were identified in which the students interacted in the target language, Spanish. From each of these three classrooms, two tasks were selected, resulting in a total of six different tasks. These tasks were selected because at least two pairs or groups of three to four students interacted without the teacher’s involvement. The tasks were audio-recorded with dictaphones placed on the tables and video-recorded with a camera positioned at the back of the classroom.
Upon reviewing the recordings, it became clear that the students began speaking immediately after the teacher's instructions, raising the question of how much speaking time should be included. Then, in the initial stages of the analysis, it was found that, after approximately two and a half minutes, at least one round of each activity (e.g. a game) had been completed. By this point, all students had had the opportunity to speak at least once. Furthermore, similar interactional patterns recurred in subsequent rounds, indicating that substantially different functions were unlikely to emerge in the language. Therefore, it was decided to start transcribing directly after the teacher’s instruction and stop after two minutes and thirty seconds.
In total, the data selection from the six tasks resulted in 20 conversations and approximately 50 minutes of transcribed spoken language, representing the voices of 54 students. Some conversations were excluded due to teacher interference, insufficient length or poor audio quality. All conversations were transcribed unless they were deemed inaudible.
In the present study, the concept task refers to classroom activities in general, and not only to communicative tasks with a focus on meaning, in contrast to exercises with a focus on language forms (Ellis, 2003; Littlewood, 2004). The six selected spoken tasks vary in the extent to which they are meaning- or form-focused, but they were selected according to the criterion that the students spoke the target language with each other in a non-assessed classroom situation and without interference by the teacher. The tasks were labelled TV-genres, Preferences, Holidays, Question Game, Shopping Clothes and Guessing Game by the researcher and typically included questions and answers on everyday topics (Appendix I).
In all six tasks, the instructions included images and/or written target-language input, and the conversations demonstrated that the students’ spoken language was generally based on these words, phrases or images. The aim of the present study is not to measure spoken target-language production quantitatively, nor to compare varying methodological designs and their spoken output, but rather to qualitatively investigate the functions of students’ spoken language when solving peer interaction tasks.
4.2 Analysis of the Conversations
When analysing the 20 conversations distributed among the six tasks, the first step was to transcribe the conversations and identify all spoken language (2). Although it was sometimes difficult to distinguish spoken interaction when everyone spoke at the same time, the video-recordings enabled identification of the speakers within each pair or small group. The amount of spoken output differed substantially between conversations, and consequently, the length of the transcripts varied directly according to the total volume of spoken output in each conversation. This outcome was influenced by several factors, e.g. the number of students who participated and what each student said. The present study does not investigate individual differences, nor does it compare conversations in order to identify better or poorer performances.
The learners used both the target language, Spanish, and the majority language, Swedish, and it was therefore necessary to identify which language was used to express different functions. The target language was highlighted in green, while the majority language – and the few English words students used – were not marked.
When proceeding with the analysis, two methodological tools from the field of Conversational Analysis (CA) were applied, namely the identification of adjacency pairs (AP) and, subsequently, sequences. These were suitable units of analysis because they often occur in the same language in bilingual interaction (Filipi & Markee, 2018; Sato & Ballinger, 2016). An adjacency pair, in its minimal and unexpanded form, is characterised as composed of two related turns performed by two different speakers. A sequence is formed by one or several adjacency pairs that are interrelated in a coherent way (Schegloff, 2007: 13). Typical forms of adjacency pairs include questions-answers, offer / request-acceptance / rejection, invitation / suggestion-acceptance / rejection or greeting-greeting. These were identified in the data and subsequently grouped into sequences when they were functionally coherent in a way that differed from the surrounding APs. In the present study, a function was identified when one or more APs worked together towards a recognisable communicative goal. The analysis was carried out in collaboration with other researchers, one with expertise in interaction. When classifications were uncertain, agreement was reached that the classroom context and the nature of the task should determine the categorisation.
Research in CA and language alternation takes as its point of departure that language choice serves different purposes, and that alternation between languages therefore constitutes a useful resource for language learners (Kunitz, 2018). This means that language choice in bilingual conversation is an activity in its own right, contributing to the structuring of interaction, both within and between sequences (Filipi & Markee, 2018, p. 46). In the present study, this behaviour is particularly relevant, as it aims to identify which language is chosen for which function in peer interaction tasks.
The next step was to explore the language choice in one AP or even throughout various APs in a sequence. Hence, the typical sequence was delimited not only by language alternation but also by features such as pauses, hesitations and rising intonation. The sequences were functionally coherent in the sense that the APs that formed them served the same communicative function.
Example 1 (3) below displays an excerpt from the Preferences task, with two different types of adjacency pairs, which also correspond to different languages. In this excerpt, one adjacency pair is considered to be one sequence, as it serves the same function. In Adjacency Pair 1, SA suggests something in the majority language, which is accepted by SB and relates to how the task should be performed. In Adjacency Pair 2, SB starts performing the actual task by asking a question in the target language, which is then answered by SA. The translation also reflects the linguistic errors:
Speaker | Utterance | Translation | Adjacency Pairs |
SA | Ska vi börja fråga först? | Shall we start asking first? | 1 |
SB | Jaha ... | Ah, yes… | |
SB | Te gustan… dibujos animados? | Do you like… cartoons? | 2 |
SA | Sí, sí me gusta. | Yes, yes, I do |
Example 1: Student Negotiation of Task Procedure and Start of Role-Play.
Example 2, on the other hand, displays a sequence that includes three adjacency pairs. It is excerpted from the beginning of the Guessing game task, in which the students organise the activity by clarifying the instructions. In this sequence, speakers S1, S2 and S3 coordinate the game before it starts, which forms a coherent unit:
Speaker | Utterance | Translation | Adjacency Pairs (AP) | Sequence |
S2 | Jag ska välja en då? | Shall I choose one then? | 1 | 1 |
S1 | Ja… | Yes… | ||
S3 | Ska vi gå så här, laget runt? | Are we going like this, around the team? | 2 | |
S2 | Ja… | Yes… | ||
S2 | Jag har valt mitt | I have chosen mine | 3 | |
S1 | Ok | Ok |
Example 2: Three different APs constituting one sequence (AP 1 and 2: the question-answer; AP 3: suggestion-acceptance)
When investigating the functions of spoken language, there was a need to find commonalities in the sequences and categorise them, and it became evident that the research questions needed to be addressed in a more inductive way. Hence, from this point onwards, instead of applying methodological tools from CA, an inductive approach was chosen, and thematic analysis (Braun & Clarke, 2006) was used as the method of analysis. In this iterative and reflective process, the first step was to determine the communicative role of the language in each sequence. The second step was to identify if, and how, the function in one sequence differed from the next one.
On only two occasions totalling less than one minute, the students spoke about topics that were completely unrelated to the assigned task. These two instances were consequently excluded from the data, as the language used in these cases was completely 'off-topic'. This means that the students’ spoken language was used to complete the task for 49 of the 50 minutes of transcribed conversation.
5 Results
The purpose of the present study is to examine peer interaction tasks and to identify the functions of spoken language when those tasks were completed. Moreover, it aims to investigate the choice of language in the different functions, particularly when alternating between the target language and the majority language. The analysis showed that the students completed the assigned tasks while using both the target language, Spanish, and the majority language, Swedish. In two of the five classrooms, no spoken peer interaction tasks were observed; therefore, only the data from the remaining three classrooms were analysed.
Through the iterative thematic process, two main categories were identified based on the communicative role of the language: the students were either speaking without detectable problems in order to perform the assigned task, or speaking to address and solve problems in the task. When the problems were solved, the students proceeded to speak and perform the task. These categories mostly aligned with the choice of either the majority or the target language. The two overarching categories were then defined as either continuously solving the task or supporting the task completion. The target language was mostly spoken when the task was completed in an uninterrupted, continuous and seamless way, while the majority language was mostly spoken when problems arose and needed to be addressed and solved. The two overarching categories were defined as either continuously solving the task or supporting task completion. Continuous functions involved uninterrupted interaction, whereas supportive functions addressed and solved problems during task performance.
Given that the scope of functions that supported solving the task was broad, there was a need to define them in greater detail. The speakers either discussed how to perform the task or were puzzled about language-related issues. In both cases, there were explicit questions or prompts; however, when related to language difficulties, these could also be either a brief pause, rising intonation to indicate a question, or an “eh…” in the hesitation.
Two sub-categories were identified within the supportive functions, representing metaphorical stumbling blocks along the path of task completion: clarifying procedural problems and solving language problems. In the case of solving language problems, two further sub-categories emerged, relating to either reception or production.
The different categories and sub-categories were identified in all six tasks, albeit not equally distributed between the different conversations.
5.1 Continuous Functions
The continuous function was spoken language, which facilitated continuous and uninterrupted interaction. In these sequences, the task was performed seamlessly, without any detectable problems, unmotivated delays or hesitations. The students appeared to understand both the task and each other and co-constructed meaning while doing so. Learners generally used the target language in this category, as is displayed in the following excerpt from the Shopping clothes task.
In Example 3, Student 1 and Student 2 played the roles of customer and shop assistant. Throughout this sequence, the speakers interacted continuously in the target language while completing the task. There were no apparent interruptions or obstacles during the performance.
Speaker | Utterance | Translation |
S1 | ¿Qué colores? | Which colours? |
S2 | ¿Qué? | What? |
S1 | ¿Qué colores tienes? | Which colours do you have? |
S2 | Ah, tenemos zapatillas de naranja y de verde | Oh, we have sneakers of orange and of green |
S1 | De verde | Of green |
S2 | De verde, no, rojo | Of green… no, red |
S1 | Rojo… | Red.. |
S2 | De rojo | Of red |
S1 | Eh, me gusta los zapatillas.. naranja | Eh, I like the orange sneakers. |
However, on very few occasions in the data, students seamlessly alternated between the target and majority language without changing the topic. These instances were rare and then were marked as a new sequence, but with the same function. The alternation was neither signalled nor marked by the speaker or peers and thus did not cause any interruptions in the interaction. In other words, the speaker alternated without indicating any problems, obstacles or hesitations. The majority language was then used in the sequence until the speaker alternated back to the target language, again without indicating any problems or providing explicit reminders to switch languages.
Examples 4 to 6 below display continuous functions in the majority language in three different tasks. The speakers responded to the content expressed by their interlocutors through affirmations or questions, alternating between languages in the process. The arrows indicate the beginning and end of the sequence, and, accordingly, the use of the majority language. In Example 4, S16 responded to the interlocutor's opinion and continued the interaction in the majority language, not indicating any problems, doubts or obstacles. Similarly, in Example 5, S17 asked a question about hair colour and continued the interaction in the majority language. Again, no problems, doubts or obstacles were indicated. Similarly, in Example 6, S20 thought aloud and continued the interaction in the majority language without indicating any problems, doubts or obstacles.
Speaker | Utterance | Translation |
S15 | Tävlingsprogram, eller Idol | Competition programmes, or Idol [name of a TV-programme] |
S16 | Njae…. No, no me gustan | Well… no, I don't like them |
S15 | Sí, ¡Muy bien! | Yes, very good! |
S16 → | Tycker du? | Do you think so? |
S15 | Nej, alltså, jag tycker det var bra att du sa att du… | No, well, that is, I think it was good that you said that you… |
S16 ← | Ja, sí… | Yes, yes… |
Speaker | Utterance | Translation |
S17 | Sí, ja. soy baja, tiene… pelo | Yes, yes, I am short, has … hair |
S18 | [inaudible] | [inaudible] |
S17 | Igen! | Again! |
Ja | Yes | |
S18 | Tiene? | Has? |
S17 → | [laughter] eller vad har jag för hårfärg? | [laughter] or what colour is my hair? |
S18 | Rubia | Blond |
S19 | Fake rubia | Fake blond |
S18 ← | Jaa… y ojos verde y azul | Yes… and green and blue eye |
Speaker | Utterance | Translation |
S20 | Eh… ¿está… es frutas? | Eh… is it…is it fruits? |
S21 | No. | No. |
S20 → | Ingen frukt… | No fruit… |
Räknas grönsaker in som frukt? | Are vegetables counted as fruits? | |
S21 | Nej, grönsaker och frukt är inte samma sak. | No, vegetables and fruits are not the same thing. |
In other words, when solving the task continuously, no problems or difficulties were indicated by the speakers, nor were there any observable reactions from peers when an interlocutor alternated between languages.
5.2 Supportive Functions
In the second main category, supportive functions, language did not complete the task seamlessly; instead, it was used to support the process of solving the task. These functions fell into two types: those that clarified procedural problems and those that addressed language-related problems. Both types helped when solving the task, and the sequences in this category were generally carried out in the majority language.
5.2.1 Clarifying Procedural Problems
Spoken language used to clarify procedural problems involved talking about the respective task and how it was supposed to be carried out. This meta-language included questions and answers, prompts, or requests for peers to behave in a certain way. The language served as a tool for performing the actual performance task, typically occurring at the beginning of each interaction as a way of organising it. Such sequences also appeared later in the conversations and took various forms. Such sequences also appeared later in the conversations, taking various forms. They could be questions directed at peers, as in the first line of Example 7, or prompts such as 'It is your turn', as in Example 8. They could also be affirmations such as 'It is my turn' or 'OK, we're done now' to finish a task. They were always produced in the majority language. Two excerpts from different tasks are given below in Examples 7 and 8:
Speaker | Utterance | Translation |
S1 → | Ska vi gå så här laget runt? | Shall we, like, go round the team? |
S2 | Ja… | Yes… |
S1 | Jag har valt mitt | I have picked mine |
S1 ← | Ok.. eh…¿De qué está hecho de metal? | Ok.. eh… Of what, is it made of metal? |
S2 | No | No |
Speaker | Utterance | Translation |
S3 | Te gustan fútbol? O no? [laughter] Y... ¿baloncesto? | Do you like football? Or not? [laughter] and basketball? |
S4 | Mmm, a veces | Mmm, sometimes |
S3 → | Det är du | It´s your turn |
S4 | Ja, ok … | Yes, ok… |
5.2.2. Solving Language Problems
The second sub-category was related to linguistic problems and their repair. Just like the procedural functions described above, language was used as a tool to perform and progress through the task. These sequences included solving linguistic problems that had been identified and were always marked or signalled in some way by the peers or by the speakers themselves. For example, problems were indicated by a pause, an ‘eh’, or a rising intonation, when not explicitly expressed in words. Sequences in which language problems were solved also generally occurred in the majority language.
When linguistic problems were addressed, they were generally repaired, most often with the help from peers, either by answering an explicit question or by taking the floor when the speaker paused or expressed hesitation, e.g. through an ‘eh…’, or an interrogating intonation. They could also be repaired through autocorrection. It is worth noting that the description of ‘repairing a problem’ refers to the interactional behaviour and did not always lead to linguistic accuracy, as grammatical structures or vocabulary were sometimes ‘repaired’ incorrectly. Two types of language problems were identified: those related to reception and those related to production. These are described more closely below.
5.2.2.1. Solving Language Problems Related to Reception
When identifying and repairing language problems related to reception, the difficulties concerned either the speaker’s own comprehension or that of their peers. As explained above, these problems were always signalled and thus caused a minor or major interruption in the interaction. Sequences of this type were always carried out in the majority language, as can be seen below. In the Examples 8 and 9, students S6 and S8 explicitly asked about the meaning of words:
Speaker | Utterance | Translation |
S5 | Eh… yo es baja.. y tengo pelo moreno | Eh… I is short… and has dark hair |
S6 → | Vad betyder baja | What does ”baja” [short] mean? |
S5 | Kort, bajo för killar y azules… nej, ojos azules | Short, ”bajo” [short] for boys and blued… no, blue eyes |
S6 | Jaha, så det blir baja, bajo? | Ok, well, so it is “baja”, “bajo” [short]? |
S5 | Ja | Yes |
Example 9: Transcript from the Question Game Task
Speaker | Utterance | Translation |
S7 | Andrea.. ir…a Suecia | Andrea… go… to Sweden |
S8 → | Åker de till Sverige? | Do they go to Sweden? |
S7 | Ja | Yes |
S8 | Ja, ok | Yes, ok. |
S7 → | Y…ir…avión… … | And… go… aeroplane |
Example 10: Transcript from the Holidays Task
5.2.2.2. Solving Language Problems Related to Production
The other type of language problems was related to production, i.e. when speakers could not express what they wished to say, or when peers recognised that the speakers could not express what they wished. Just like all functions that supported solving the task, this was either explicitly expressed in words or marked by a pause, an ‘eh’, or an interrogative intonation, thereby causing a minor or major interruption in the interaction. These sequences were also normally, but not always, conducted in the majority language. In Example 11 below, S10 addressed a problem explicitly.
Speaker | Utterance | Translation |
S9 | ¿Qué buscas? Vad letar du efter? | What are you looking for? What are you looking for? |
S10 → | Un… eh…un camiseta …nej, vänta, så säger man inte… camiseta… | An… eh…a t-shirt… No wait, that´s not how you say it… t-shirt… |
S9 | Du kan säga [inaudible] nej | You can say… [inaudible] No |
S10 → | Un camiseta… verde | A … green t-shirt. |
Language problems related to production could also be addressed and solved completely in the target language, although this occurred less frequently. This is exemplified in Example 11 below, where S12 autocorrected the conjugation of the verb gustar (‘to like’) by adding the consonant ‘N’, while S12 continued the interaction and added a comment to the previous answer:
Speaker | Utterance | Translation |
S11 | Hmm, ¿te gusta … noticia? | Hmm, do you like…news? |
S12 | No. No no. No me gusta | No no no, I don´t like it |
S11 → | O, ¿te gustaN noticias? | Or, ¿do you like the newS |
S12 | Es muy aburrido [laughter] | It is very boring [laughter] |
On some occasions, the speakers translated their own utterances, and it was not always possible to determine whether the purpose was to ensure their peers’ comprehension or to clarify their own intended meaning in production. Nonetheless, the function was clearly related to resolving language problems, as shown in Example 11 (S9) above and Example 13 (S14) below:
Speaker | Utterance | Translation |
S13 | ¿Te gusta… te gusta el tenis? | Do you like, do you like tennis? |
S14 → | Eh… más o menos… halv | Eh… more or less, half |
S13 | Ok | ok |
In summary, these results demonstrate how spoken language is used when students complete peer interaction tasks in a foreign language classroom. Spoken language is used for both continuous and supportive functions. The target language is generally used for continuous functions, while the majority language is used for supportive functions.
The following section discusses these results in the context of previous research and relevant theoretical concepts.
6 Discussion
The present study has shown how spoken language is used for different functions when solving peer interaction tasks, and that either the target language or the majority language may be used for these purposes. It is clear that the learners did complete the tasks they were assigned, even outside the teachers’ earshot, by using both the target and the majority language. The assumption that students may fail to complete tasks or ignore instructions when working in pairs therefore cannot be sustained (Brooks & Donato, 1994; Nyström, 2025a). On the contrary, the students collaborated in a helpful way, and during 49 of the total 50 minutes of transcribed interactions, their spoken language was either used to continuously complete the task or to support solving the task.
In the following section, the results are discussed and interpreted with particular attention to three key patterns that contribute new knowledge about students’ spoken language use in the classroom.
6.1 Spoken Interactive Classroom Tasks for Encouraging Target Language Use
The first pattern, spoken interactive classroom tasks for encouraging target language use, concerns the communicative tendency of students to enter and exit the task, often combined with alternating between target and majority language, as part of their way of acting and role-playing. Given the language-learning classroom context, they appeared either to carry out the actual school task or to speak for a communicative purpose as themselves. In other words, both roles were required for speaking the target language. This type of alternation and role-playing has been identified under different labels in previous research. For example, Hancock (1997) calls them layers, while Toth & Gil‐Berrio (2022) call them episodes. In the present study, the notion of function is applied, emphasising the different language purposes that students draw on when solving classroom tasks.
In the data, the students spoke the target language in a seamless and continuous way, as displayed in Example 3 above, although on a few occasions, the majority language was used in the same way, i.e. without any indication of problems or obstacles. Consequently, the alternation between languages did not elicit any reaction from their counterparts. After one or several utterances in the majority language, someone would always return to the target language again as a way of confirming that it is a classroom task. This shift was also done seamlessly and never explicitly expressed, although it could have been, for example, by reminding peers to speak Spanish. Recent Swedish research suggests that spontaneous, realistic conversations and teachers’ feedback are key to increased speaking in foreign language classrooms (Skolforskningsinstitutet, 2024). Neither of these aspects was investigated in the present study; instead, the results emphasise that tasks explicitly designed to make students speak do, actually, lead them to speak the target language. The students did, in fact, successfully complete the tasks, which proves that they had the necessary competence to do so. In terms of Sociocultural Theory, communicative interaction like this is both the source and the outcome of internalisation processes (Van Compernolle, 2015). This means that the students’ spoken use of the target language, like in example 2 above, constitutes evidence of real language learning.
Another characteristic of school tasks is that speakers are provided with scaffolding resources, e.g. written language in the textbook. As displayed above, the task design involved scaffolding through written target language structures, e.g. words and phrases, which served as tools that enabled participation in the target language. This has also been explained in previous research (e.g., Konzett-Firth, 2020; Storch & Sato, 2020).
6.2 Language Choice as a Functional Resource
The second pattern lies at the core of the investigation, i.e. the choice of language in the identified functions. It was clear that procedural problems were clarified in the majority language and served an organising purpose that enabled the performance of the assigned task. Previous research has identified similar processes and referred to them as task management (DiCamilla & Antón, 2012; Swain & Lapkin, 2000), confirming how they generally, like in the present study, are carried out in the majority language in contexts where learners share this language as their primary means of communication.
According to Sociocultural Theory (Lantolf et al., 2018; Lantolf & Thorne, 2006; Van Compernolle, 2015), language itself is the principal human mediating tool for thinking and learning, and thus constitutes the basis of collaborative dialogue (Swain, 2000). In the present study, whenever language was spoken to support solving the task, it could be categorised as such dialogues. In these moments, the speakers engaged in problem-solving and knowledge-building, and the majority language generally served as the medium that scaffolded and mediated target-language learning (Van Lier, 2004). Although collaborative dialogues can, of course, be conducted entirely in the target language, in this study, they were generally conducted in the majority language, which aligns with substantial previous research (e.g., Antón & DiCamilla, 1998; Donato, 1994; Swain & Lapkin, 1998).
With regard to functions that addressed language problems, those related to perception were predominantly carried out in the majority language, probably to ensure that the peers would understand and be able to provide support. However, problems related to production were sometimes addressed and solved in the target language, as shown in Example 12, where Student 11 identifies his own grammatical error and self-repairs in ‘Or do you like the newS?’ In this case, the speaker refrains from ensuring that peers have understood, probably because the problem has already been solved independently. On several occasions, the speakers explained or auto-translated themselves, as shown in Example 13 above – partly in the target language, and then partly in the majority language. It is not possible to distinguish in the transcript whether it is with the purpose to ensure the peers’ comprehension or a way to think aloud in private speech to mediate thinking (Lantolf et al., 2018; Van Compernolle, 2015). Nevertheless, Example 11 above displays clear private speech when S10 interrupts and attempts to autocorrect himself without explicitly asking for help: “No wait, that’s not how you say it”.
One might assume that the majority language was often used due to deficient linguistic proficiency. Perhaps, as with S16 in Example 4 or S17 in Example 5, the content of questions such as 'Do you think so?' or 'What colour is my hair?' was too difficult to express in the target language, even though this was not indicated in any way, for example through hesitation. One counterargument can, however, be identified in Example 6 above, since it is likely that S20 was proficient enough to say “no fruit” in the target language, given that the student had expressed more complex language earlier in the conversation. An alternative interpretation, then, is that expressing the actual content was more important for the speaker than proceeding strictly in the target language. Here, the private speech (thinking aloud) probably mediates meaning for the speaker himself, and the language constitutes a tool for thinking (Lantolf et al., 2018). These results suggest that language alternation is not necessarily a lack of competence, but can, in fact, be a mediating tool for thinking and for expressing complex thoughts, e.g., the question uttered in Example 6, “Are vegetables counted as fruit?” This question is a way of proceeding with the guessing game and asking the peer about how close the right answer is. All in all, the interactional moves identified by Foster & Ohta 2005 and originally described by Foster (1998), i.e. clarification requests, comprehension checks and confirmation checks, are clearly included in the functions that supported task-solving in the present study.
When the students were continuously solving the respective task, the participants neither negotiated nor solved any problems; rather, they were engaged in meaning-making for the interlocutor about the content, i.e. the actual topic of the conversation (Foster & Ohta, 2005; Lantolf et al., 2018; Wertsch, 1991). Interestingly, the speakers appeared to agree that it made sense for them to be ‘allowed’ to speak both the target and the majority language when doing so. Similarly, problems were often solved in the majority language without any explicit reminders from the peers to switch back to the target language. The speakers appeared to agree on the language choice, instead of feeling obliged to speak the target language. This seamless alternation between languages resembles the concept of translanguaging, where speakers employ all their available linguistic resources to navigate social and cognitive demands (e.g., Fuster & Bardel, 2024; García et al., 2018). Interpreted this way, the present study contributes to the research field of translanguaging. The students relied on the majority language and, on a few occasions, English as resources to complete the task – a finding that aligns with recent research showing how a variety of languages can scaffold peer interaction in the foreign language classroom and support language learning (Lämsä-Schmidt, 2024).
6.3 Collaboration Opens Learning Possibilities
The third pattern highlights how peer interaction tasks in the language classroom open up different kinds of learning possibilities, not only linguistic ones. These include, for example, getting to know a peer and discovering that he or she does not like watching competition programmes, as shown in Example 3, or describing one’s own hair colour, as in Example 5. Recent research has shown that learners appreciate collaboration and peer tasks because such tasks allow them to get to know their peers better (Fredrika Nyström, 2025a).
That being said, this study demonstrates many possibilities for language learning in the classroom. Language problems that were addressed and solved together were most likely related to the speakers’ linguistic proficiency, while speakers who were proficient (or confident) enough to repair their own language probably did so while speaking. Previous research has identified the same functions, most commonly referred to as Language-related episodes (Swain & Lapkin, 1998). In the present data, students were never left without help if they failed to solve a linguistic problem themselves; instead, their peers always stepped forward to assist. There are numerous examples of how language problems were solved in the present study, like in Example 9, when S5 taught a peer the Spanish word for ‘short”, and in Example 10 above, when S9 helped S10 to find the Spanish word for ‘T-shirt”.
In addition, the students negotiated the roles of novice-expert dynamically throughout their conversations, which more closely resembled the SCT-orientated Interaction approach (Gass & Mackey, 2014) than the more cognitively orientated Interaction hypothesis (Long, 1996). Accordingly, communicative interaction is key for mediating both thinking and learning, which then, as in the present study, is not determined solely by each individual’s linguistic proficiency (Gass & Mackey, 2014; Van Compernolle, 2015). The task itself was the starting point, but depending on the interactional patterns in each group, the learning possibilities differed between them. In these positioning processes, speakers treated their peers as collaborative partners who both offered and accepted roles – findings that also align with previous research (Sato & Ballinger, 2016; Storch & Sato, 2020). The roles that students adopted are documented above when S1 and S3, respectively, took on organising roles in Examples 7 and 8, while S5 and S9 appeared to be linguistic experts in Examples 9 and 11.
Within education generally, SCT is recognised as a solid, well-established theory, and the results of the present study emphasise the value of applying it specifically within language methodology. The study also shows that students can benefit when teachers organise peer activities in the second foreign language classroom.
7 Conclusion
The present study has illuminated important aspects of language learning in the classroom, explaining how peer interaction tasks are completed through collaboration, which, in turn, provides opportunities for language learning. It has identified how language is used in two overarching categories, namely those that involve functions that continuously complete the task and those that support completing the task. Among the functions that support task-solving, there were those that clarified procedural problems, while others solved language problems. In addition, the findings highlight the importance of distinguishing between these different functions when carrying out spoken tasks, given that students choose their language guided by the interactional purpose.
Teachers may feel a lack of control and question whether peer interaction is worthwhile, especially given limited lesson time and the possibility of hearing more majority language than they would prefer. To increase the amount of target language use in the classroom, teachers can provide more explicit instruction on useful structures and vocabulary, specifically when language is spoken with the function to support solving the task. Be that as it may, the current findings suggest that students apply themselves to completing the tasks presented to them, which aligns with previous research (Källermark Haya, 2015; Mashayekh, 2022; Nyström, 2025a, 2025b).
Although its results are not generalisable, the present study demonstrates how peer interaction tasks can lead to spoken target-language use, enable students to act as resources for each other and open up various learning possibilities in the foreign language classroom.
Appendix
Topic (6 tasks) | Data (20 recordings) | Type of Interaction | Type of Scaffolding in Books or on the Board |
TV Genres | 3 recordings, voices of 6 students | Practicing the structure of the verb gustar (‘to like’) by asking and answering questions about likes and dislikes about TV-genres | Written on the board:¿Te gusta/n?, Sí, me gusta/n. No, no me gusta/n. (‘Do you like?’, ‘Yes I like’, ‘No I don´t like’) Written in the students’ notebooks: dibujos animados, deporte, noticias, series, telecomedias, telenovelas, concursos, películas, documentale. (‘cartoons, sport, news, series, movies, documentaries’) |
Prefer- ences | 3 recordings, voices of 7 students | Practicing the structure of the verbs gustar, (‘to like’) encantar, (‘to like a lo’) interesar (‘to interest’), doler (‘to hurt’) by asking and answering questions | Written on the board: ¿Te gusta/n…?,Te interesa/n…?, Te encanta/n?, ¿Te duele/n…? El golf, tenis, fútbol, la guitarra, helado, los pies, la cabeza, estos libros, las canciones, los gatos, la geografía. (‘Do you like…? Are you interested in…?, Do you love…? Do/does your… hurt? Golf, tennis, football, guitar, ice cream, feet, head, these books, songs, cats, geography’) |
Holidays | 4 recordings, voices of 13 students | Co-creating a story of an imaginary person’s holiday. The only task without a question- and answer structure. | Written on the board: Habla sobre sus vacaciones. Nombre, actividades, personas, comida, ropa, medios de transporte, etc. (‘Talk about their holidays. Name, activities, persons, food, clothes, means of transport, etc.’) Written on pieces of paper: either Andrea or Alex, and: avión, fútbol, fiesta, Suecia, trabajar, coche, familia, bici, tenis (‘airplane, football, party, Sweden, to work, car, family, bike, tennis’) (4) |
Question game | 3 recordings, voices of 12 students | Playing a game where one student throws the dice, moves the marker, reads the cor- responding question aloud and answers it. Varying sorts of questions – many of which open-ended –, which required answers on a personal level, e.g.: “What did you have for breakfast?” or “What are the most important qualities in a friend?” | None |
Shopping clothes | 5 recordings, voices of 9 students | Performing a role play about buying clothes in a shop. The students take turns in acting as shop assistant and customer. | Written on the board: mochila, zapatillas, libro de horror, bolso (‘backpack, sneakers, book of horror, bag’). In the textbooks: pictures of clothes alongside the written phrases: Quiero probar una talla más pequeña. ¿Tenéis una talla más grande?, ¿Hay esta camiseta en otro color?, Perfecto, me lo/la llevo, ¿Qué talla tienes? Tengo la talla M. (‘I would like to try a smaller size. Do you have a bigger size? Do you have this T-shirt in another colour? Perfect, I´ll take it, What size are you? I am a size M.’) (Solé Prieto, Thalin, & Östervald 2019a: 24–25 for details). |
Guessing game | 2 recordings, voices of 7 students | Playing a guessing game, and figuring out what item the peer is thinking about. | The textbook has pictures of various items alongside the written phrases: ¿Es algo que hay en el colegio?, ¿Es algo que se puede comer? ¿Es redondo?, ¿Es la manzana? (‘Is it something in school? Is it something you can eat? Is it round? Is it an apple?’) (Solé Prieto, Thalin, & Östervald 2019b: 30–31 for details). |
References
Alfredsson, C., & Lutteman, A. (2010). Colores. 9, Lärarhandledning (1. uppl. ed.). Stockholm: Natur & Kultur.
Antón, M., & DiCamilla, F. (1998). Socio-cognitive functions of L1 collaborative interaction in the L2 classroom. Canadian modern language review, 54(3), 314–342.
Barfod Lund, S., Jakobsen, A.-S., Skovgaard Andersen, M., & Wacher Kjærgaard, H. (2023). Elevperspektiver på fremmedsprog fra grundskole til gymnasium – En motivations- og barriereanalyse. Retrieved from Denmark.
Braun, V., & Clarke, V. (2006). Using thematic analysis in psychology. Qualitative research in psychology, 3(2), 77–101. doi:10.1191/1478088706qp063oa
Brooks, F. B., & Donato, R. (1994). Vygotskyan Approaches to Understanding Foreign Language Learner Discourse during Communicative Tasks. Hispania, 77(2), 262–274. doi:10.2307/344508
Council of Europe. (2020). Common European framework of reference for language learning, teaching, assessment- Companion volume with new descriptors. In C. o. Europe (Ed.), (pp. 254).
DiCamilla, F. J., & Antón, M. (2012). Functions of L1 in the collaborative interaction of beginning and advanced second language learners. International journal of applied linguistics, 22(2), 160–188. doi:10.1111/j.1473-4192.2011.00302.x
Donato, R. (1994). Collective scaffolding in a second language. In J. P. Lantolf & G. Appel (Eds.), Vygotskian approaches to second language research (pp. 33–56): Greenwood Publishing Group.
Ellis, R. (2003). Task-based language learning and teaching. Oxford: Oxford University Press.
Fernández Dobao, A. (2016). Peer interaction and learning: A focus on the silent learner. In M. Sato & S. Ballinger (Eds.), Peer interaction and second language learning: Pedagogical potential and research agenda (pp. 33–61). doi:doi 10.1075/lllt.45.02fer
Filipi, A., & Markee, N. (2018). Conversation analysis and language alternation: capturing transitions in the classroom (Vol. 295). Amsterdam; Philadelphia: John Benjamins Publishing Company.
Finndahl, I. (2023). Young students’ language choice in Swedish compulsory school–Expectations, learning and assessment. (Doctoral degree monography). Gothenburg University, Göteborg.
Foster, P. (1998). A Classroom Perspective on the Negotiation of Meaning. Applied linguistics, 19(1), 1–23. doi:10.1093/applin/19.1.1
Foster, P., & Ohta, A. S. (2005). Negotiation for meaning and peer assistance in second language classrooms. Applied linguistics, 26(3), 402–430.
Fuster, C., & Bardel, C. (2024). Translanguaging in Sweden: A critical review from an international perspective. System, 121, 103241.
García, O., Wei, L., & Palgrave, M. (2018). Translanguaging : language, bilingualism and education. Basingstoke; New York: Palgrave Macmillan.
Gass, S. M., & Mackey, A. (2014). Input, interaction, and output in second language acquisition. In Theories in second language acquisition (pp. 194–220): Routledge.
Goffman, E. (1974). Frame analysis: An essay on the organization of experience: Harvard University Press.
Granfeldt, J., Erickson, G., Bardel, C., Sayehli, S., Ågren, M., & Österberg, R. (2023). Speaking French, German and Spanish in Swedish lower secondary school: A study on attained levels of proficiency. Apples-Journal of applied language studies, 17(2), 91–121.
Hancock, M. (1997). Behind classroom code switching: Layering and language choice in L2 learner interaction. TESOL quarterly, 31(2), 217–235.
Konzett-Firth, C. (2020). Co-adaptation processes in plenary teacher-student talk and the development of L2 interactional competence. Classroom discourse, 11(3), 209–228. doi:10.1080/19463014.2019.1597744
Kunitz, S. (2018). L1/L2 alternation practices in students’ task planning. In Conversation Analysis and Language Alternation (pp. 107–128): John Benjamins Publishing Company.
Källermark Haya, L. (2015). Agency in the Spanish language classroom [Elektronisk resurs] Student and teacher choices, actions and reports when students search for information online as part of a theme. Department of Language Education, Stockholm University, Stockholm.
Lantolf, J. P., Swain, M., & Poehner, M. E. (2018). The Routledge Handbook of Sociocultural Theory and Second Language Development: Taylor and Francis.
Lantolf, J. P., & Thorne, S. L. (2006). Sociocultural theory and the genesis of second language development. Oxford: Oxford UP.
Lightbown, P. M., & Spada, N. (2015). How languages are learned (4. ed.). Oxford: Oxford University Press.
Littlewood, W. (2004). The Task-Based Approach: Some Questions and Suggestions. ELT Journal, 58(4), 319–326. doi:10.1093/elt/58.4.319
Llovet Vilà, X. (2016). Language teacher cognition and practice about a practical approach : the teaching of speaking in the Spanish as a foreign language classroom in Norwegian lower secondary school. University of Bergen, Bergen.
Long, M. H. (1996). The Role of the Linguistic Environment in Second Language Acquisition. In W. Rithcie & T. Bahtia (Eds.), Handbook of Second Language Acquisition (pp. 413–468). San Diego Academic Press.
Lämsä-Schmidt, H. (2024). Translanguaging-to-Learn During Collaborative Dialogue in Task-Based Interaction: The Potential of Utilizing Arabic as a Heritage Language for Additional Language Learning. Critical multilingualism studies, 11(1), 351–393.
Marian, K. S., & Kunitz, S. (2017). "Well if we're wrong it's your fault": Negotiating participation in the EFL classroom. Travaux neuchâtelois de linguistique(67), 49–77.
Mashayekh, S. (2022). Agentic learners and task negotiation: How is task response constructed. Studies in educational evaluation, 73, 101149.
Nilsson, M. (2020). Young learners' perspectives on English classroom interaction : Foreign language anxiety and sense of agency in Swedish primary school. Stockholms Universitet, Stockholm.
Nilsson, T., Harjanne, P., & Rosell Steuer, P. (2019). Between theory and practice. Apples Apples - Journal of applied language studies, 13(1), 113–131.
Nyström, F. (2025a). Didactic dilemmas and possibilities: Teachers’ and students’ perceptions of speaking Spanish in foreign language education in Swedish lower secondary school. Educare (4), 163–198.
Nyström, F. (2025b). Opportunities for speaking: the if, how and when of students’ spoken target language in the second foreign language classroom. Moderna språk, 119(2), 194–215.
Nyström, F., Säfström, S., & Söderblom, E. (2019). Hur får jag dem att prata? En klassrumsstudie i spanska. Skolportens numrerade artikelserie för undervisning, lärande och ledarskap, 9, 28.
Ohta, A. S. (2001). Second language acquisition processes in the classroom : learning Japanese. Mahwah, N.J: Lawrence Erlbaum Associates.
Reichert, T., & Liebscher, G. (2012). Positioning the expert: Word searches, expertise, and learning opportunities in peer interaction. The modern language journal, 96(4), 599–609.
Sato, M., & Ballinger, S. (2016). Peer interaction and second language learning: Pedagogical potential and research agenda. In L. l. L. teaching (Ed.), (pp. 409).
Sato, M., & Viveros, P. (2016). Interaction or collaboration. Group dynamics in the foreign language classroom. In Peer Interaction and Second Language Learning (pp. 91–112). Amsterdam: John Benjamins Publishing Company.
Schegloff, E. A. (2007). Sequence organization in interaction. Cambridge: Cambridge University Press.
Skolforskningsinstitutet. (2024). Muntlig kommunikation i moderna språk - lärares undervisning. Retrieved from Solna: https://www.skolfi.se/forskningssammanstallningar/systematiska-forskningssammanstallningar/muntlig-kommunikation-moderna-sprak/
Solé Prieto, M., Thalin, U., & Östervald, M. (2019a). Qué bien 9 Textos (Första upplagan ed.). Stockholm: Sanoma Utbildning.
Solé Prieto, M., Thalin, U., & Östervald, M. (2019b). ¡Qué bien! 9 Ejercicios. Stockholm: Sanoma Utbildning.
Stetsenko, A. P. (1999). Social interaction, cultural tools and the zone of proximal development: In search of a synthesis. In S. Chaiklin, M. Hedegaard, & U. Juul Jensen (Eds.), Activity Theory and Social Practice (pp. 235–252). Århus: Aarhus University Press.
Stoltz, J. (2011). L'alternance codique dans l'enseignement du FLE: Étude quantitative et qualitative de la production orale d'interlocuteurs suédophones en classe de lycée.
Storch, N. (2002). Patterns of interaction in ESL pair work. Language learning, 52(1), 119–158.
Storch, N., & Sato, M. (2020). Comparing the same task in ESL vs. EFL learning contexts: An activity theory perspective. International journal of applied linguistics, 30(1), 50–69.
Swain, M. (2000). The output hypothesis and beyond: Mediating acquisition through collaborative dialogue. In Sociocultural theory and second language learning (Vol. 97, pp. 97–114).
Swain, M., & Lapkin, S. (1998). Interaction and second language learning: Two adolescent French immersion students working together. The modern language journal, 82(3), 320–337.
Swain, M., & Lapkin, S. (2000). Task-based second language learning: the uses of the first language. Language teaching research, 4(3), 251–274. doi:10.1177/136216880000400304
Toth, P. D., & Gil‐Berrio, Y. (2022). Exploring “intersubjectivity negotiation episodes” and “language related episodes” in second‐language peer interaction. Language learning, 72(S1), 275–313.
Van Compernolle, R. A. (2015). Interaction and second language development: A Vygotskian perspective (Vol. 44): John Benjamins Publishing Company.
Van Lier, L. (2004). The Ecology and Semiotics of Language Learning A Sociocultural Perspective (1st ed. 2004. ed.). Dordrecht: Springer Netherlands.
Vetenskapsrådet. (2024). God forskningssed 2024. Vetenskapsrådet.
Vold, E. T., & Brkan, A. (2020). Classroom discourse in lower secondary French-as-a-foreign-language classes in Norway: Amounts and contexts of first and target language use. System (Linköping), 93, 102309. doi:10.1016/j.system.2020.102309
Vygotsky, L. S. (1978). Mind in Society Development of Higher Psychological Processes: Harvard University Press.
Wertsch, J. V. (1991). Voices of the mind: Sociocultural approach to mediated action: Harvard University Press.
Xi, J., & Lantolf, J. P. (2021). Scaffolding and the zone of proximal development: A problematic relationship. Journal for the theory of social behaviour, 51(1), 25–48.
Author:
Fredrika Nyström
PhD Student
Department of Education
Uppsala University
Sweden
ORCID-ID: 0009-0007-5266-8626
Email: Fredrika.nystrom@edu.uu.se
___________________
(1) In the Swedish school system, English is referred to as the first foreign language and is compulsory starting from year three (9–10 years old).
(2) Pauses were indicated with an ellipsis, and laughter and unidentifiable words were written in square brackets [ ].
(3) The respondents’ linguistic errors were retained in the utterances.
(4) The task was inspired by those presented by Alfredsson & Lutteman (2010).