The effect of task-based peer interaction and pre-task instruction on young EFL learners’ explicit and implicit knowledge of past tense: An intervention study

Abstract

The present study explores whether a focused task-based peer interaction pedagogical intervention leads to increased explicit and implicit knowledge of past simple tense among young learners of English as a foreign language (EFL) and whether adding pre-task explicit grammar instruction or explicit interactional instruction or both has any impact on the results. Four groups of 6th grade EFL learners aged 11–12 years participated in an 8-week pedagogical intervention (50 minutes per week) while a fifth control group only participated in the testing sessions (n = 36). Groups were divided according to the intervention received: grammar+interaction strategy (n = 26), grammar (n = 26), interaction strategy (n = 24) and task-only (n = 24). The four groups who underwent treatment also participated in past tense focused task-based peer interaction in the second part of the weekly intervention session. All groups completed a bi-modal untimed grammaticality judgement task (GJT) (explicit knowledge) and a self-paced reading task (SPR) (implicit knowledge) before and after the intervention. Results show that the use of pre-task instruction of both linguistic and interactional type together with focused peer-interaction tasks contributes to the children’s explicit learning of past tense. Children’s implicit knowledge of past tense did not show any significant development in any of the intervention types. Findings are discussed in relation to the pedagogical intervention carried out and the development of implicit and explicit knowledge.

Keywords

EFL explicit knowledge implicit knowledge peer interaction pre-task instruction young learners

I Introduction

The effects of interaction on second language (L2) development are central to the study of instructed second language acquisition (ISLA) (Mackey, 2006, 2012; Loewen, 2015; Long, 2017). The original interaction hypothesis (Long, 1996), now the interaction approach (Gass & Mackey, 2020), includes a number of processes related to L2 learning, namely exposure to input, production of output, noticing, negotiation of meaning, feedback provision and modified output, which might be favoured to a higher or lesser extent in the context of the L2 classroom. The effects of such processes on L2 learning are mediated by a number of variables which are the focus of recent and present research. These include the characteristics of learners, the characteristics of the tasks employed to elicit interaction, the type of linguistic structures examined and the learning context where such interaction processes occur (Loewen & Sato, 2018).

The type of interaction produced and how it is conducive to learning is affected by whether the learners interact with a teacher or with peers (Philp et al., 2014) or whether the learners are adults, teenagers or young learners (Oliver, 1998; Oliver et al., 2017). Regarding task characteristics, whether a task is focused or unfocused (i.e. whether it prompts a specific linguistic structure or not) (Ellis, 2003; Ellis et al., 2006; Lyster, 2004), how learners prepare for a given task and whether they are given pre-task instruction (Ellis, 2018; Ellis et al., 2020) or whether they pay attention to language during task performance or receive corrective feedback, (Lyster et al., 2013) all have an impact on the effects of interaction on L2 learning. The type of linguistic feature that is taken to represent L2 development as a result of interaction is also relevant (Mackey & Goo, 2007). Feature saliency (Gass et al., 2017) or learner readiness to learn a given linguistic feature (Pienemann, 2005) are also factors that might determine development. The learning context also plays a fundamental role. Even within instructed contexts, whether learners interact in second language or immersion contexts or they do so in a foreign language context will affect interaction and its effect on L2 development to a great extent (García Mayo, 2021).

Although the effects of interaction have been widely researched, there are still a number of aspects within the mediating variables in need of further investigation. Child learners, particularly those in foreign language classroom contexts and interacting with their peers, still represent an under researched learning context (García Mayo, 2021). More specifically, classroom intervention studies exploring the effect of task-based peer interaction and pre-task instruction on such child learners’ L2 knowledge are even scarcer. Bearing in mind these research gaps, this study aims to analyse whether including focused task-based peer interaction in weekly English as a foreign language (EFL) primary school sessions has any effects on the learners’ explicit and implicit knowledge of simple past tense. We will also determine whether adding pre-task explicit instruction of past tense with regular and irregular verbs (grammar group), or pre-task explicit instruction of interactional and corrective feedback strategies (interaction strategy group) or both (grammar+interaction strategy group) during the weekly teaching intervention has any further effects on the learners’ development of past tense. The following sections will review how young EFL learners take part in and benefit from task-based peer interaction and how the use of tasks, interaction and pre-task instruction might foster grammatical development.

II Young EFL learners and peer-interaction

While young learners (YLs) are an ever-increasing group of L2 learners, the outcomes of language teaching policies in FL settings are still under-researched in comparison to older age groups (i.e. secondary school or university students) or same-age groups in second language settings (Collins & Muñoz, 2016). Findings from older learners or second language learning settings cannot be extrapolated to YLs on account of age-related, cognitive, affective and literacy differences and on account of the differences in the amount and quality of exposure that learners receive in second language contexts. Primary school learners (i.e. 6 to 12 year-olds) find themselves in the middle childhood developmental period, during which children undergo crucial changes in their physical, cognitive and socio-affective development. Children increase their reasoning skills as well as their working memory capacity and self-regulation strategies (DelGiudice, 2018). Such changes inevitably influence their L2 learning experience, which still mainly benefits from great amounts of rich and varied input and opportunities to use the target language in a meaningful way, but gradually involves more explicit learning mechanisms and greater metalinguistic awareness (Holmes & Myles, 2019). In FL contexts, access to plentiful and quality input and opportunities for language use tend to be restricted, so fostering children’s communicative interaction and creating further opportunities to increase language use in the classroom is crucial to ensure progress (García Mayo, 2021).

Research on child EFL peer interaction during task completion has increased over the past few years and has mainly been carried out in task-supported language teaching contexts, which include the use of meaningful tasks but are not exclusively based on task-based language teaching (TBLT) lessons (García Mayo, 2021). Young EFL learners have been shown to be able to negotiate for meaning in the L2, producing instances of conversational adjustments, repetitions, and collaboration and feedback strategies with age, instructional setting, proficiency pairing and task repetition as mediating factors (Azkarai & Imaz Agirre, 2016; García Mayo & Imaz Agirre, 2017; García Mayo & Lázaro Ibarrola, 2015; Lázaro Ibarrola & Azpilicueta Martínez, 2015; Lázaro Ibarrola & Hidalgo, 2017; Pinter, 2006, 2007; Pladevall-Ballester & Vraciu, 2020). Children’s collaborative patterns and their attention to form while carrying out peer interaction tasks have also been analysed (Azkarai & Kopinska, 2020; Calzada & García Mayo, 2020; García Mayo & Imaz Agirre, 2016, 2019; Pladevall-Ballester, 2021) and the evidence suggests that they mainly engage in collaborative patterns and they can indeed focus on and negotiate language form, both of which might be affected by task modality.

Although research on child EFL peer interaction is increasing, the majority of studies are not classroom-based and/or do not include any type of classroom intervention or measures of L2 knowledge and development. Knowing to what extent peer interaction might affect L2 grammatical development among young learners and whether pre-task instruction might mediate such effects would greatly contribute to our understanding of the benefits of peer interaction.

III Task-based peer interaction, pre-task instruction and language development

While the use of tasks in language learning fosters communication among learners, research has cast doubt on whether the use of unfocused tasks alone (i.e. the strong version of TBLT) without the provision of focus on form (FonF) actually promotes the learning of specific language structures (Ellis, 2016; R. Sato, 2010; J. Willis, 1996). The use of focused tasks in different conditions allows researchers to test whether and to what extent learning of the target structure occurs (see, amongst others, Doughty & Varela, 1998; Ellis et al., 2006; Lyster, 2004). Yet learners might not be developmentally ready to learn the target feature through performing tasks (Pienemann, 2005) and designing a task that makes it essential for learners to produce a specific grammatical form is complex (Loschky & Bley-Vroman, 1993). In order to enhance the learning potential of the use of focused tasks, certain task conditions can be manipulated at different moments of the task cycle (Ellis et al., 2020). One of these moments, the pre-task stage, motivates and prepares learners for task performance and has been particularly studied in relation to task performance itself as well as motivation and learning improvement (see, amongst others, Ellis, 2009; Li et al., 2016; Ortega, 1999; D. Willis & Willis, 2007). Such pre-task planning might come in the form of form-focused instruction in an explicit or implicit manner. Pre-task explicit instruction has been sometimes rejected by pedagogical research within the TBLT field (Long, 2015) arguing that any FonF should be reactive and occur while learners perform the task (in the form of corrective feedback). Other researchers see explicit teaching of target forms as a necessary step prior to their use in task performance or as essential for their subsequent development. From a theoretical perspective, pre-task instruction is supported by DeKeyser’s (2007, 2015, 2017) skill acquisition theory, by which explicit knowledge (which entails explicit instruction) of a given L2 feature precedes practice activities where such knowledge is proceduralized and eventually automatized if there are extensive opportunities to use the target form in communicative contexts. Pre-task instruction would then establish learners’ declarative knowledge, which would be put into practice during task performance.

Previous research on the effects of pre-task explicit instruction on the explicit and implicit knowledge of specific grammatical structures is not abundant and similar studies based on in-class interventions with young EFL learners are, to the best of our knowledge, inexistent. Li et al. (2016) explored the effects of explicit instruction and oral tasks on the explicit and implicit knowledge of the past passive by Chinese middle school EFL learners. The four experimental groups participated in a two-hour treatment where they performed two dictogloss tasks in groups and an oral reporting session. One group only performed the oral tasks, a second group received explicit instruction and carried out the tasks, a third group received within-task feedback and a fourth group received both within-task feedback and pre-task instruction. A control group only took part in the pre, post and delayed posttests, namely a grammaticality judgement task (GJT) and an elicited imitation test (EIT). The comparison between the first and second groups, which is relevant to our study, resulted in the group that received explicit instruction prior to the performance of the tasks obtaining greater gains in the GJT (i.e. explicit knowledge) than the group who just performed the tasks. Implicit knowledge was not affected by any type of intervention, but when results were analysed separately according to whether participants had previous knowledge of the structure, those who had prior knowledge and had received within-task feedback (i.e. third group) showed some improvement in the implicit task. In a very similar study, Li et al. (2018) compared three groups of Chinese EFL learners who were split into control, task-only and explicit instruction+tasks conditions on their explicit and implicit knowledge of the past passive. Using the same tests, the researchers found that the condition that obtained higher results was the one with explicit instruction on the measure of explicit knowledge, while no effects were observed in implicit knowledge.

Drawing on Li et al.’s (2016, 2018) studies, Shintani (2019) carried out a similar study with low to intermediate university Japanese EFL students but included a number of instructional sessions (i.e. 7) and a condition in which the learners did not perform tasks and were only explicitly instructed. She also included a production test to measure accurate use of the target structure (i.e. past counterfactual conditional) as well as an Error Correction (EC) test to measure explicit knowledge. Relevant to our study are the results on the explicit knowledge test. The group that received explicit instruction and performed the tasks improved the most, followed by the group that only received explicit instruction and by the task-only group. The control group was only significantly lower than the group with explicit instruction and task performance. The author suggests that explicit instruction is helpful for low proficiency learners, who necessarily should be then given opportunities to produce the target features in a freer way.

More recently, Khezrlou (2021) investigated the immediate and delayed effect of task repetition (picture narrative tasks) with and without explicit instruction on the explicit and implicit knowledge of past tense among young adults at a language institute in Iran by means of an untimed GJT and an EIT. Two groups participated in the study, the first one being exposed to task repetition only and the second one receiving task repetition and explicit instruction between the first and the second task repetition on consecutive days. Both groups were pre-tested, post-tested after each task repetition and post-tested two weeks after the last task repetition. Findings indicate that the group that received explicit instruction together with task repetition improved significantly at the immediate and delayed explicit post-tests whereas the group that was only exposed to task repetition improved only after the first immediate test but obtained lower results afterwards. As for implicit knowledge, the group that only received task repetition did not significantly improve at any testing times, whereas the group receiving explicit instruction did significantly improve after the second task repetition.

Even scarcer is the research on the effects of pre-task interaction strategy instruction on grammatical development and, to the best of our knowledge, it is inexistent with regard to explicit and implicit L2 knowledge. M. Sato and Loewen (2018) explored the impact of metacognitive instruction on clarifications requests and recasts together with activities including the two types of implicit corrective feedback on university EFL learners’ use of third person singular -s and possessive determiners his/her in picture description tasks. Results showed that metacognitive instruction with corrective feedback activities led to higher use of the target structures than either type of corrective feedback activities alone. Research on interactional and metacognitive instruction has mainly focused on its effects on the actual use of corrective feedback (Fujii et al., 2016; M. Sato & Lyster, 2012) and interactional strategies (M. Sato, 2020) in subsequent tasks, but also on learners’ accuracy and fluency and their beliefs about corrective feedback (see, amongst others, M. Sato & Lyster, 2012). With young learners, M. Sato and Dussuel Lam (2021) found that interactional instruction led to a higher use of the L2 and a greater metacognitive knowledge of how oral communication works, although the children’s willingness to communicate was not seen to improve.

Further research is needed on the effect of focused tasks, peer-interaction and pre-task grammar and interactional explicit instruction on language development and specifically among young learners. Following a pre-post test design with an 8-week teaching intervention including a control group and four experimental groups, the following two research questions were posed:

Research question 1: Does focused task-based peer interaction lead to increased explicit and implicit knowledge of simple past tense among young EFL learners?

Research question 2: If gains are observed, are there differences based on type of intervention?

These two research questions will allow us to explore if adding focused task-based peer interaction in a weekly EFL session during 8 weeks in primary education has any effects on the learners’ explicit and implicit knowledge of regular and irregular simple past tense. We will also determine whether adding pre-task explicit instruction of regular and irregular past tense (grammar group), or explicit instruction of interactional and corrective feedback strategies (interaction strategy group) or both (grammar+interaction strategy group) during the weekly teaching intervention has any further effects on the learners’ development of past tense. Past tense was chosen as it is one of the key grammar components of the 6th grade EFL curriculum and it is formally taught in this grade for the first time. We expect gains in all intervention groups (Khezrlou, 2021; Li et al., 2016, 2018; Shintani, 2019; Williams & Evans, 1998), particularly in the explicit task and potentially in the implicit knowledge task. Inconclusive group effects were found in previous research with other age populations, but we do hypothesize that explicit grammar instruction will have an effect on the young learners’ knowledge of simple past tense.

IV Methods

1 Participants

The participants were 136 young (N = 136) EFL 6th grade learners of English (69 females and 67 males), aged 11–12 years with an A2 level of English proficiency (M = 49.8; SD = 17.5) as measured in the Oxford Placement Test for Young Learners,¹ studying in three Catalan state schools sharing sociodemographic characteristics in the Metropolitan area of Barcelona, Spain. The learners from the different schools were comparable in terms of number of hours of previous and current in-classroom exposure to English and in terms of level of English.² The learners were Catalan/Spanish bilinguals and the accumulated in-class exposure to English prior to the study was of 525 hours approximately over the span of nine years (since the beginning of pre-school). As for the role of English in their daily lives, 40.77% of the learners had taken extracurricular English lessons prior to the study, versus 59.23% who had never taken them (with fairly similar proportions across groups). Along similar lines, 44.62% were occasionally involved in leisure activities in English (e.g. videogaming, watching TV), whereas 55.38% were not. One of the three schools acted as a Control group (C) (N = 36), who only took the pre and post tests. The children in this group were not taught the past tense and did not participate in any kind of peer-interaction tasks during the study. The children of the other two schools were divided into four intervention groups, which were named according to the different types of intervention they received in the first part of each weekly session: the grammar+interaction strategy group (Gr+I) (N = 26), the grammar group (Gr) (N = 26), the interaction strategy group (I) (N = 24) and the task-only group (T) (N = 24), meaning that the Gr+I group received grammar and interactional strategy pre-task instruction, the Gr group received grammar pre-task instruction, the I group received interactional strategy pre-task instruction and the T group did not receive any pre-task instruction (for a detailed explanation, see Section IV.2). In the second part of each weekly session, all four intervention groups participated in peer interaction tasks (see Section IV.2). The teachers created the groups and ensured that they were balanced in terms of cognitive and linguistic abilities. All groups were pre and post tested with an untimed grammaticality judgement task (GJT) and a self-paced reading task (SPR) on the children’s knowledge of the simple past tense and the four intervention groups participated in the teaching intervention explained below.³

2 Intervention

The teaching intervention lasted 8 weeks with a 50-minute session per intervention group each week. These weekly sessions were divided into two parts of 25 minutes each and were carried out by the same researchers throughout the intervention period. The first 25 minutes were different in each group while the remaining 25 minutes were exactly the same for all groups during which they carried out oral peer interaction tasks (see below).

As for the first 25 minutes, the task-only group (i.e. T) did not receive any type of pre-task instruction and participated in EFL games or vocabulary activities which were unrelated to the target structure but ensured the children received the same amount of exposure to English as the other groups. These games included tic-tac-toe, charades, picture dictation, the hangman and odd-one-out, amongst others. The other three experimental groups (Gr, Gr+I and I) did receive pre-task instruction during this part of their weekly intervention sessions. Such pre-task instruction was structured around the present, practice, produce (PPP) approach to instruction (see, amongst others, Nassaji & Fotos, 2011; Shintani, 2013) in which children were presented with the focused structures in an explicit way. In the case of the Gr (i.e. grammar) group, the children were given metalinguistic explanations and rules for regular and irregular past tense verb forms in affirmative and negative syntactic contexts and were provided with first language (L1) / second language (L2) contrasts every week (i.e. present). This was followed by the controlled production stage (i.e. practice), where they carried out fill-in-the gap, error correction, translation, sentence completion and jumbled sentences exercises as well as dictation and running-dictation or irregular verb competition games. Finally, this gave way to their weekly focused task performance (i.e. produce), the second part of their weekly intervention session explained below). As for the I group (i.e. interaction strategy), the children were given explicit instruction of interactional expressions and strategies (i.e. conversational adjustments, lexical requests and corrective feedback strategies) through the use of glossaries which included examples and L1 translations of clarification requests, comprehension checks, confirmation checks, repetitions, recasts and lexical requests. We also exemplified such strategies with video modelling with videos we had created specifically for the sessions where children could watch other children interacting and using such strategies and expressions (i.e. present). Children were asked to identify those expressions in the videos by raising flashcards depicting the expressions and also to repeat the expressions. They also participated in fill-in-the gap exercises, running dictation or sentence completion exercises so as to practice the expressions in a controlled way (i.e. practice). As in the Gr group, this led way to a much freer practice stage which included their weekly task performance explained below (i.e. produce). The Gr+I group (i.e. grammar+interaction strategy) was provided with a reduced version of the same activities carried out in the previous two groups, since the amount of time was the same. Again, this group also followed the PPP approach to pre-task instruction. It is important to note that the intervention activities explained so far were not all included in every 25-minute session. Every session included the present and practice stages (and the produce stage during task performance explained below) and the type of activities varied every week. In case we used the same type of activity, the content of the activity was of course different every week (for a sample of activities, see the Appendix in supplementary material).

For all the intervention groups, the remaining 25 minutes in every session included task-based peer interaction in the form of procedural repetition of both collaborative picture narrative and information-gap focused tasks the children carried out in pairs. This was the produce stage of every session, where children who were in pre-task instruction groups could actually use the structures (i.e. Past tense and/or interactional and corrective feedback strategies depending on the group) they had been previously taught. The tasks were the same in all experimental groups and all of them were focused on the past simple tense, the target structure of the study. The picture narrative tasks were collaborative story telling tasks where children were given six to eight vignettes which they had to describe as they told the story together. The prompts were all in the past tense and they were given vocabulary help if needed. The information-gap tasks consisted of a set of questions the students had to ask and answer in relation to different timetables or different sets of pictures each member of the pair had in order to answer one last general question together with the information the learners had been exchanging. Again, all the prompts were in the past tense (for a sample of such tasks, see the Appendix in supplementary material).

3 Grammaticality judgement task

An untimed grammaticality judgement task (GJT) was designed to tap into the explicit knowledge of past tense formation in English.⁴ The GJT included a total of 50 items divided across six conditions (10 items each target condition and 5 items for the control conditions) where we manipulated type (control vs. target), grammaticality (grammatical vs. ungrammatical) and verb type (regular vs. irregular).⁵ Ungrammaticality was introduced by deleting the past morpheme -ed in regular verbs, adding the past tense morpheme -ed in irregular verbs, replacing the auxiliary didn’t by not with and without the past tense morpheme on the verb and combining the negative auxiliary with the past tense morpheme -ed. Table 1 contains an outline of the six categories as well as example items for each condition.

Table 1.

Experimental conditions and example items in the grammaticality judgement task (GJT).

Type	Grammaticality	Verb type	Example items	N
Target	Grammatical	Regular	Last month you played with my friends. Yesterday she didn’t like the film.	10
		Irregular	Last week we went to the beach.On Wednesday I didn’t meet my friends.	10
	Ungrammatical	Regular	Last year we play a game.Last summer I not painted the house.	10
		Irregular	Last month we goed to school.Last Christmas we not sleep a lot.	10
Control	Grammatical	n/a	At home we can eat pizza.	5
	Ungrammatical	n/a	At school I play can tennis with my friends.	5

As shown in Table 1, in addition to the 40 target items, the GJT contained two unrelated conditions with 10 items (ration of 1-to-2) that served as both distractors and controls. Within these two conditions, one contained 5 grammatical sentences and the other one contained 5 ungrammatical sentences. These items were used to ensure participants could do the task appropriately and that they were able to detect ungrammaticality. We carefully designed the ungrammatical sentences so that they contained clear morphosyntactic violations that the participants would be sensitive to. This was particularly important as the proportion of accuracy in the control conditions was taken as an inclusion-criteria. All items in the sentences were level-appropriate and we carefully controlled for the verbs used throughout the experiment to make sure participants knew their meaning.

The items were presented randomized for each participant, and in a written and recorded oral form. Participants were then asked to judge the grammaticality using a binary response mode. Alongside the item, they saw three different emojis signalling different responses: happy = grammatical, sad = ungrammatical and confused = I am not sure. Once a participant selected one of the emojis, they were then asked to correct the sentence if they thought the sentence was ungrammatical.⁶ Participants had been given exact instructions as to what the emojis meant and did four practice items alongside the researcher in the room so that they could ask any pertinent questions. Once they felt comfortable with the task, they proceeded at their own pace monitored by the researcher.

4 Self-paced reading task

The self-paced reading task (SPR; Just et al., 1982) was administered to tap into the implicit knowledge of past tense formation in English.⁷ The presentation of the items was done on a moving-window paradigm, whereby participants saw a region at a time and moved to the following one by pressing the space bar. There were 60 items spread across six different conditions where we manipulated type (target vs. control), grammaticality (grammatical vs. ungrammatical) and verb type (regular vs. irregular). All items contained six regions. Ungrammaticality was introduced in the same way as in the GJT in the affirmative sentences. In the negative sentences and to control for the region length, we only combined the negative auxiliary with the past tense morpheme -ed. Each sentence started with an adverbial (region 1), followed by a pronoun (region 2), a verb (region 3: Critical Region), an object (region 4: Spill Over region), and a carrier phrase divided across two regions (region 5 and 6). We made sure that sentences were as uniform as possible by controlling the lexical items and segments used in the stimuli. Table 2 shows the six conditions, as well as example items for each.

Table 2.

Experimental conditions and example items in the self-paced reading task (SPR).

Type	Grammaticality	Verb type	Example items	N
Target	Grammatical	Regular	Last month / you / played / a game / with / my friend.Last week / we / didn’t touch / the floor / with / our hands.	12
		Irregular	Last month / you / took / a photo / with / your camera.Last week / we / didn’t take / a photo / our phone.	12
	Ungrammatical	Regular	Last year / we / play / a game / with / our dad.Last year / they / didn’t visited / their dad / with / their friends.	12
		Irregular	Last week / we / taked / a photo / with / our phone.Last month / we / didn’t went / to school / with / our brother.	12
Control	Grammatical	n/a	At school / I / can play / tennis / with / my teacher.	6
	Ungrammatical	n/a	At home / you / eat can / pasta / with / your mother.	6

In addition to the experimental items, we added content questions to check for accuracy and attention on a 1-to-3 ratio. This allowed us to ensure participants were in fact reading carefully through the sentences and paying attention to the actual meaning of the sentences. We only included those participants who had at least 80% accuracy in the content questions. In order to further ensure that the task worked and that participants were, indeed, sensitive to morphosyntactic violations we explored the contrast between the two control conditions.

Upon start of the task, participants were instructed to read the sentences at a normal pace and to pay attention to the global meaning of the sentence. First, they did four practice items monitored by the researcher and were then given the opportunity to ask anything they were not sure of. As soon as the practice items were over, they were instructed to start with the main experimental task.

5 Procedure

The experiment at both pre- and post-tests took place in a classroom setting and all the tasks were completed by each participant using their personal computer. To maximize efficiency of testing, each participant was tested alongside their cohort with the supervision of a researcher. We created and administered the two experimental tasks (GJT and SPR) using Gorilla Experiment Builder (www.gorilla.sc; Anwyl-Irvine et al., 2020) as it has been shown to be highly reliable to conduct experiments where reaction times are recorded (Bridges et al., 2020). The tasks involved the use of the same set of experimental items in both the pre-test and post-test; this was done to ensure that there was no effect of lexical item across testing times. All items were randomized for each participant and during each testing session. Given that the pre-test and post-test were conducted 10 weeks apart, we considered it unlikely to observe effects resulting from participants being re-exposed to the same items. Participants completed the Oxford Placement Test for Young Learners, the ELIAS grammar test II (Kersten et al., 2012) and a questionnaire tapping into their perceptions and motivation towards peer interaction that was designed as part of the project. Additionally, caregivers also answered a background questionnaire on the linguistic background of the participants. All participants as well as their caregivers provided informed consent to take part in the study and all procedures regarding this study, as well as the main project that this study belongs to, were given ethical approval by the pertinent research ethics committee. The study followed the guidelines in the Declaration of Helsinki regarding human testing for non-medical research.

V Results

1 GJT results (pre vs. post-test)

a Data treatment GJT

To assess accuracy in the GJT, responses were coded in a binary fashion. Correct responses indicated acceptance of a grammatical sentence and rejection of an ungrammatical one, while incorrect responses indicated the opposite. To ensure the reliability of the data, we applied two trimming procedures. First, we excluded all uncertain responses where participants indicated uncertainty about the grammaticality of the sentence.⁸ This resulted in the exclusion of 16.3% of the responses at Time 1 and 12.3% of the responses at Time 2. Second, we applied an exclusion criterion on participants to ensure their sensitivity to grammaticality in the task. Specifically, we excluded participants who demonstrated less than 66% accuracy⁹ across the two control conditions. This exclusion led to the removal of 11 participants, which comprised 8.1% of the total sample size.

b Results

To ensure the task worked well and that participants could detect (un)grammaticality, we began by exploring the two control conditions where we expected participants to be able to correctly detect grammatical and ungrammatical sentences. Descriptive results revealed that participants had high levels of accuracy in both grammatical (high rates of acceptance) and ungrammatical (high rates of rejection) conditions across both Time 1 (grammatical: 91.89 [SD = 26]; ungrammatical: 80.94 [SD = 39]) and Time 2 (grammatical: 91.86 [SD = 28]; ungrammatical: 83.57 [SD = 31]). The high levels of accuracy in the control condition are taken as evidence that the task worked (see Appendix in supplemental material).

Moving on to the conditions of interest, we first examined those involving regular verbs. As shown in Table 3 and Figure 1, participants responded accurately in the regular grammatical condition at both times and in all intervention groups. At Time 1, participants in all intervention groups performed poorly in the regular ungrammatical condition, with the descriptive results indicating some improvement in certain intervention groups. In fact, the four groups with some sort of intervention showed some moderate descriptive improvement in accuracy in the ungrammatical conditions between Time 1 and Time 2 (grammar+interaction strategy: +28%; grammar: +15%; interaction strategy: +8%; task-only: +7%) while the control group did not show much descriptive improvement (+3%).

Table 3.

Descriptive statistics (Mean + SD) for the regular condition across intervention groups, times and conditions.

Intervention	Grammaticality	Regular Time 1		Regular Time 2
		Mean	SD	Mean	SD
Grammar+interaction	Grammatical	0.90	0.30	0.97	0.18
strategy	Ungrammatical	0.17	0.38	0.44	0.50
Grammar	Grammatical	0.92	0.27	0.89	0.31
	Ungrammatical	0.21	0.41	0.36	0.48
Interaction strategy	Grammatical	0.90	0.30	0.91	0.28
	Ungrammatical	0.19	0.39	0.27	0.44
Task-only	Grammatical	0.94	0.23	0.94	0.24
	Ungrammatical	0.21	0.41	0.28	0.45
Control	Grammatical	0.95	0.22	0.91	0.29
	Ungrammatical	0.12	0.32	0.15	0.36

Figure 1.

Stacked-bar chart displaying the proportion of correct/incorrect responses at both Time 1 and Time 2 for the grammatical and ungrammatical regular conditions across all groups.

To determine which intervention led to the greatest increase in accuracy, we used a generalized linear mixed-effects model with a binomial family and the lme4 (Bates et al., 2015) and multcomp (Hothorn et al., 2008) packages for post-hoc comparisons in the R environment. We conducted three separate analyses for each pair-condition (Model 1: control conditions, Model 2: regular conditions, Model 3: irregular conditions). We started by fitting the fixed effects of grammaticality (grammatical vs. ungrammatical condition), intervention group (Gr+I, Gr, I, T or C), time (Time 1 vs. Time 2), proficiency (continuous variable) and all possible interactions. We also included random intercepts for participants and item. After model selection, the model with best fit included only grammaticality, intervention group and time as fixed effects and their interaction.

The main outcomes of the model showed a main significant effect of grammaticality (estimate = −4.15, SE = .31, z = −13.26, p < .001), intervention group (estimate = .77, SE = .38, z = 2.01, p = .043) and time (estimate = 1.25, SE = .38, z = 2.01, p = .012) and multiple two-way and three-way interactions, which we further explored using planned post-hoc comparisons, which are reported in Table 4. The main effect of grammaticality indicates that participants generally had higher rates of accuracy in the grammatical condition as opposed to the ungrammatical one, the main effect of time indicates higher rates of accuracy at Time 2 against Time 1 and the main effect of intervention indicates that intervention group led to differential gains in accuracy. We further explored the significant three-way interaction of time, grammaticality, and intervention, which showed that the only group who obtained significant improvement in accuracy from Time 1 to Time 2 was the grammar+interaction strategy group, albeit the improvement was only seen in the ungrammatical condition (this contrast is shaded in grey in Table 4). Recall, however, that the proportion of correct responses in the grammatical condition was already high at Time 1 of testing, leaving less room for improvement. All other intervention groups performed similarly at both times of testing.

Table 4.

Relevant post-hoc comparisons for the model on the regular conditions.

Looking at the irregular verb conditions, Table 5 and Figure 2 display the results, which indicate that at Time 1, all participants had relatively high levels of proportion of accuracy in the irregular grammatical condition (0.86 [SD = 0.33]), and low accuracy in the ungrammatical condition (0.20 [SD = 0.40]), regardless of their intervention group. Moving to Time 2, there appears to be a moderate improvement in the grammatical condition (0.93 [SD = 0.23]), with the proportion of accuracy reaching ceiling levels for the four groups who received some intervention. However, there does not seem to be much of an observable improvement in the ungrammatical condition (0.24 [SD = 0.42]). Overall, the results suggest that while the participants showed progress in the irregular grammatical condition, they still struggled with the ungrammatical one.

Table 5.

Descriptive statistics (Mean + SD) for the irregular condition across intervention groups, times and conditions.

Intervention	Grammaticality	Irregular time 1		Irregular time 2
		Mean	SD	Mean	SD
Grammar+interaction	Grammatical	0.85	0.36	0.95	0.21
Strategy	Ungrammatical	0.24	0.43	0.33	0.47
Grammar	Grammatical	0.81	0.39	0.96	0.20
	Ungrammatical	0.25	0.44	0.20	0.40
Interaction strategy	Grammatical	0.86	0.35	0.94	0.24
	Ungrammatical	0.21	0.41	0.23	0.42
Task-only	Grammatical	0.89	0.32	0.95	0.21
	Ungrammatical	0.24	0.43	0.28	0.45
Control	Grammatical	0.92	0.27	0.89	0.31
	Ungrammatical	0.13	0.34	0.18	0.38

Figure 2.

Stacked-bar chart displaying the proportion of correct/incorrect responses at both Time 1 and Time 2 for the grammatical and ungrammatical irregular conditions across all groups.

To explore the effect of intervention group on accuracy in the irregular conditions, we performed the same analysis and procedure as the one done for the regular condition. The final model with best fit included only grammaticality, intervention group and time. The results indicate a significant main effect of grammaticality (estimate = 1.86, SE = .26, z = 7.01, p < .001) and intervention (estimate = .82, SE = .32, z = 2.53, p < .001). The main effect of grammaticality indicated that irrespective of intervention group and time, participants gave higher proportions of accuracy to the grammatical condition than the ungrammatical one. The main effect of intervention indicated that some intervention groups had higher accuracy scores overall than some other groups. In fact, the group with the lowest overall accuracy was the control group. There was also a significant three-way interaction of grammaticality, time and intervention group, which we explored by looking at the relevant post-hoc contrasts shown in Table 6. The two significant contrasts for the grammar+interaction strategy and the grammar group in the grammatical condition (shaded in grey) show that these two groups obtained higher proportions of accuracy at Time 2 in the grammatical condition only. None of the other groups showed a significant improvement from Time 1 to Time 2 in either of the conditions.

Table 6.

Relevant post-hoc comparisons for the model on the irregular conditions.

2 SPR results (pre vs. post-test)

a Data treatment SPR

As mentioned above, comprehension questions were included during the SPR task to ensure participants’ attention. We excluded 5 children whose accuracy on these questions was below 80%. Following this, we performed outlier trimming using the method outlined by Keating and Jegerski (2015). The low cut-off point was set at 100 ms, and any values lower than this were replaced with 100ms. For the high cut-off point, we chose the mean plus two standard deviations, resulting in a data loss of 4.79%. We then transformed the raw reaction times to residual reaction times, which neutralized the effects of region length to facilitate the comparison across regions.

b Results

To analyse the SPR data, we ran linear mixed effects models using the lme4 (Bates et al., 2015) and multcomp (Hothorn et al., 2008) packages for post-hoc comparisons in R. We conducted three separate analyses for each pair-condition (Model 1: control conditions, Model 2: regular conditions, Model 3: irregular conditions). Within each analysis, we explored the effects and interactions of grammaticality (grammatical vs. ungrammatical condition), intervention group (Gr+I, Gr, I, T, C), time (Time 1 vs. Time 2), region (Critical Region (CR) vs. Spill Over (SO) Region) and proficiency at Time 1. We utilized similar model selection procedures as outlined above and reported the model of best fit.

Our first analysis aimed to establish that the task indeed captured grammatical sensitivity in our participants. We explored the control conditions where we hypothesized that if the task worked, participants would exhibit a slowing-down effect in the ungrammatical condition at both Time 1 and Time 2, regardless of the intervention group. Figure 3 illustrates the contrast between the conditions for all groups combined – all descriptive statistics including residual reaction times (RTs) (means and SDs) for both the critical and spill-over region can be found in the Appendix in supplementary material.

Figure 3.

Line graph illustrating the mean residual reaction times (RTs) for all regions across both grammatical and ungrammatical control conditions (all groups together) at Time 1 of testing.

The results clearly show that all participants significantly slowed down in the ungrammatical condition, indicating that they were sensitive to the grammatical violation in those conditions as evidenced in the significant main effect of grammaticality (Estimate = 89.65, SE = 11.32, t = 7.92, p > .001) in the final model with the best fit. It is worth mentioning that there is also a significant main effect of region (Estimate = 34.49, SE = 20.73, t = −12.91, p > .001) and time (Estimate = −267.61, SE = 11.31, t = 3.95, p > .001). The region main effect indicates that participants are generally slower at the spill over region and the time main effect indicates that participants are generally quicker over all at Time 2 of testing irrespective of grammaticality.

As for potential changes in the regular conditions, our hypothesis was that if any of the interventions had led to improvement, there would be a significant slowing down effect in the ungrammatical condition at Time 2. As is visually appreciated in Figure 4 and can be seen in the descriptives in the Appendix in supplementary material, none of the groups show sensitivity to the ungrammatical condition at Time 2.

Figure 4.

Line graph illustrating the mean residual reaction times (RTs) for all regions across both grammatical and ungrammatical regular conditions (divided by group).

The statistical model with the best fit for this pair-condition included time and intervention type, as well as their interaction. All other fixed effects and interactions did not improve the fit of the model, indicating that grammaticality, region or proficiency did not explain any of the variance. The model showed a significant main effect of time (Estimate = −277.42, SE = 15.08, t = −18.39, p > .001), which indicates that participants have significantly quicker readings times at Time 2 as opposed to Time 1, as shown in Table 7.

Table 7.

Relevant comparisons for the model on the regular conditions.

Time 1 vs. Time 2: Regular condition
Grammar+interaction strategy	Estimate = 163.11, SE = 17.1, z = 9.59, p < .001
Grammar	Estimate = 292.64, SE = 17.7, z = 16.53, p < .001
Interaction strategy	Estimate = 217.62, SE = 18.1, z = 12.05, p < .001
Task-only	Estimate = 325.04, SE = 18.1, z = 17.97, p < .001
Control	Estimate = 277.42, SE = 15.1, z = 18.39, p < .001

Our hypothesis for the irregular pair-conditions was that a positive effect from the intervention types would result in a slowing down effect in the ungrammatical condition at Time 2. Figure 5 provides a visual representation of the data for all groups at Time 1 and Time 2.

Figure 5.

Line graph illustrating the mean residual reaction times (RTs) for all regions across both grammatical and ungrammatical irregular conditions (divided by group).

To build the model, we used similar procedures as with the regular conditions, and the best-fit model included time and intervention type as fixed-effects, along with their interaction. None of the other factors were found to contribute significantly to the model. The results indicated a similar outcome to the regular conditions, with a main effect of time (Estimate = −316.71, SE = 15.74, t = −20.11, p > .001) indicating that all groups were significantly quicker at Time 2 as opposed to Time 1 irrespective of all other variables considered. This is shown in all significant contrasts presented in Table 8.

Table 8.

Relevant comparisons for the model on the irregular conditions.

Time 1 vs. Time 2: Irregular conditions
Grammar+interaction strategy	Estimate = 148.52, SE = 17.8, z = 8.39, p < .001
Grammar	Estimate = 303.11, SE = 18.5, z = 16.41, p < .001
Interaction strategy	Estimate = 254.15, SE = 18.8, z = 8.36, p < .001
Task-only	Estimate = 319.98, SE = 18.8, z = 16.94, p < .001
Control	Estimate = 316.71, SE = 15.7, z = 20.11, p < .001

VI Discussion

Our study set out to explore whether a focused task-based peer interaction 8-week treatment led to increased explicit and implicit knowledge of past simple tense among young EFL learners and whether adding pre-task explicit grammar instruction or explicit interactional instruction or both had any impact on the results. Considering that we were working with young learners, we wanted to make sure that tasks employed in the study effectively captured the children’s sensitivity to (un)grammaticality. Our results in the control conditions in both tasks and at the two testing times confirm that the tasks captured acceptance/rejection of grammaticality/ungrammaticality in the GJT and a slowing down effect in the CR or SO of control ungrammatical sentences in the SPR.

The two research questions in the study will be explored together, first in relation to explicit knowledge (GJT) and then in relation to implicit knowledge (SPR). Findings from the GJT in the regular condition indicate a high proportion of accurate judgements of grammatical sentences at both times and by all groups, possibly indicating some pre-study familiarity with the target structure. This cannot be attributed to a response bias towards grammaticality, as shown in our initial assessment of participants’ sensitivity to grammaticality in both the control grammatical and ungrammatical conditions. We observed high accuracy (indicating high acceptance for grammatical items and high rejection for ungrammatical items), which suggests that the higher-than-expected accuracy in the experimental conditions should not be attributed to a response bias. If this was a general bias effect, such effect should be seen across conditions. However, we understand that the over-acceptance of the ungrammatical conditions might be due to an inclination to accept what is not familiar to the participant. As for the judgements of ungrammatical regular conditions, the four treatment groups showed low accuracy at Time 1 and moderate descriptive improvement at Time 2, while the control group did not show any development, suggesting some degree of treatment effects. The Gr+I group significantly improved their judgement of regular ungrammatical sentences, indicating that pre-task explicit instruction of the past tense together with explicit instruction of interactional and corrective feedback strategies led to greater gains in explicit knowledge. In the irregular past condition, the control group showed the lowest accuracy scores and the two groups with pre-task explicit grammar instruction (Gr and Gr+I) significantly improved their accurate judgements of grammatical sentences, reaching ceiling levels at Time 2. No significant improvements were observed in the judgements of ungrammatical irregular past sentences, suggesting that there was an additional layer of complexity involving lexical memorization of the irregular forms. The overall results in the explicit task are generally in line with previous research with adult and adolescent populations, where the condition that favours greater gains in explicit knowledge is the one that combines explicit pre-task instruction and task performance (Khezrlou, 2021; Li et al., 2016, 2018; Shintani, 2019; Williams & Evans, 1998), lending support to claims that low proficiency learners in limited exposure contexts need to draw on explicit instruction and practice opportunities for learning to be effective (Shintani, 2019). As was the case with Shintani (2019) and one of the groups in Khezrlou (2021), our learners were actually exposed to task procedural repetition (Bygate, 2018) and explicit instruction for a number of sessions. Yet our study explores young learners, who might surely need more time and more intensity of exposure to benefit from peer interaction tasks in general and from pre-task instruction in particular. Our young learners find themselves in a very limited context of exposure to the FL, as evidenced by the limited accumulated and weekly exposure, and they had never participated in task-based peer interaction before. The group that only participated in focused tasks (i.e. the task-only group) and the one who took part in the tasks and received interactional strategy instruction (i.e. Interaction strategy group) might even need more intensive exposure, as interactional instruction and/or focused tasks alone for 25 minutes a week might not provide with enough opportunities for incidental learning and might not elicit the use and practice of the target structure which would then lead to improvement in explicit L2 knowledge.

Results from the SPR, the task tapping on the children’s implicit knowledge, did not provide evidence of any relevant developments. Although the task captured sensitivity to ungrammaticality in the control conditions, children did not show clear signs of implicit knowledge development for the past tense, the linguistic domain under investigation. In the regular condition, grammatical and ungrammatical sentences obtained very similar reaction times at Time 1 and at Time 2 with a significant main effect of time by which reaction times were quicker across intervention groups. This suggests the children were more familiar with the sentences and the task but did not react differently to grammatical and ungrammatical sentences, indicating no development in implicit knowledge of the structure being examined. As for the irregular condition, a very similar picture emerges, with a significant time effect across groups. Yet, at Time 2, the two groups that had received pre-task grammar instruction (i.e. Gr and Gr+I) displayed a very moderate and only descriptive trend which indicates a slight slowing down effect in the CR for the ungrammatical sentences. It is only a trend that does not show signs of statistical significance but it might point towards an incipient potential development or a pre- developmental stage. These results coincide with Li et al.’s (2016, 2018) studies in that implicit knowledge did not seem to be affected by any treatment type, while Khezrlou (2021) did find that the group with explicit instruction and task repetition obtained higher results with implicit knowledge after the second post test and at the delayed post test. However, Khezrlou’s participants were young adults, the implicit knowledge test was different (i.e. an EIT) and their intervention had been more intense (i.e. on consecutive days over a week).

Our findings suggest that explicit pre-task instruction and the use of focused tasks throughout the treatment period were beneficial for young learners’ explicit knowledge of at least the structure under study. All treatment groups, who participated in focused tasks every week, showed some degree of development in explicit knowledge and the groups that improved the most in the explicit test were the ones which received pre-task grammar instruction as well as interactional instruction or pre-task grammar instruction alone. As seen in 4.2, the children in these groups were presented with the structures under study in an explicit way, following a PPP instructional approach (see, amongst others, Nassaji & Fotos, 2011; Shintani, 2013). They were provided with metalinguistic explanations as well as L1/L2 contrasts. This was followed by controlled production exercises where they practised the structures and then were encouraged to produce them during task performance. This approach to instruction is closely linked to DeKeyser’s (2007, 2015, 2017) skill acquisition theory, according to which these children were provided with declarative knowledge every week which they could then put into practice through focused task procedural repetition over the 8-week intervention. This enhanced their knowledge of the structure, which is reflected in their results in the explicit task. Our findings seem to confirm that in an input-limited environment such as the one explored in this study, explicit teaching is particularly useful for young learners and can actually enhance their (explicit) language development (see, amongst others, Harley, 1998; Lichtman, 2016; Milton & Alexiou, 2006). The fact that the children’s implicit knowledge was not found to clearly develop suggests the learners might not have had enough time, exposure intensity and/or communicative practice opportunities for knowledge to become automatized. As is known, implicit learning is gradual and requires large amounts of input and time (Roehr-Brackin, 2024).

The group who received linguistic as well as interactional pre-task instruction was the group who showed greatest improvement in explicit knowledge, in both the regular ungrammatical and the irregular grammatical conditions. This group was followed by the group who only received pre-task grammar instruction. The Gr+I group appears to have benefitted most from the use of tasks. During their weekly task practice part of the session, they were encouraged to practise the structure they had received instruction on (in the same way as in the Gr only group), use the interactional expressions provided in pre-task instruction, and even provide corrective feedback on the structure while performing the tasks. The weekly type of instruction including both structure-related and interactional strategies and feedback-related instruction seems to have provided the children with a favourable environment to develop their declarative knowledge even further and practice it during the tasks (even if the amount of time devoted to pre-task instruction of grammar and interaction was obviously lower than in the Gr only or I only groups, which devoted all their instruction time to grammar or interactional strategies alone).

VII Conclusions

This study has contributed new data on the effects of task-based peer interaction and pre-task explicit instruction on the development of explicit and implicit knowledge of past tense among young learners, an under researched population in the area of study (García Mayo, 2021; Roehr-Brackin, 2024). Results showed that the use of pre-task instruction of both linguistic and interactional type together with focused peer-interaction tasks contributed to the children’s explicit learning of past tense. Particularly in young learner EFL contexts of limited exposure, knowing to what extent peer-interaction focused tasks can be used to foster grammar development and that pre-task instruction might further contribute to such learning greatly contributes to the implementation of tasks in the primary EFL classrooms. It is often suggested that production tasks might not be suitable for beginners or young learners under limited input conditions (see Ellis, 2020) but the present study suggests that their use is not only possible but also beneficial in combination with some degree of explicit instruction on the focused structure and on interactional and corrective feedback strategies, confirming previous research on the benefits of explicit teaching for young learners (Harley, 1998; Lichtman, 2016; Milton & Alexiou, 2006).

A number of limitations need to be acknowledged. Although the study has provided an 8-week intervention, a much longer period might have been needed in such limited exposure environments for results to be more robust. Particularly for the groups that did not receive pre-task explicit instruction, a more intense and longer treatment might have been more fruitful, and it might have even shown effects on the children’s implicit knowledge in other groups as well. However, a longer intervention working on the same target structure would not be ecologically valid or feasible within the set curriculum. It would be desirable to see if incorporating similar interventions in a regular manner would lead to longer-term gains. A delayed post-test was not possible but it might have also added relevant data to the study of L2 knowledge development. In addition, further research should test additional structures and use a wider variety of tests to confirm or reject our findings.

Supplemental Material

sj-docx-1-ltr-10.1177_13621688241267367 – Supplemental material for The effect of task-based peer interaction and pre-task instruction on young EFL learners’ explicit and implicit knowledge of past tense: An intervention study

Supplemental material, sj-docx-1-ltr-10.1177_13621688241267367 for The effect of task-based peer interaction and pre-task instruction on young EFL learners’ explicit and implicit knowledge of past tense: An intervention study by Elisabet Pladevall-Ballester, Eloi Puig-Mayenco and Montserrat Capdevila in Language Teaching Research

Footnotes

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the Spanish Ministry of Science and Innovation (PID2019-107328GB-I00) and the funded research group GReLA (Grup de Recerca en Lingüística Anglesa: Ús i Adquisició en Contextos Multilingües), Grup de Recerca Consolidat de la Generalitat de Catalunya (2021SGR00684).

ORCID iDs

Elisabet Pladevall-Ballester

Eloi Puig-Mayenco

Supplementary material

Supplementary material for this article is available online.

Notes

References

Anwyl-Irvine

A.L.

Massonié

Flitton

Kirkham

N.Z.

Evershed

J.K.

(2020). Gorilla in our midst: An online behavioural experiment builder. Behavior Research Methods, 52, 388–407.

Azkarai

Imaz Agirre

(2016). Negotiation of meaning strategies in child EFL mainstream and CLIL settings. TESOL Quarterly, 50, 844–870.

Azkarai

Kopinska

(2020). Young EFL learners and collaborative writing: A study on patterns of interaction, engagement in LREs, and task motivation. System, 94, 102338.

Bates

Mächler

Bolker

Walker

(2015). Fitting linear mixed-effects models using lme 4. Journal of Statistical Software, 67, 1–48.

Bridges

Pitiot

MacAskill

Peirce

(2020). The timing mega-study: Comparing a range of experiment generators, both lab-based and online. PeerJ, 8, e9414.

Bygate

(2018). Learning language through task repetition. John Benjamins.

Calzada

García Mayo

M.P.

(2020). Child EFL grammar learning through a collaborative task. In Suzuki

Storch

(Eds.), Languaging in language learning (pp. 19–39). John Benjamins.

Collins

Muñoz

(2016). The foreign language classroom: Current perspectives and future considerations. The Modern Language Journal, 100, 133–147.

DeKeyser

R.M.

(2007). Practice in a second language: Perspectives from applied linguistics and cognitive psychology. Cambridge University Press.

10.

DeKeyser

R.M.

(2015). Skill acquisition theory. In VanPatten

Williams

(Eds.), Theories in second language acquisition: An introduction (pp. 94–112). Routledge.

11.

DeKeyser

R.M.

(2017). Knowledge and skill in ISLA. In Loewen

Sato

(Eds.), The Routledge handbook of instructed second language acquisition (pp. 15–32). Routledge.

12.

DelGiudice

(2018). Middle childhood: An evolutionary-developmental synthesis. In Halfon

Forrest

C.B.

Lerner

R.M.

Faustman

E.M.

(Eds.), Handbook of life course health development (pp. 95–107). Springer.

13.

Doughty

Varela

(1998). Communicative focus on form. In Doughty

Williams

(Eds.), Focus on form in classroom second language acquisition (pp. 114–138). Cambridge University Press.

14.

Ellis

(2003). Task-based language learning and teaching. Oxford University Press.

15.

Ellis

(2009). The differential effects of three types of task planning on the fluency, complexity, and accuracy in L2 oral production. Applied Linguistics, 30, 474–509.

16.

Ellis

(2016). Focus on form: A critical review. Language Teaching Research, 20, 405–428.

17.

Ellis

(2018). Reflections on task-based language teaching. Multilingual Matters.

18.

Ellis

(2020). Task-based language teaching for beginner-level young learners. Language Teaching for Young Learners, 2, 4–27.

19.

Ellis

Loewen

Erlam

(2006). Implicit and explicit corrective feedback and the acquisition of L2 grammar. Studies in Second Language Acquisition, 28, 339–368.

20.

Ellis

Skehan

Shintani

Lambert

(2020). Task-based language teaching: Theory and practice. Cambridge University Press.

21.

Fujii

Ziegler

Mackey

(2016). Peer interaction and metacognitive instruction in the EFL classroom. In Sato

Ballinger

(Eds.), Peer interaction and second language learning: Pedagogical potential and research agenda (pp. 63–89). John Benjamins.

22.

Gass

Spinner

Behney

(2017). Salience in second language acquisition and related fields. In Gass

Spinner

Behney

(Eds.), Salience in second language acquisition (pp. 1–18). Routledge.

23.

García Mayo

M.P

. (2021). Child interaction in task-supported EFL/CLIL contexts. In Ahmadian

Long

M.H.

(Eds.), The Cambridge handbook of task-based language teaching (pp. 397–415). Cambridge University Press.

24.

García Mayo

M.P.

Imaz Agirre

. (2016). Task repetition and its impact on EFL children’s negotiation of meaning strategies and pair dynamics. An exploratory study. The Language Learning Journal, 44, 451–466.

25.

García Mayo

M.P.

Imaz Agirre

. (2017). Child EFL interaction; age, instructional setting and development. In Enever

Lindgren

(Eds.), Researching the complexity of early language learning in instructed contexts (pp. 249–68). Multilingual Matters.

26.

García Mayo

M.P.

Imaz Agirre

. (2019). Task modality and pair formation method: Their impact on patterns of interaction and attention to form among EFL primary school children. System, 80, 165–175.

27.

García Mayo

M.P.

Lázaro Ibarrola

. (2015). Do children negotiate for meaning in task-based interaction? Evidence from CLIL and EFL settings. System, 54, 40–54.

28.

Gass

S.M.

Mackey

(2020). Input, Interaction and Output in L2 Acquisition. In VanPatten

Keating

G.D.

Wulff

(Eds.), Theories in second language acquisition: An introduction (pp. 192–222). Routledge.

29.

Harley

(1998). The role of focus-on-form tasks in promoting child L2 acquisition. In Doughty

C.J.

Williams

(Eds.), Focus on form in classroom second language acquisition (pp. 156–174). Cambridge University Press.

30.

Holmes

Myles

(2019). White Paper: Primary Languages Policy in England – The Way Forward. http://www.ripl.uk/policy/

31.

Hothorn

Bretz

Westfall

Heiberger

R.M.

(2008). Multcomp: Simultaneous inference in general parametric models: R package version 1.0-0. R Foundation for Statistical Computing.

32.

Just

Carpenter

Woolley

(1982). Paradigms and processes in reading comprehension. Journal of Experimental Psychology: General, 111, 228–238.

33.

Keating

G.D.

Jegerski

(2015). Experimental designs in sentence processing research: A methodological review and user’s guide. Studies in Second Language Acquisition, 37, 1–32.

34.

Kersten

Piske

Rohde

, et al. (2012). ELIAS grammar test II. Studies on Multilingualism in Language Education, 1.

35.

Khezrlou

(2021). Explicit instruction through task repetition: Effects on explicit and implicit knowledge development. Language Awareness, 30, 62–84.

36.

Lázaro Ibarrola

Azpilicueta Martínez

. (2015). Investigating negotiation of meaning in EFL children with very low levels of proficiency. International Journal of English Studies, 15, 1–21.

37.

Lázaro Ibarrola

Hidalgo

M.A

. (2017). Benefits and limitations of conversational interactions among young learners of English in a CLIL context. In García Mayo

M.P.

(Ed.), Learning foreign languages in primary school: Research insights (pp. 86–102). Multilingual Matters.

38.

Ellis

Kim

(2018). The influence of pre-task grammar instruction on L2 learning: An experimental study. Studies in English Education, 23, 831–857.

39.

Ellis

Zhu

(2016). Task-based versus task-supported language instruction: An experimental study. Annual Review of Applied Linguistics, 36, 205–229.

40.

Lichtman

(2016). Age and learning environment: Are children implicit second language learners? Journal of Child Language, 43, 707–730.

41.

Loewen

(2015). Introduction to instructed second language acquisition. Routledge.

42.

Loewen

Sato

(2018). Interaction and instructed second language acquisition. Language Teaching, 51, 285–329.

43.

Long

M.H.

(1996). The role of the linguistic environment in second Language acquisition. In Bhatia

T.K.

Ritchie

W.C.

(Eds.), Handbook of language acquisition (pp. 413–468). Academic Press.

44.

Long

M.H.

(2015). Second language acquisition and task-based language teaching. Wiley.

45.

Long

M.H.

(2017). Instructed second language acquisition (ISLA): Geopolitics, methodological issues, and some major research questions. Instructed Second Language Acquisition, 1, 7–44.

46.

Loschky

Bley-Vroman

(1993). Grammar and task-based methodology. In Crookes

Gass

(Eds.), Tasks and language learning (pp. 123–163). Multilingual Matters.

47.

Lyster

(2004). Differential effects of prompts and recasts in form-focused instruction. Studies in Second Language Acquisition, 26, 399–432.

48.

Lyster

Saito

Sato

(2013). Oral corrective feedback in second language classrooms. Language Teaching, 46, 1–40.

49.

Mackey

(2006). Feedback, noticing and instructed second language learning. Applied Linguistics, 27, 405–430.

50.

Mackey

(2012). Input, interaction, and corrective feedback in L2 learning. Oxford University Press.

51.

Mackey

Goo

(2007). Interaction research in SLA: A meta-analysis and research synthesis. In Mackey

(Ed.), Conversational interaction in second language acquisition: A series of empirical studies (pp. 407–453). Oxford University Press.

52.

Milton

Alexiou

(2006). Language aptitude development in young learners. In Abello-Contesse

R. DeKeyser

, et al. (Eds.), The age factor in L2 acquisition and teaching (pp. 177–192). Peter Lang.

53.

Nassaji

Fotos

(2011). Teaching grammar in second language classrooms: Integrating form-focused instruction in communicative contexts. Routledge.

54.

Oliver

(1998). Negotiation of meaning in child interactions. The relationship between conversational interaction and second language acquisition. Modern Language Journal, 82, 372–386.

55.

Oliver

Philp

Duchesne

(2017). Children working it out together: A comparison of younger and older learners collaborating in task based interaction. System, 69, 1–14.

56.

Ortega

(1999). Planning and focus on form in L2 oral performance. Studies in Second Language Acquisition, 21, 109–148.

57.

Philp

Adams

Iwashita

(2014). Peer interaction and second language learning. Routledge.

58.

Pienemann

(2005). Cross-linguistic aspects of processability theory. John Benjamins.

59.

Pinter

(2006). Teaching young language learners. Oxford University Press.

60.

Pinter

(2007). Some benefits of peer-peer interaction: 10-year-old children practicing with a communication task. Language Teaching Research, 11, 189–207.

61.

Pladevall-Ballester

(2021). Pair dynamics and language-related episodes in child EFL task-based peer interaction. Language Teaching for Young Learners, 3, 189–213.

62.

Pladevall-Ballester

Vraciu

(2020). EFL child peer interaction: Measuring the effect of time, proficiency pairing and language of interaction. Studies in Second Language Learning and Teaching, 10, 449–472.

63.

Roehr-Brackin

(2024). Explicit and implicit knowledge and learning of an additional language: A research agenda. Language Teaching, 57, 68–86.

64.

Sato

(2020). Metacognitive instruction for collaborative interaction: The process and product of self-regulated learning in the Chilean EFL context. In Lambert

Oliver

(Eds.), Using tasks in second language teaching: Practice in diverse contexts (pp. 215–236). Multilingual Matters.

65.

Sato

Dussuel Lam

(2021). Metacognitive instruction with young learners: A case of willingness to communicate, L2 use, and metacognition of oral communication. Language Teaching Research, 25, 899–921.

66.

Sato

Loewen

(2018). Metacognitive instruction enhances the effectiveness of corrective feedback: Variable effects of feedback types and linguistic targets. Language Learning, 68, 507–545.

67.

Sato

Lyster

(2012). Peer interaction and corrective feedback for accuracy and fluency development: Monitoring, practice, and proceduralization. Studies in Second Language Acquisition, 34, 591–626.

68.

Sato

(2010). Reconsidering the effectiveness and suitability of PPP and TBLT in the Japanese EFL classroom. JALT Journal, 32, 189–200.

69.

Shintani

(2013). The effect of focus on form and focus on forms instruction on the acquisition of productive knowledge of L2 vocabulary by young beginning-level learners. TESOL Quarterly, 47, 36–62.

70.

Shintani

(2019). The roles of explicit instruction and guided practice in the proceduralization of a complex grammatical structure. In DeKeyser

R.M.

Prieto Botana

(Eds.), Doing SLA research with implications for the classroom: Reconciling methodological demands and pedagogical applicability (pp. 83–106). John Benjamins.

71.

Williams

Evans

(1998). What kind of focus and on which forms? In Doughty

Williams

(Eds.), Focus-on-form in classroom second language acquisition (pp. 139–155). Cambridge University Press.

72.

Willis

(2007). Doing task-based teaching. Oxford University Press.

73.

Willis

(1996). A framework for task-based learning. Longman.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

2.07 MB