Abstract
This study examines how English as a Lingua Franca (ELF) interlocutors employ communication strategies during task-based interaction and how these strategies relate to ELF-aware communicative orientations of intelligibility, adaptability, and multilingual–multicultural awareness. Ten university students learning English as an additional language completed a closed information-gap task. Interactional data were video-recorded, transcribed, and analyzed through a combination of distributional analysis and interactional analysis, and complemented by stimulated recall to examine learners’ strategic awareness. The findings show that, beyond commonly reported strategies, participants frequently engaged in co-creating messages, deconstructing complex information, and adaptive self-repair. These strategies enabled learners to collaboratively clarify meaning, adjust their linguistic production in response to interactional demands, and draw on shared multilingual resources. The analysis illustrates how these strategies support three ELF-aware communicative orientations: intelligibility through collaborative clarification and reformulation, adaptability through self-initiated monitoring and adjustment, and multilingual awareness through the use of shared linguistic resources. The findings highlight strategic competence as an interactionally enacted resource that supports collaborative and adaptive meaning-making during task-based interaction, and underscore the pedagogical value of tasks that promote these interactional processes.
Keywords
Introduction
Communication strategies (CSs) are essential for successful interaction when interlocutors face linguistic limitations or misunderstandings. Research in second language acquisition (SLA) and English as a Lingua Franca (ELF) has shown that CSs enable speakers to negotiate meaning, compensate for linguistic gaps, and co-construct meaning across diverse linguistic backgrounds (Bjorkman, 2014; Cogo and Pitzl, 2016; Mauranen, 2006). However, their pedagogical integration into instructed L2 contexts remains limited, as most ELF research has focused on spontaneous conversations rather than structured classroom interaction (Savignon, 2017).
Recent developments in ELF-aware pedagogy (Bayyurt and Sifakis, 2015; Jenkins, 2015) have emphasized communicative orientations prioritizing intelligibility, adaptability, and multilingual–multicultural awareness. These orientations provide a useful perspective for understanding how learners manage communication in multilingual settings. However, there is limited empirical research examining how such orientations are enacted through learners’ strategic behavior in task-based interaction.
To address this gap, the present study explores how multilingual interlocutors employ CSs during task-based interaction and analyzes how these strategies function in interaction and reflect ELF-aware communicative orientations. Drawing on interactional data and stimulated recall interviews, the study provides an empirically grounded account of learners’ strategic behavior that can inform ELF-aware task design and pedagogy. The primary contribution of this study lies in its fine-grained interactional analysis of how CSs are enacted in task-based contexts, with ELF-aware orientations serving as an interpretive lens for understanding these processes.
Literature review
CSs in SLA and ELF research
CSs in SLA are generally understood as resources that learners use to negotiate meaning, compensate for linguistic limitations, and resolve communication breakdowns during interaction. A common distinction is drawn between compensatory strategies addressing linguistic gaps and interactional strategies for facilitating negotiation and conversational management (Littlemore, 2003). From an interactionist perspective, such strategies contribute to interlanguage development by creating opportunities for modified output and feedback (Fernandez-Garcia, 2006; Long, 1996).
A substantial body of SLA research has examined CSs through taxonomy development, frequency analysis, and instructional intervention. This work has identified a wide range of strategy types and demonstrated the pedagogical value of strategy instruction (e.g., Dörnyei and Scott, 1997; Lam, 2006; Nakatani, 2005). However, it has largely treated CSs as individual responses to linguistic limitations, focusing on categorization and instructional outcomes rather than on how strategies emerge and develop through interaction between interlocutors.
Research in ELF has extended this perspective by examining communication among speakers from diverse linguistic and cultural backgrounds. In such contexts, CSs function not only to resolve difficulties but also to maintain intelligibility and sustain interactional flow (Cogo and House, 2018; Mauranen, 2006; Pietikainen, 2018). Studies have shown that ELF interlocutors frequently engage in collaborative practices such as reformulation, repetition, and clarification to co-construct understanding (e.g., Cogo, 2009), highlighting the shared responsibility for meaning-making and the flexible use of multilingual resources (Cogo and House, 2018; Cogo and Pitzl, 2016).
Despite this shift toward an interactional perspective, most ELF research has focused on naturally occurring communication in informal, academic, or professional settings. Less attention has been paid to pedagogically structured task-based interaction, where learners must coordinate meaning while completing communicative tasks. As a result, how CSs are deployed as interactional resources in task-based contexts remains underexplored. Examining CS use in such settings can provide insight into how learners collaboratively manage meaning under pedagogically relevant conditions.
CS instruction and ELF-aware pedagogy
Interest in the pedagogical implications of CSs has grown in recent years. Research indicates that explicit CS instruction can enhance learners’ awareness of strategic options and support more effective communication (Lam, 2006; Milliner and Dimoski, 2024; Nakatani, 2005; Peker and Erdemir, 2021). At the same time, developments in ELF research have informed pedagogical approaches that emphasize communication in linguistically diverse environments.
ELF-aware pedagogy integrates insights from ELF communication into language teaching and teacher education (Bjorkman, 2013; McBride, 2021; Seidlhofer, 2011; Soruc and Griffiths, 2023). Rather than priortizing conformity to native-speaker norms, it emphasizes communicative clarity, adaptability, and the ability to interact across diverse linguistic contexts (Bayyurt and Sifakis, 2015; Sifakis and Bayyurt, 2024). This shift moves the focus from linguistic accuracy alone to how interlocutors establish and sustain understanding in interaction.
ELF awareness has been conceptualized as encompassing linguistic, attitudinal, and pedagogical dimensions (Sifakis, 2019), which can be operationalized as observable communicative orientations. Three dimensions are particularly relevant to the present study: intelligibility, adaptability, and multilingual–multicultural awareness. Intelligibility prioritizes clarity and shared understanding over native-like accuracy (Jenkins, 2015). Adaptability refers to learners’ ability to adjust linguistic and interactional resources in response to interlocutors’ needs (Cogo and House, 2018; Seidlhofer, 2011). Multilingual–multicultural awareness emphasizes sensitivity to linguistic and cultural diversity and the strategic use of shared resources to support communication (Baker, 2015; Sifakis, 2019).
Despite increasing interest in ELF-aware pedagogy, empirical research has rarely examined how CSs are employed in task-based interaction in relation to these communicative orientations. Existing studies have tended to focus on teacher beliefs or curricular design (e.g., Bayyurt and Sifakis, 2015; Galloway and Rose, 2018), while task-based research has typically evaluated learner performance in terms of accuracy, fluency, and complexity. Consequently, the relationship between CS use and ELF-aware communicative orientations in task-based interaction remains insufficiently documented.
The present study addresses this gap by examining how multilingual learners employ CSs during task-based interaction and how these strategies relate to the three ELF-aware communicative orientations: intelligibility, adaptability, and multilingual–multicultural awareness. By focusing on observable strategic behavior in interaction, the study provides an interactionally grounded understanding of CS use and offers insights relevant to task design and strategy instruction in multilingual language classrooms.
Research questions
To address these issues, this study investigates CS use in task-based ELF interaction. The study is guided by the following research questions:
How do multilingual learners employ CSs in task-based interaction? How do these strategies reflect ELF-aware communicative orientations, specifically intelligibility, adaptability, and multilingual–multicultural awareness?
Methodology
Participants
Ten undergraduate students (five L1 Chinese and five L1 Korean) from a medium-sized university in Seoul participated in the study. All were learning English as an additional language and had not previously resided in English-speaking countries. Their Test of English for International Communication (TOEIC) scores ranged from intermediate to advanced levels. 1 They were randomly paired across L1 backgrounds, with each dyad consisting of one Chinese and one Korean student. This cross-linguistic pairing was purposefully designed to simulate authentic ELF interaction, in which interlocutors draw on diverse linguistic repertoires rather than a shared L1. Although the sample size was relatively small, it allowed for a fine-grained qualitative analysis of interactional processes, consistent with previous ELF discourse studies that prioritize depth and detail over breadth (e.g., Cogo, 2009; Kaur, 2011). The study involved no instructional intervention and posed no foreseeable risk to participants. All participants were fully informed of the purpose and procedures of the study and provided voluntary written consent prior to participation. They were informed of their right to withdraw at any time without penalty. Data were anonymized during transcription and analysis.
Task design
A closed information-gap task, “spot-the-difference,” was employed to elicit spontaneous strategy use and negotiation sequences. Each pair received two similar images containing four to five subtle differences and was instructed to identify the differences through verbal communication, without showing their images to each other. The task design required meaning-focused communication under information-gap conditions, creating an environment conducive to negotiation and strategic behavior. A closed task was intentionally chosen to ensure comparability across dyads and to observe how participants employed CSs within a constrained communicative space. Such tasks also reveal learners’ ability to sustain intelligibility and adapt communicatively when opportunities for open expression are limited, both of which are key aspects of ELF-oriented strategic competence. Each session (lasting between 13 and 17 min) was conducted in English without pre-task rehearsal to maintain spontaneity. All interactions took place in a language laboratory and were video recorded for further analysis.
To gain insights into participants’ reflections on strategy use, stimulated recall sessions (Gass and Mackey, 2016) were conducted with each participant the following day. Immediately following each task session, the researcher reviewed the recordings and identified segments for stimulated recall based on three criteria: (a) instances where communication appeared effortful or disrupted, (b) moments of collaborative meaning construction, and (c) episodes where participants used non-verbal or multilingual resources. During each recall session, participants viewed each selected segment and were asked to describe what they were thinking at that moment, including their intentions, reasoning, and interpretation of the interaction. This session was conducted in participants’ L1 to ensure that they could fully express their thoughts. The recall data were analyzed alongside the corresponding interactional episodes to clarify each participant's strategic orientations and decision-making processes.
Data analysis
All interactional and stimulated recall data were transcribed verbatim and analyzed. The analysis adopted a qualitatively driven mixed-methods approach in which frequency counts were integrated with interactional analysis. Frequencies and percentages of CSs were calculated to provide an overview of distributional patterns. These numerical trends were not treated as indicators of statistical significance or as the basis for interpretative claims; rather, they served to contextualize the qualitative analysis and illustrate the relative prominence of strategies observed in the dataset. Interpretative claims were grounded primarily in turn-by-turn interactional analysis, supported by stimulated recall data clarifying participants’ strategic orientations. This approach captures both distributional tendencies and the interactional functions of strategies in task-based ELF interaction.
The analysis proceeded in two interrelated phases: (a) identification and categorization of CSs and (b) interpretative analysis to clarify how these CSs were oriented toward ELF-aware dimensions. In the first phase, CS identification was conducted to determine how interlocutors managed communicative difficulties during task-based interaction. To guide the initial coding, widely recognized CS categories in SLA and ELF research were consulted (e.g., Cogo and Pitzl, 2016; Dörnyei and Scott, 1997; Mauranen, 2006). These categories are summarized in Table 1.
Common CSs discussed in previous research.
Note. CSs: communication strategies.
During data analysis, the coding process remained open and iterative, allowing for the identification of additional strategies that were not fully represented in existing taxonomies. CSs were categorized based on how participants attempted to prevent, address, or repair communicative difficulties during interaction. The unit of analysis was a CS episode, defined as a sequence of turns in which a communicative difficulty emerged and was negotiated or resolved. Each episode could include more than one CS.
In the second phase, each identified CS episode was then examined to determine how it reflected the three ELF-aware orientations. Assignment of orientations to CS episodes was interpretive, based on the interactional function observed in context and triangulated with stimulated recall comments. A single episode could reflect one or multiple orientations simultaneously, depending on its interactional context. Table 2 summarizes the coding criteria used to operationalize each orientation.
ELF-aware orientations and coding criteria.
Note. ELF: English as a Lingua Franca.
Stimulated recall data were analyzed systematically for all identified CS episodes. For each episode, recall comments were coded for evidence of strategic awareness, communicative intention, and perceived interactional difficulty. These codes were then compared with the interactional evidence to strengthen interpretive validity.
Excerpt 1 illustrates how CSs were identified and interpreted in relation to ELF-aware communicative orientations:
B: And the second one is chased after the angry Snoopy. P: Chase? [Clarification request] B: On the right side. Chase after. So is being chased after the angry Snoopy. [Reformulation] P: Really? B: So Snoopy is chasing the bird. [Reformulation] P: No. Ah…yeah.
This episode was coded as involving reformulation and clarification requests (Table 1). The participants collaboratively reconstructed the intended meaning through successive turns, using lexical adjustment (“chased after” → “chasing”) and confirmation. Such exchanges represent classic examples of meaning negotiation, where interlocutors detect and repair misunderstanding through interactive effort.
In the second analytic phase, this episode was interpreted in relation to ELF-aware communicative orientations. From this perspective, the interaction can be understood as reflecting an orientation toward intelligibility (Table 2), as both interlocutors explicitly sought mutual understanding, and toward adaptability (Table 2), as B flexibly modified linguistic forms in response to P's cues (lines 3 and 5).
Excerpt 2 illustrates a case involving different CSs and ELF-aware dimensions:
L: How about their shoes? The left girl wear sneaker. And maybe his, his shoes, above his shoes maybe stripes K: Yeah. I don’t know. It is light. Its name is … but it is jjori [Korean, meaning flip-flops]? L: [nodding] K: And the right side...right side of the black-haired girl. Put on white shoes.
This exchange features translanguaging (Table 1), as K compensates for lexical gap by using the Korean term jjori.
In the second analytic phase, the episode was interpreted through the lens of ELF-aware dimensions. It reflects multilingual–multicultural awareness (Table 2), with K drawing on shared multilingual resources to maintain interaction, and adaptability, through modification of lexical choices. However, intelligibility was not achieved: according to stimulated recall data, L later reported not understanding jjori but had nodded to maintain conversational flow rather than seek clarification.
To ensure analytical credibility, 20% of the dataset was independently coded by a second researcher, yielding substantial interrater reliability (κ = .78). Coding discrepancies were discussed until consensus was reached. Stimulated recall data provided an additional layer of triangulation, linking participants’ thinking processes along with their observable communicative behavior.
Results
CSs
This section addresses the first research question concerning the types and distribution of CSs observed in the dataset. The results first present overall distributional tendencies of CS use, followed by qualitative analysis of selected episodes illustrating their interactional functions. All instances of CS were systematically identified and categorized according to their frequency. Table 3 summarizes the results.
CSs identified in the present study.
CSs that are less frequently discussed in previous research and therefore not included in Table 1.
Note. CSs: communication strategies.
As shown in Table 3, the most frequently occurring strategies were self-repair (26.6%), followed by co-creating messages (20.7%) and paraphrase/reformulation (16.7%). Strategies traditionally emphasized in SLA research, such as clarification request and confirmation check, occurred less frequently. Likewise, translanguaging and topic abandonment were relatively rare, each accounting for less than 3% of the total.
Three strategies—self-repair, co-creating messages, and deconstructing difficult information—marked in Table 3 and categorized as less prominently emphasized in previous research, were selected for detailed qualitative analysis. Although these strategies are not entirely absent from earlier literature, they have received comparatively limited attention relative to classic negotiation moves. In the present dataset, however, they collectively accounted for over half of all CSs (53.2%) and thus warrant closer examination.
This distribution suggests that, while learners continue to draw on well-documented CSs to negotiate meaning, they also engage extensively in strategies that promote collaborative construction and refinement of messages, indicating a dynamic, problem-solving orientation in ELF interactions. The following sections present illustrative excerpts to demonstrate how these strategies function in interaction.
The first focal strategy is co-creating messages. This strategy refers to instances in which interlocutors collaboratively construct meaning by jointly producing lexical or propositional content, rather than merely correcting or questioning an utterance. Excerpt 3 illustrates this process through lexical support:
A: One tree, and the boy … uh … [pause] B: Is hiding behind … A: Hide. Hide behind the tree.
In this exchange, A's hesitation signals difficulty in lexical retrieval. B offers a candidate verb (“hide”), which A immediately incorporates. This joint construction allows the utterance to be completed without disrupting the interactional flow.
Excerpt 4 provides another example of co-creating messages, though realized through a slightly different mechanism:
A: Three. The … the sandbox … sand what … sandbox? B: Sandbox. Ah yeah. A: I see one blue bucket. And red … red some … B: Something. Yeah.
In this exchange, A signals uncertainty through self-repair and incomplete phrasing. B first acknowledges A's lexical candidate and subsequently provides minimal collaborative support (“something”), thereby sustaining the interaction despite incomplete lexical formulation. Stimulated recall data further indicated that B recognized A's intended meaning even in the absence of full lexical accuracy.
In both excerpts illustrating co-creating messages, one interlocutor seeks assistance when encountering difficulty retrieving a lexical item. This need is signaled through rising intonation, pauses, or the use of incomplete phrases. The partner then contributes by collaboratively constructing the message. Notably, although the co-created language may not conform to target-like norms, it remains intelligible to both interlocutors. This shared understanding provides a foundation for further elaboration and facilitates the progression of the interaction.
Another prominent strategy identified in the dataset is deconstruction of difficult information. This strategy involves breaking down complex information into smaller, more manageable components, as shown in Excerpt 5:
A: Yes and there is a bell um … there is a bell … there is a bell on the dog's Snoopy's house. B: Um … A: Do you see? B: The dog house? A: Yes. B: There is some bell? A: Yes. B: What is the color of the bell? A: Yellow. B: Where is it? A: Um … Upper side.
In Excerpt 5, A's repeated pauses and self-repairs indicate production difficulty. B's subsequent questions progressively unpack the message (e.g., “dog house?” “what color?” “where?”), allowing interlocutors to jointly reconstruct meaning. B reported in the stimulated recall that she could extract only partial information from the first sentence, so she checked it piece by piece. Through this deconstructing process, complex content is redistributed across turns, allowing the speakers to simplify syntactic demands while achieving mutual understanding.
The final CS that warrants detailed analysis is self-repair. Self-repair enables interlocutors to monitor and modify their own speech without external prompts, thereby maintaining fluency while pre-empting potential communication breakdowns. Excerpt 6 illustrates a self-repair occurring within a single speaking turn:
C: Oh no. Pink shirts. Pink … Yeah yeah shirts. I think it's shirts. Because she has a [kʌlɚ]. [kʌlɚ]. [kʌlʌ]. [Pause]. [Puts hands on his neck.] L: Yes. In my picture she is carrying some … tray.
In this exchange, C repeatedly adjusts his utterances while searching for lexical resources, attempting to confirm the intended meaning as he explains the “shirts.” The iterative self-repairs involving the pronunciation of “kʌlɚ” as “kʌlʌ” illustrate the strategy C employed to refine his phonological production, particularly when dealing with similar-sounding words that caused confusion.
Excerpt 7 illustrates another instance where self-repairs occur across multiple turns:
A: Yeah, one is in the sky. One is in the chair. B: Chair yes. A: Long chair. B: Yeah, long chair. A: Bench. Bench. B: Yes, yes. A: And on the top of the, there are two birds on the bench.
Two self-repairs occur in Excerpt 7. The first involves a reformulation from “chair” to “long chair,” and the second further refines “long chair” to “bench.” Although these repairs unfold across multiple turns, B only provides minimal response without disrupting interaction. This example shows how self-repair can extend across turns without disrupting interactional progress. In the stimulated recall session, B indicated an inclination to refine his word choice to enhance clarity, and A demonstrated comprehension of all the words.
Overall, the findings indicate that ELF interlocutors in task-based interaction rely not only on reactive negotiation moves but also on proactive and co-constructive strategies. These practices create a flexible and collaborative space for meaning-making that sustains interactional flow.
CSs and ELF-aware orientations
This section reports on how the CSs identified in RQ1 were interpreted in relation to three ELF-aware communicative orientations: intelligibility, adaptability, and multilingual–multicultural awareness (Bayyurt and Sifakis, 2015; Jenkins, 2015; Sifakis, 2019). These orientations were assigned based on the interactional function of each CS episode, supported by stimulated recall data that clarified participants’ strategic intentions. Table 4 summarizes how each CS aligns with the three communicative orientations.
Realization of ELF-aware orientations across CSs.
Note. ELF: English as a Lingua Franca; CSs: communication strategies.
Table 4 shows the distribution of CSs across the three ELF-aware communicative orientations, with adaptability being the most prominent (61.7%), followed by intelligibility (57.2%) and multilingual–multicultural awareness (38.7%). This distribution suggests that learners not only maintained understanding but also actively modified their linguistic production in response to interactional demands.
Notably, self-repair was interpreted as reflecting adaptability in all observed instances, suggesting a consistent pattern that speakers frequently adjusted their linguistic output through self-monitoring and modification, rather than relying solely on interlocutor-initiated negotiation. Adaptability was also evident in paraphrasing (94.6%) and deconstructing difficult information (53.9%), indicating that speakers employed multiple adjustment strategies to maintain interactional continuity.
Intelligibility was most strongly reflected in collaborative strategies such as co-creating messages (95.7%) and deconstructing difficult information (84.6%). These strategies involved joint clarification and reformulation across turns, supporting mutual understanding through shared interactional effort. In contrast, more traditional negotiation moves such as clarification requests (19.4%) and confirmation checks (45.5%) showed weaker alignment with intelligibility, suggesting that signaling non-understanding alone did not always lead to sustained collaborative resolution.
Multilingual–multicultural awareness was concentrated primarily in interactionally collaborative strategies, particularly co-creating messages (87%) and translanguaging (100%), while appearing far less frequently in self-initiated strategies such as self-repair (16.9%). This distribution suggests that multilingual orientation emerged mainly through joint interactional work, where participants drew on shared linguistic resources to sustain communication.
Taken together, the results indicate that ELF-aware communicative orientations were realized through distinct yet complementary strategic pathways. Adaptability was primarily realized through speakers’ self-initiated monitoring and adjustment, intelligibility was maintained through collaborative clarification and reformulation, and multilingual awareness through the shared mobilization of multilingual resources during interaction.
Discussion
Strategic patterns in task-based ELF interaction
The findings indicate that the CSs employed differ notably from those most frequently reported in earlier ELF research. Previous studies have typically emphasized strategies such as clarification requests, confirmation checks, and paraphrasing (e.g., Cogo and House, 2018; Mauranen, 2006; Pietikainen, 2018), largely based on analyses of naturally occurring conversation in informal, academic, or professional conversational settings. These contexts are typically open-ended and less constrained by task-specific requirements. In contrast, the present data demonstrated a broader and more interactionally embedded repertoire, including adaptive self-repair, co-creating messages, and deconstructing difficult information. Rather than relying primarily on reactive repair following overt breakdowns, learners engaged in ongoing, proactive management of meaning. These patterns indicate that participants actively monitored their output, making real-time adjustments to maintain interactional alignment and pre-empt potential breakdowns.
This difference can be related to features of the task design. The information-gap task required both participants to achieve a shared outcome through mutual exchange of visual information. The task's symmetrical structure positioned both interlocutors as equally responsible for maintaining communication. Within this interactional environment, strategies such as co-creating messages and deconstructing difficult information enabled participants to collaboratively refine meaning through lexical prompting, partial formulations, and incremental confirmation. These interactional moves facilitated communicative progress even when precise linguistic forms were not immediately available, highlighting the role of collaborative adjustment in sustaining task completion (Excerpt 4).
The findings also provide insight into the function of self-repair in task-based ELF interaction. In earlier SLA-oriented work, self-repair has often been discussed either as a marker of disfluency (Buckwalter, 2001) or as evidence of monitoring and noticing (Kormos, 2008). In the present data, however, self-repair functioned primarily as an adaptive resource for maintaining alignment between intended meaning and linguistic expression. Immediate repairs within a single turn (Excerpt 6) reflected speakers’ efforts to improve form–meaning alignment, while delayed repairs across turns (Excerpt 7) demonstrated how shared task context supported autonomous reformulation. Unlike listener-initiated repairs reported in earlier studies (e.g., Wong, 2000), these repairs were predominantly self-initiated and strategically deployed. Stimulated recall data reinforced this interpretation, as participants reported “intentionally changing words to make them clearer.”
The three dominant strategies identified in this study reflect varying degrees of interactional agency, ranging from individual monitoring to collaborative scaffolding. Self-repair exemplifies speakers’ capacity for autonomous adaptation, whereas co-creating messages and deconstructing difficult information highlight shared responsibility for maintaining meaning. These findings suggest that CSs in task-based ELF interaction function not only as responses to breakdown but also as proactive resources for maintaining communicative alignment.
Interactional implications for ELF-aware communication
The findings also revealed variation in how different CSs were associated with ELF-aware communicative orientations. Intelligibility was most strongly associated with collaborative strategies such as co-creating messages and deconstructing difficult information. These strategies enabled interlocutors to jointly refine meaning across multiple turns, illustrating intelligibility as an interactional process achieved through ongoing clarification and adjustment. This practice aligns with ELF perspectives that treat mutual understanding as a shared responsibility rather than an individual achievement (Jenkins, 2015; Mauranen, 2006).
By contrast, more traditional negotiation strategies such as clarification requests and confirmation checks, widely discussed in SLA research (Long, 1996), showed weaker alignment with intelligibility. While these strategies signaled nonunderstanding, they did not consistently lead to extended interactional sequences that supported meaning elaboration (Hu and Park, 2023). This pattern highlights a key distinction between reactive negotiation moves that signal comprehension problems and proactive, collaborative strategies that sustain meaning construction. In this task-based context, intelligibility was therefore supported more strongly by joint elaboration than by breakdown-repair sequences alone.
Adaptability emerged as the most prominent orientation overall and was most consistently associated with self-repair. This finding suggests that speakers frequently adjusted their linguistic output through ongoing self-initiated monitoring during interaction. Such adjustments included simplifying expressions, restructuring utterances, and paraphrasing difficult information in response to perceived interactional demands. This pattern supports Seidlhofer’s (2011) view of ELF communication as a process of continuous adjustment. Importantly, stimulated recall reports corroborated this interpretation: participants consciously simplified, rephrased, or restructured their speech “to make sure we are on the same page.” These findings indicate that adaptability operates at both the level of observable interactional behavior and the level of underlying strategic awareness.
Multilingual–multicultural awareness was less frequent overall but was particularly evident in co-creating messages and translanguaging. In these instances, participants drew on shared multilingual resources to sustain interaction and rapport, suggesting that multilingual awareness emerged primarily through joint interactional work. Consistent with ELF research (e.g., Baker, 2015; Cogo, 2009), translanguaging functioned as a strategic communicative resource that supported interactional continuity rather than signaling linguistic deficiency.
The current findings highlight the interactional nature of strategic competence in ELF communication. Rather than functioning solely as compensatory responses to linguistic limitation, CSs operated as flexible resources that enable learners to maintain intelligibility, adapt to interactional demands, and mobilize multilingual repertoires during task-based interaction.
From a broader perspective, these findings extend current research on ELF and task-based interaction by demonstrating how strategic behavior is systematically distributed across distinct communicative orientations in pedagogically structured interaction. By showing that different strategies support different dimensions of ELF-aware communication, this study offers empirical evidence that strategic competence involves both individual adaptive processes and collaborative meaning construction within task-based interaction.
Conclusion
This study examined how ELF interlocutors employ CSs during task-based interaction and how these strategies reflect ELF-aware communicative dimensions. The findings reveal a systematic strategic profile characterized by collaborative and adaptive meaning management. Strategies such as co-creating messages
The analysis further showed that CSs differed in how they aligned with ELF-aware communicative orientations. Strategies involving collaborative elaboration and reformulation were closely associated with intelligibility and adaptability, while multilingual resource use supported multilingual–multicultural awareness. These patterns demonstrate that the strategic significance of CSs is not determined solely by frequency, but the extent to which they facilitate alignment, adaptation, and mutual understanding during interaction.
This study contributes to ELF-informed SLA research by suggesting that, unlike prior ELF studies primarily based on naturally occurring interaction, CSs in task-based contexts extend beyond compensatory functions, serving as central mechanisms for sustaining alignment and coordinating meaning in task-based interaction. Task-based interaction provides a productive context for developing this competence, as it creates communicative conditions that require learners to sustain intelligibility, adapt to interlocutors, and mobilize available communicative resources. While the findings are grounded in an information-gap task, they point to broader implications for how structured interactional conditions can elicit diverse strategic behavior in classroom-based language learning. By showing how CSs function as interactional resources supporting ELF-aware orientations, this study offers support for integrating ELF-aware perspectives into task-based language pedagogy and underscores the value of strategy-oriented approaches to communicative development.
Pedagogical implication
The findings suggest several ways in which CS instruction can be integrated into task-based classroom practice from an ELF-aware perspective.
First, task design plays a critical role in promoting strategic and collaborative communication. The information-gap task examined in this study required learners to exchange complementary information and jointly resolve communicative challenges. These task conditions created opportunities for co-creating messages and deconstructing difficult information, which supported intelligibility and collaborative meaning construction. Pedagogically, this suggests that tasks requiring mutual dependency and incremental information exchange (e.g., spot-the-difference, jigsaw tasks) may be particularly supportive for developing collaborative strategic competence.
Second, self-repair should be recognized as a strategic resource rather than treated primarily as a sign of disfluency. In the present study, self-repair supported adaptability by enabling learners to monitor and adjust their utterances in real time without disrupting interaction. Pedagogical practices that tolerate reformulation, hesitation, and revision may help learners develop greater flexibility and confidence in maintaining intelligibility during interaction.
Third, multilingual resources can support communication and reflect multilingual–multicultural awareness. Instances of translanguaging demonstrated how shared linguistic repertoires can sustain interaction and reflect multilingual–multicultural awareness. Pedagogically, activities that encourage learners to draw on their full repertoires, such as multilingual planning, strategic L1 use for problem-solving or cross-linguistic comparison, may foster sensitivity to linguistic diversity while strengthening learners’ ability to manage communication effectively.
Finally, CS instruction can be integrated into task-based teaching by explicitly drawing learners’ attention to how particular strategies support intelligibility, adaptability, and collaborative meaning-making. Rather than focusing exclusively on linguistic accuracy, instruction can foreground strategic competence as the ability to monitor understanding, adjust language, and work collaboratively to achieve communicative goals. This approach aligns with the communicative realities of multilingual interaction and supports learners’ development as effective and adaptable users of English across diverse contexts.
Several limitations of this study should be acknowledged. First, the dataset involved a small number of participants engaged in a single information-gap task in one instructional context, which limits the transferability of the findings to other settings and task types. Second, the participants shared relatively similar educational backgrounds; broader linguistic, cultural, and proficiency diversity may yield different patterns of strategy use. Future research could extend this line of inquiry through longitudinal designs, multiple task types, and more diverse learner populations, in order to further examine how ELF-oriented strategic competence emerges over time and how it is shaped by task demands, interlocutor configurations, and institutional contexts.
Footnotes
Funding
The authors received no financial support for the research, authorship, and/or publication of this article.
Declaration of conflicting interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.