I. INTRODUCTION
New understandings of multiliteracy, which go beyond the traditional print medium, are critical as literacy and technology continually shape one another (Benjamin, 2014). Multimodal technology fosters multiliteracy practices in language learning environments that allow English learners to explore various modes for conveying meaning. This shift plays a pivotal role as multimodality is increasingly regarded as an essential component of ELT training, pedagogy, and education policy (Grapin & Llosa, 2020). Despite the consistent prevalence of semiotics as an analytical tool in media and substantial research on disciplinary meaning-making, the effectiveness of multimodality—both linguistic and nonlinguistic modes—for EFL writers remains inconclusive (Yi et al., 2019). Grapin (2019) emphasized that both linguistic and nonlinguistic modes are equally valued to effectively cultivate learners’ communicative competence. However, Qu (2017) argued that nonlinguistic modes are often seen as merely supporting linguistic instruction. For example, while gestural communication conveys meaning through bodily movement, linguistic meaning relies on grammar, sentence structure, metaphors, and word choice (New London Group, 1996). In contrast, another study confirmed that nonlinguistic modes more commonly support language learning (Grapin, 2019). In particular, silent movie clips can enhance learners’ writing performance by stimulating motivation through visual images and encouraging creative sentence construction (Kasper & Singer, 2001). Similarly, Kim (2014) found that nonlinguistic modes, such as images and silent movie clips, may lead to varying learning outcomes or interpretations depending on how they are employed. Static images engage learners in cognitively demanding narrative construction and temporal sequencing, fostering linguistic elaboration, while silent animations provide implicit temporal cues that ease processing and facilitate more immediate meaning-making (Mayer, 2001; Sadoski & Paivio, 2001). Nonetheless, empirical research directly comparing these two modalities within narrative writing contexts remains limited. Specifically, narrative tasks have been shown to facilitate the development of linguistic competence by encouraging learners to consciously focus on grammatical structures, lexical choices, and overall language form (Verdugo & Belmonte, 2007). When mediated through multimodal input, such as images or animations, these tasks can further support academic writing by stimulating metalinguistic reflection. Given that each mode entails its own semiotic potentials and constraints (Kress, 2009), evaluating their effectiveness across contexts is inherently complex. Thus, further research is needed to identify the relative impact of media on meaning-making and to determine its most effective application in writing classrooms. To investigate the effects of linguistic and nonlinguistic input on narrative writing and cognitive engagement, this study utilized three instructional materials frequently employed in EFL contexts: an animated video with sound representing a linguistic mode and two nonlinguistic modes—an animated video without sound and a digitized picture book without text.
Furthermore, considering that multimodality leaves room for negotiating meaning, a writer’s creative expression can be profoundly shaped by their identities, contexts, and purposes (Curwood, 2012) in collaborative narrative writing. Since writing is inherently a social activity facilitated by technology (Cheung, 2022), the role of collaborative writing in co-constructing knowledge through multimodal resources has become important. However, considerable research has examined the merits of collaborative writing rather than focusing on meaning negotiation. Such investigations focused on enhancing motivation (Liu & Lan, 2016), facilitating language negotiation and mutual scaffolding (Akoto, 2021; Li & Zhu, 2011; Storch, 2019), offering affordances for using technology in the writing process (Abrams, 2019; Jiang, 2018; Smith, 2019; Strobl, 2014), and fostering both metacognitive effectiveness and writer autonomy (Azkarai & Kopinska, 2020; Durksen et al., 2017; Villarreal & Lázaro-Ibarrola, 2022; Wang et al., 2024). Additionally, peer interaction has been shown to help address linguistic gaps and promote self-reflection (Caux & Pretorius, 2024; Jin et al., 2022). However, how individual writing skills are shaped through collaboration using Google Docs to engage with various media remains under-researched. Given that one of the key benefits of collaborative writing lies in the opportunity to model peers’ cognitive and metacognitive strategies (Teng, 2021), it is crucial to investigate how the integration of different types of media can function as a scaffold for enhancing learners’ writing development. This study investigates how learners negotiate meaning during collaborative writing, with a focus on their perceptions and satisfaction of three different media modes, their collaborative experiences, and their individual writing outcomes using Google Docs. Accordingly, this study addresses the following three research questions.
1. What are participants’ perspectives on the use of the three media modes in collaborative writing?
2. To what extent are participants satisfied with the use of media modes in collaborative writing?
3. Does collaborative writing using media modes enhance participants’ individual writing performance?
II. PEDAGOGICAL FRAMEWORK
This research employs Liang and Lim’s (2020) pedagogical framework, which consists of three domains—critical, creative, and technical—in the context of video production. The creative domain encourages English learners to generate ideas with empathy and experimentation based on the principles of the Stanford Design Thinking Model (Meinel & Leifer, 2020). The technical domain enables students to turn their ideas into digital projects by providing the production and editing skills needed for video creation. Finally, the critical domain helps learners make thoughtful semiotic choices through a metalanguage, allowing them to reflect on how various modes—such as visuals, sound, and text—combine to convey meaning in their final performance. Likewise, this study adopted the three domains to examine their pedagogical impact (Figure 1). The critical domain was explored by applying metalanguage in the analysis of three modes. The creative domain was addressed through collaborative composition, and the technical domain was implemented by supporting peer interaction using Google Docs for individual writing.
1. Metalanguage Through Multimodality
The critical domain helps learners understand how three different modes—images, animation with sound, and animation without sound—create meaning through a shared metalanguage that articulates the function and effect of these semiotic choices (Liang & Lim, 2020; Unsworth & Mills, 2020). Multimodal composing allows L2 learners to enhance their English language proficiency and develop metalinguistic awareness by engaging in authentic writing tasks, as the incorporation of various modes helps develop language skills and enables them to interpret multiple modes at the same time (Kim, 2016). This pedagogy embraces multimodal meaning-making that can help learners develop language proficiency by expressing their own meaning (Nelson, 2008). Specifically, semiotic knowledge across multiple modes enhances writers’ authorial agency. In the process of recontextualization, this agency transforms rhetorical structures through the use of semiotic resources (Barton & Potts, 2013; Cimasko & Shin, 2017; Jiang, 2017). Thus, it enhances language skills by guiding learners to integrate language use with media that can enhance their writing. Building on this foundation, this study highlights the importance of understanding how language learners interpret each mode, both linguistic and non-linguistic, as a medium to create meaning for their writing proficiency.
2. Collaborative Multimodal Composing
Collaborative learning fosters reflective thinking and supports language learners in constructing and negotiating meaning through interaction (Zhang et al., 2024). Similarly, situated within sociocultural theory (Vygotsky, 1978), multimodal collaborative writing enhances the expression of complex ideas, improves language use, and enriches content during the writing process, where interpreting semiotic modes is crucial. The incorporation of media resources not only provides scaffolding for writers by enhancing their content knowledge and engagement with writing, but also significantly boosts their motivation (Chen, 2021; Dobao, 2012). As such, this scaffolding process plays a crucial role in alleviating writers’ difficulties and reinforcing their motivation. Multimodal collaborative writing has underscored the importance of fostering more engaging and innovative writing beyond traditional print-based methods to construct and communicate meaning within sociocultural and multiliteracies frameworks (Allagu, 2022). As it involves thinking, acting, and communicating to engage in group learning and to participate in literacy across diverse social contexts through various media (Miller, 2015), exploring which mode can develop greater writing performance is regarded as pivotal in collaborative writing. Thus, this study emphasizes writers’ experiences with peer interactions to provide a deeper understanding of how linguistic and nonlinguistic modes can be negotiated based on their perceptions and experiences for writing development.
3. Google Docs as a Digital Tool
Utilizing Google Docs offers a significant affordance that enables interaction and communication through peer editing and peer review (Alharbi, 2020; Hoang & Hoang, 2022) as a collaborative writing medium. Throughout the process, Google Docs provides scaffolding that helps writers support each other’s development. Notably, Google Docs provides a crucial avenue that enables online interactions both synchronously and asynchronously (Abrams, 2016), therefore allowing language learners to actively engage in collaborative writing inside and outside the classroom.
This indicates that Google Docs can effectively expand the learning space beyond classroom boundaries, thereby helping learners develop autonomy as they write collaboratively on shared documents and contribute to the structure and content of their writing (Elola & Oskoz, 2010). Another notable feature is that it enables the sharing of semiotic modes within a collective digital environment. Additionally, as a Word-based platform, it allows learners to attach images or generate links, and from the perspective of L2 writers, it provides functions that assist in identifying simple typos or errors. With the various affordances of Google Docs, this study aims to examine individual writing production in order to capture the potential impact of collaborative writing on Google Docs as a space for meaning negotiation through semiotic media resources.
4. Individual Performances in Collaborative Writing
From a pedagogical perspective, writing is viewed as a process rather than a final product, implying that learners require various scaffolding activities (Hoang & Hoang, 2022). During collaboration, it is necessary to identify how different semiotic media resources may affect individual writing outcomes as learners engage in co-constructing knowledge. Many previous studies have acknowledged that collaborative writing skills can be transferred to individual writing proficiency. For example, web-based collaborative writing significantly enhanced L2 learners’ individual writing performance since they perceived the collaborative experience as beneficial for their writing development (Bikowski & Vithanage, 2016). Similarly, Jafari and Ansari (2012) reported that Iranian EFL learners engaged in collaborative writing outperformed their peers in grammatical accuracy, suggesting that interaction supports form-focused language development. Jiang and Eslami (2021) also found that Chinese EFL learners who participated in computer-mediated collaborative writing using Google Docs demonstrated significant improvements in individual writing performance, with fluency gains among lower proficiency learners and enhanced accuracy among higher proficiency learners. These studies provide insights into how collaborative writing may enhance individual learning, yet only overall writing was evaluated as a reflection of writing quality. Similarly, collaboration was further amplified by peers, but doubts remain regarding the linguistic accuracy of peer feedback (Abdelhalim, 2024). Although prior studies have validated the positive effects of collaborative writing, learners who lack confidence in English writing may not experience the same benefits. Furthermore, positive learning outcomes are difficult to achieve when learners face conflicting opinions or when topics are insufficiently engaging (Felipeto, 2019). Accordingly, this study investigates how learners engage in meaning-making through various media interpretations in computer-mediated collaborative writing and how this process influences the development of their individual writing performance.
III. METHOD
1. Participants
Convenience sampling was used for the participants who attended the first-year college composition courses for Korean EFL learners. All 90 freshmen (53 males; 37 females) from various majors such as Humanities, Electrical Engineering, Education, Public Administration, and Architecture participated. The class met twice a week for 90 minutes each time. Their first language is Korean, and their English proficiency levels, determined by their TOEIC scores prior to the course, ranged from A2 to B2 according to the Common European Framework of Reference for Languages (CEFR) (A2: n = 28; B1: n = 43; B2: n = 19). Almost half of them (n = 44) lacked prior writing experience, and the group that used images had significantly less writing experience. Only six had writing experience, compared to the other two groups. The consent forms were obtained when gathering their responses on Google Forms excluding personal identification.
2. Research Procedures
The researcher, as an instructor, provided consistent guidance to all participants using the same textbook during the semester. Although the narrative tasks were delivered through different media modes, all participants adhered to uniform academic writing guidelines to ensure consistency in task expectations. Upon assessing their English writing proficiency at the beginning of the semester, participants wrote a short narrative story “A Self-Introduction Over Time” on the pretest in Week 2 during class. A background questionnaire was also designed to obtain information on the participants’ English proficiency via Google Forms in Week 2 (see Appendix A). The instructor randomly and evenly assigned five participants to six different groups based on their English proficiency and mode-preference survey. Two groups were assigned to each of the three modes (Figure 2). The linguistic mode involved an animation with audio, including dialogues and sound effects (Group A). The two non-linguistic modes comprised an animation without audio (Group B) and a series of images without written text (Group C).
For the animation with sound and conversation, The Lorax (Pups na chas, 2023), a six-minute video based on Dr. Seuss’s book, was selected as a lingustic mode. Shape up with Shaun (Shaun the Sheep Official, 2015) was utilized for the animation without sound as a non-linguistic mode. The researcher extracted sounds from the animation and used it as a movie clip without sound, also about six minutes long. Hiking (Oswald, 2021), the 40-page digitized picture book, was employed for the image group as another non-linguistic mode. In each class the instructor provided explicit guidance on how to organize ideas logically to write narrative essays, use a range of academic vocabulary, and apply diverse and accurate grammatical structures in their writing based on the IELTS Writing Task 2 criteria, which were also used to assess the participants’ pretest and posttest. The instructor created six Google Docs links with embedded media for each group and posted them on the class website for collaborative writing. Additionally, considering that this process can be effectively collaborative when roles are clearly defined among group members (Abdelhalim, 2024), Google Sheets, created by the instructor, was shared with the participants to document their individual contributions, ensuring accountability for their writing activities after each class such as leader or scheduler. The participants also used the KakaoTalk group chat, a South Korean messaging app, to facilitate idea exchange and write essays outside of class hours. In Week 3, the participants first constructed the story on worksheets with their groups by drawing images and writing plots (Figure 3) and then began writing an essay on Google Docs in the classroom.
In Weeks 4-7 and 9-10, excluding Week 8 reserved for midterms, participants wrote an essay with their group, establishing narrative structures, character development, and crafting the stories’ introductory, main, and concluding sections. Afterwards, participants were paired with another group that had used the same media mode to exchange feedback on narrative structure, vocabulary, grammar, and syntax. The peer review process was guided by a rubric for IELTS Task 2 provided by the instructor 1. Each group was required to collaboratively produce written feedback based on their discussions on Google Docs. In Weeks 10-11, participants revised their drafts using the feedback and submitted their final versions in Week 12. The second survey was also carried out (see Appendix B) via Google Forms to assess their satisfaction with each of the three modes, group work, and peer feedback. To assess the impact of digital collaborative writing on individual outcomes, a posttest was administered during Week 13. Participants completed an in-class individual writing task on the topic “The Most Memorable Experience in My Life.” Additionally, voluntary individual interviews (see Appendix C) were conducted during the same week.
3. Data Collection and Analysis
The data were collected throughout two semesters. To assess the participants’ individual writing performance, three native English-speaking educators who earned master’s degrees in English language education and linguistics and have taught English for over five years analyzed their two writing tasks using the IELTS Task 2 rubric. Their inter-rater reliability was .913 for a movie clip with the sound group, .886 for a movie clip without sound group, and .898 for the image group using Cronbach’s α. Microsoft Excel and the SPSS statistical program were employed to quantitatively analyze the two writing samples and two survey datasets. To assess the normality of the pretest and posttest scores across the three groups, the Shapiro-Wilk test was conducted. Homogeneity of variances was also examined using Levene’s test. A one-way ANOVA was employed to determine whether there were statistically significant differences in mean scores among the groups. For within-group comparisons between pretest and posttest scores, paired samples t-tests were performed. Furthermore, a Principal Component Analysis (PCA) was conducted to explore the underlying structure of the four writing assessment criteria. The suitability of the data for PCA was verified through the Kaiser-Meyer-Olkin (KMO) measure of sampling adequacy and Bartlett’s test of sphericity.
Participants’ responses regarding their perceptions and satisfaction were analyzed through Python coding, using word frequency and sentiment analysis based on natural language processing in Google Colab. As co-occurrence frequency lists often highlight significant keywords (Gries, 2021), word frequency analysis was used to capture participants’ perspectives. For word frequency analysis, the KoNLPy library with the Okt tokenizer was employed to extract nouns. Sentiment analysis was also employed to analyze participants’ satisfaction, which was then compared with the results from the Likert scale. To do this, the Copycats model was downloaded from Hugging Face, and it demonstrated an accuracy of 0.93 when benchmarked against the NSMC (Naver Sentiment Movie Corpus) dataset.
4. Results
1) Perspectives on the Three Media Modes in Collaborative Writing
Word frequency analysis was employed based on participants’ responses (Table 1). The movie clip with sound mode showed a recurring use of “word (6),” “sentence (5),” “expression (4),” “dialogue (3),” “listening (2),” “vocabulary (1),” and “grammar (1).” The participants felt using a movie clip with sound had positive and negative aspects since it included dialogue. Some replied that understanding the movie clip was difficult due to their lack of linguistic knowledge. On the other hand, some showed positive effects, such as using grammar and vocabulary words from the clip to write a story based on their interpretation. In the personal interview transcription, interviewees indicated that the use of the sound movie clip elicited more positive responses among those with higher proficiency levels.
Participant 1 (English level A2): Understanding the movie clip content was important to influence my writing, but it was challenging to listen to and interpret the sentences in the movie clip.
Participant 2 (English level B2): After watching the audiovisual video, I realized that words I already knew were used in various ways depending on the context, and it was helpful to apply those words in my writing.
However, the participants commented that the storyline inhibited them from creating a new story, and if someone already knew the plot, it constricted their creativity, making it difficult to generate innovative ideas. Some also perceived that watching a movie clip required too much time to recall the story.
Participant 3: One of the members was already familiar with the movie clip’s content. This pre-existing knowledge posed certain limitations in developing an original storyline.
Participant 4: I find that there is little room to invent an original story.
Additionally, as for the movie clip without sound mode, the words “group (8),” “feedback (7),” “imagination (7),” “grammar (4),” “collaboration (3),” and “interest (1)” appear frequently. The participants considered the absence of sound to facilitate enhanced imaginative thinking and gave them greater freedom in writing to develop their own expressions. This, in turn, enabled them to infer actions and improve their writing proficiency. Unlike the group that watched the movie clip with sound, they did not show a strong negative attitude toward creating a story. Overall, participants reported high satisfaction with this method, adding that it fostered more creative engagement and a deeper connection with the movie clip. Their interview answers are below.
Participant 5: I liked being able to express my own ideas in my own words or grammar by watching the movie clip without sound.
Participant 6: Watching a video without sound allowed me to freely express my imagination without restrictions while working with my group members.
Lastly, participants found that utilizing images, a picture book, to create a story enhanced their writing and creativity while some struggled with the content that was too simplistic. On the other hand, the use of still images also promoted independent thinking and helped strengthen sentence construction through associative thinking. In the word frequency, the series of images resulted in a greater use of “situation (5),” “word (2),” “sentence (2),” “association (1),” and “creativity (1).” Images allow for more flexibility in narrative construction, as they provide space for creative input between them, unlike a movie clip. The participants expressed both positive and negative feelings towards the use of images.
Participant 7: It was difficult to choose the appropriate words and the story in the series of images book was too simple to create a narrative.
Participant 8: Using images allowed for more creative and diverse thoughts compared to other media, which greatly helped with writing since it allowed me to think and organize the content on my own and did not require me to retain much information about the content.
2) Satisfaction With Collaborative Writing Using Media Modes
Participants’ overall responses to writing across the three modes were relatively comparable. In terms of satisfaction regarding the positive influence on their writing development, movie clip with sound ranked the highest, followed by a movie clip without sound, and finally, images. On the other hand, participants felt that the use of a movie clip without sound, a movie with sound, and images contributed to greater satisfaction in collaborative writing using media. The sentiment analysis results, similar to the participants’ responses on the Likert scale, demonstrated parallel outcomes (Figure 4).
Similarly, Figure 5 illustrates the correlations between each mode and participants’ satisfaction in terms of writing improvement, English proficiency, collaboration, and peer feedback. Notably, Group A reported the highest satisfaction with writing improvement, which may be attributed to the positive feelings elicited by linguistic input. However, satisfaction related to collaboration and peer feedback was lower compared to the other two groups. Groups B and C revealed a similar pattern, with the highest satisfaction with collaboration, followed by writing improvement, peer feedback, and English proficiency. Participants responded that using a movie clip without sound did not provide a predefined storyline, which required collaborative efforts to negotiate meaning and create a story by determining appropriate contexts together. Conversely, participants who used a movie clip with sound reported that they often divided the given plot into segments and assigned them to each individual. Group C showed the lowest overall satisfaction. Although their collaboration received the highest score in the correlation results, participants felt that the storyline conveyed through images was too simple to create meaningful content, which led to relatively lower satisfaction with writing improvement and peer feedback compared to the other groups. Participants’ overall reaction toward collaborative writing was that the experience provided opportunities to practice English and observe different writing styles. Nevertheless, participants found it challenging due to a lack of familiarity with group members and felt it did not significantly improve their writing skills.
However, more active participation was encouraged to avoid inconveniencing others, leading to the acquisition of new expressions and vocabulary, particularly during group editing and brainstorming. Some participants encountered difficulties in collaboration when group members were absent from class, requiring them to work outside of class time. Thus, one group leader suggested that each group member determine their own writing contribution to encourage more active participation rather than assigning roles in the group. Regarding the use of Google Docs, most participants considered it an effective tool for collaborative writing, as it provided access to a word-processing environment and enabled media attachment and link sharing, although concerns about the security of online collaboration were viewed as a limitation.
3) Individual Differences in Collaborative Writing Using Media Modes
The results of the Shapiro-Wilk test indicated that all three groups met the assumption of normality on the pretest: Group A (W = 0.972, df = 30, p = .602), Group B (W = 0.976, df = 30, p = .704), and Group C (W = 0.971, df = 30, p = .574). Levene’s test confirmed homogeneity of variances across the groups, F(2, 87) = 0.537, p = .586. A one-way ANOVA showed no statistically significant differences in pretest scores among the three groups, F(2, 87) = 2.179, df = 30, p = .119, η² = 0.066. For the posttest scores, the normality assumption was violated for Group B (W = 0.850, df = 30, p < 0.001), while it was satisfied for Group A (W = 0.932, df = 30, p = .057) and Group C (W = 0.970, df = 30, p = .526). However, Levene’s test confirmed the assumption of homogeneity of variances, F(2, 87) = 1.601, p = .208.
The data demonstrated adequate suitability for factor analysis, as indicated by a KMO value of 0.799 and a statistically significant result from Bartlett’s test of sphericity, χ²(6) = 290.660, p < .001. Based on the Kaiser criterion (eigenvalue > 1.0), a single principal component was extracted (Task response, Lexical resources, Grammatical range and accuracy, and Coherence and cohesion). A one-way ANOVA revealed no statistically significant differences in posttest scores among the groups, F(2, 87) = 0.595, p = .554, η² = 0.078.
According to the t-test results from the three raters, all three groups demonstrated improvements in individual writing skills from the pretest to the posttest (Table 2). The group working with videos with sound showed significant differences in four areas: task performance, coherence and cohesion, lexical resources, and grammatical range and accuracy. The group using videos without sound also exhibited significant differences, excluding task response. The group using images showed significant improvements in task performance and grammatical range and accuracy. However, no statistically significant improvement was found in task response for the group using a movie clip without sound, nor in coherence and cohesion or lexical resources for the image group. Regarding overall scores, calculated by averaging the four domains, the group with the movie with sound (t(29) = 3.37, p < .01, the group with the movie without sound (t(29) = 2.44, p < .05), and the group using images (t(29) = 3.76, p < .01) all demonstrated significantly enhanced writing performance.
IV. DISCUSSION
This comprehensive study investigated the impact of three different media modes in collaborative writing on the writing proficiency of Korean EFL learners, as well as their perceptions of modality and experiences of collaboration. It highlighted the role of content learning through semiotic resources in English writing. Overall, this study revealed that despite varying impacts and satisfaction levels across media modes, both linguistic and nonlinguistic modes contributed to improved writing performance in collaborative contexts. These results echo the finding that as media entails cognitive processing, requiring individuals to employ skills such as analysis, evaluation, categorization, induction, deduction, synthesis, and abstraction (Kim, 2014; Martens, 2010), writers can actively construct meaning through their own interpretation. In addition, narrative tasks are frequently employed to facilitate communication through media by helping learners notice, comprehend, and produce language. Accordingly, story creation across different media modes can be leveraged to support meaning negotiation in collaborative writing. Based on the analysis, this study presents four key findings that offer insights into how different media modes influence EFL learners’ writing performance through collaborative writing using Google Docs.
Firstly, the linguistic mode—a movie clip with sound—had a stronger impact on writing development, as measured by the four skills in the IELTS rubric, than the nonlinguistic modes—a movie without sound and a series of images. Participants indicated that vocabulary, grammar, and the story structure presented through the media supported their writing. However, they felt that their creativity was constrained due to the predetermined content. Specifically, in this mode, participants’ English proficiency may affect their ability to interpret and summarize the media content when generating a narrative. Thus, their performance in multimodal learning contexts can also be shaped by their prior linguistic competence. The movie clip without sound was also found to positively influence writing development, despite certain limitations in fulfilling the task response criteria. Nonetheless, as learners inferred characters’ actions and searched for their own expressions to construct a story without interference from preset dialogue (Kim, 2016), this mode appeared to enhance motivation, reduce anxiety toward learning, and promote meaning negotiation within a given context, ultimately contributing to writing development. In contrast, images yielded better outcomes in task response, likely due to the simplicity of the storyline, but resulted in the weakest overall writing performance. Despite the divergent findings, which may be attributed to inconsistencies in how media influence learning outcomes, the results clearly demonstrate that participants’ writing skills were significantly enhanced through the deployment of three different semiotic resources as scaffolds.
Secondly, non-linguistic modes displayed greater impact on co-construction of knowledge than linguistic mode based on participants’ satisfaction for collaborative writing. Among the three modes, a movie clip without sound yielded relatively higher outcomes. This result suggests that this mode may maximize learning outcomes in collaborative writing by providing learners with opportunities to engage interactively and develop their writing skills, along with scaffolding and motivational support offered by semiotic resources (Lim & Polio, 2020; Storch, 2005). It also supports the idea that non-linguistic modes, through shared interpretation and construction of meaning, can play a more significant role in the co-construction of knowledge than linguistic modes, particularly when language learners report higher satisfaction with collaborative writing. This aligns with the view that cognition and emotion are not isolated, static, or purely individual constructs, but rather integrated and highly contextualized processes shaped through social interaction (Gisbert & Font, 2008). This perspective is also reflected in social constructivism, which highlights dialogue-based interaction as a key process for internalizing shared knowledge through negotiated meaning (Topping, 2005).
Thirdly, peer feedback was not significantly related to writing outcomes. Although learners showed development in writing through meaning negotiation during collaborative learning, their satisfaction with peer feedback was relatively low. In this study, feedback primarily referred to the interaction that occurred after the writing was completed, which may explain the limited influence of peer feedback on writing performance. Writers’ experiences with collaboration were also an invaluable factor that may have influenced their writing performance. Collaborative learning may serve as an effective form of indirect enhancement, enabling self-reinforcement systems that stimulate intrinsic motivation. In addition, experience in collaborative writing was also important, as psychosocial relationships among peers can differ according to the structure of the learning task (Gisbert & Font, 2008). In individual interviews while some learners reported benefits from collaborative writing, such as shared interpretation and language exposure, others expressed difficulties related to uneven participation and limited engagement.
Lastly, Google Docs served as a useful digital tool for collaborative writing, providing cognitive scaffolding to support writers and facilitating the expansion of the learning environment. It enhances learning both within and beyond the classroom by enabling students to collaborate more effectively, exchange diverse viewpoints, and achieve learning goals (Zhou et al., 2012). Given this, the study demonstrated that Google Docs plays a significant role in improving learners’ writing skills.
V. CONCLUSION
This research probes how Korean EFL learners employed their writing skills and engaged with various semiotic resources during collaborative writing processes. The findings highlight divergences between individual and collaborative writing, particularly in the use of multimodal resources. From the perspective of cognitive load theory, learners demonstrate different learning responses depending on the media information mode they encounter (Mayer, 2001). This aligns with the notion that meaningful learning occurs when learners effectively organize and structure media input, underscoring the significance of the learning process. Thus, language educators and researchers need to design instructional activities that consider learners’ cognitive abilities and utilize media effectively as scaffolding.
Given the increasing importance of critical literacy involving multimodal texts and the ability to effectively create multimodal compositions, the results of this study offer valuable insights for advancing multimodal literacy among EFL learners through both linguistic and nonlinguistic modes in collaborative writing. Nevertheless, several relevant issues remain to be explored. These include how writers’ preferences and outcomes differ across varying media modes according to their English proficiency levels, and the extent to which these preferences shape their collaborative writing experiences. The content produced across media modes may have been differentially influenced by learners’ varying degrees of interest in each mode. Furthermore, how learners engage with writing activities using platforms such as Google Docs beyond classroom contexts has not been fully examined. Additionally, further investigation is needed into how individual writing improvements occur and the challenges associated with accurately assessing the impact of peer feedback. Addressing these aspects in future research will provide deeper insights into the dynamics of collaborative writing within media-based learning environments.
PDF Links
PubReader
ePub Link
Full text via DOI
Download Citation
Print
