To understand the experience of immersive digital composing, this study investigated embodied writing practices in virtual reality (VR) environments through a mixed-methods analysis of 32 participants who completed VR writing tasks using virtual keyboards and three-dimensional handwriting tools. Drawing on established embodiment theories from writing studies, i.e., Perl’s felt sense theory, Weedon and Fountain’s embodied genres framework, Haas’s embodied practice conception, and Wysocki and Arola’s multimodal embodiment theory, the research revealed systematic patterns of embodied disruption and adaptation. Key findings demonstrated universal typing disruption across all participants contrasted with variable handwriting adaptation, sophisticated spatial environmental requirements, and an unexpected inverse relationship between VR expertise and adoption likelihood. While individual productivity writing faced significant embodied resistance, participants identified strong potential for collaborative and creative applications that use VR’s spatial affordances. The study advances embodiment theory by providing empirical validation of theoretical claims while establishing VR as a valuable methodological tool for investigating embodied writing processes that traditional methods cannot systematically capture.
As generative artificial intelligence becomes normalized in L2 writing, questions of authorial voice and textual ownership require analysis beyond broad claims of benefit or risk. This study reanalyzes the focal-case subcorpus of a semester-long EFL writing dataset drawn from English-major undergraduates at a public university in Jordan. The analysis is limited to 24 focal participants because only these cases contained the linked evidence needed to trace authorship across process and text: interviews, reflections, AI-use logs, prompt-output examples, Essay 2 draft histories, focal profile indicators, and paired first and final drafts. Using a qualitative-dominant mixed-methods multiple-case design, the study combines codebook thematic analysis of the process corpus with a structured paired-draft voice-comparison layer applied to the source-based Essay 2 drafts. Cross-case analysis identified three recurring pathways: revoiced assistance, support sliding toward dependence, and strategic nonuse to preserve control. The bounded quantitative strand suggested stronger final-draft voice where agency and verification were higher and direct text uptake lower. AI did not itself diminish voice; risk arose when fluent output displaced writer judgment and evaluative control.
This study examines how design thinking (DT) can function as a conceptual and analytical framework for supporting students’ critical AI literacy practices in writing and technical communication classrooms. Drawing on focus group interviews, individual reflection essays, and design artifacts from undergraduate technical writing students who collaboratively designed and customized AI chatbots using Poe, the study explores how students enacted responsible and user-centered design practices through each stage of the DT process. Findings reveal that 1) the empathize and define stages helped students identify algorithmic biases and frame inclusivity as part of user-centered design; 2) the ideate and prototype stages translated critical awareness into design decisions around prompt engineering, knowledge base curation, and visual accessibility; and 3) the test stage allowed students to revise their chatbot design through iterative usability testing and user feedback. Collectively, the DT process positioned students as human-in-the-loop co-creators who exercised rhetorical agency through prompt design, knowledge curation, and bias mitigation while designing customized chatbots. The article ends with pedagogical implications for integrating DT into critical AI pedagogy, chatbot design, and composition instruction.
