This study introduces the development of a short-form English learning engagement scale. Based on Hoi & Le Hang's scale, it was validated through 362 pre-test and 3,238 formal samples, retaining 12 items that ultimately cover four dimensions: behavioral, cognitive, emotional, and social engagement. The scale demonstrated satisfactory reliability and validity, and its correlations with academic emotions and gender were confirmed, providing a reliable assessment tool for secondary school English teaching. An ethnographic case study examined the introduction of Lefebvre's rhythmanalysis theory into Chinese university English flipped classrooms, finding that the abrupt transition from traditional to flipped modes and cultural factors initially caused rhythmic dissonance. Through interventions integrating ideological-political education and production-oriented activities, teachers restored normal classroom rhythms, demonstrating the pedagogical implications of rhythmanalysis for second language teaching. A study on exemplification expressions in applied linguistics research papers, based on a self-built corpus of 80 papers using Frame Semantics, identified 13 markers activating five frames including Increment with 22 elements total. An exemplification framework containing four frame elements—Type, Instance, and others—was proposed accordingly, revealing its rhetorical and cognitive functions. A four-year narrative inquiry traced a university English teacher's trajectory in MALL contexts, presenting an identity evolution from "conservative beginner" to "natural user" to "enthusiastic follower" alongside an emotional progression from anxiety to acceptance to confidence. It highlights the need for emotional support in technology integration and offers implications for teacher professional development and technology融合. An analysis of voice negotiation in high-quality second language argumentative writing constructed a three-level analytical framework covering discourse, paragraph, and clause levels. Systematic analysis of 14 award-winning argumentative essays regarding generic structure, engagement resources, and lexicogrammar revealed the importance of hybrid generic stages and strategic engagement for authorial voice construction, with pedagogical recommendations proposed. A comparative evaluation of ChatGPT and Juzi Hanyu regarding word definitions and example sentences found that GPT exhibited problems including missing senses, incorrect part-of-speech labeling, redundancy, and circular definitions among near-synonyms. While its example sentences were practical, they lacked diversity and metalinguistic awareness. It is recommended that human-machine collaboration be adopted at the current stage to enhance its application value in dictionary compilation.