Uysal, Derya2026-01-242026-01-2420261871-18711878-0423https://doi.org/10.1016/j.tsc.2025.102099https://hdl.handle.net/20.500.12868/5805Artificial intelligence in education increasingly leverages chatbots to provide on-demand support across diverse academic contexts. This study examined a curriculum-aligned chatbot in Structure of English, a first-year ELT course, focusing on student engagement and endline cognitive load through a Cognitive Load Theory lens. The design was an embedded single-case study with three units of analysis: behavioral engagement, affective engagement, and endline cognitive load. Participants were first-year ELT majors at a public university. Instruments included interaction logs, per-interaction satisfaction ratings (1-5), the 10-item Leppink Cognitive Load Scale (alpha = 0.849), and two open-ended questions on effectiveness, limitations, and improvement. Results showed that students submitted 6010 prompts ranging from 1 to 332 words (similar to 1-2236 characters), indicating both quick checks and multi-sentence inquiries. Weekly chatbot use tracked the assessment calendar: after a quiet start, it stabilized (approximate to 250-500 queries/week) before the midterm, declined post-midterm, and surged sharply before the final (approximate to 2500), indicating that exam timing strongly shaped behavioral engagement. Mean satisfaction was 4.28/5. Perceived usefulness emphasized independent learning (anytime/anywhere clarification), syllabus-aligned explanations, and better usability than the coursebook. Limitations included typo sensitivity, coverage gaps, and occasional irrelevant or repetitive replies. Endline cognitive load scores were IL = 6.01 (high), EL = 3.64 (low to moderate), and GL = 5.89 (medium). In conclusion, a course-aligned, outside-class chatbot was associated with frequent, varied engagement and positive perceptions, while yielding a CL profile typical of dense, abstract content-high IL, low to moderate EL, and medium GL.eninfo:eu-repo/semantics/closedAccessChatbotCognitive load theoryLinguistics courseLearning engagementPre-service EFL teachersArticle10.1016/j.tsc.2025.102099602-s2.0-105024611452Q1WOS:001642014000001Q1