Students’ attitudes, challenges, and needs in English reading comprehension: Foundations for AI-integrated corpus-based language pedagogy

Abstract

Reading comprehension is essential in English language learning, yet many EFL learners continue to struggle with vocabulary, implicit meanings, and complex academic texts. Addressing these challenges requires innovative pedagogical approaches that go beyond surface-level instruction. This study reports on a needs analysis conducted to support the development of a Corpus-Based Language Pedagogy model integrated with AI for reading instruction. Data were collected through a questionnaire administered to Indonesian EFL students, focusing on their reading attitudes, challenges, and instructional expectations. The results reveal that students generally hold positive attitudes toward reading in English, particularly when authentic texts and interactive activities are incorporated. However, they experience notable difficulties with unfamiliar vocabulary, complex grammatical structures, and lengthy passages. Current classroom practices are often limited to surface-level comprehension, which students perceive as inadequate for developing critical and analytical reading skills. In contrast, learners express the need for authentic corpus-based materials, relevant vocabulary, and contextualized reading strategies. They also highlight the importance of integrating technology and AI to enhance engagement and facilitate deeper comprehension. The study concludes that a corpus-based approach integrated with AI can provide authentic language input, expand lexical knowledge, and foster data-driven learning. The proposed model, therefore, offers a promising framework for improving students’ reading competence in the EFL context.