|
Abstract
|
Vocabulary knowledge is a central component of second language (L2) proficiency, and within vocabulary, collocational competence plays a decisive role in achieving fluent, accurate, and idiomatic language use (Nation, 2022; Wray, 2002). Collocations—frequent and conventional word combinations such as make a decision or heavy rain—constitute a substantial proportion of natural discourse and academic texts (Howarth, 1998; Simpson-Vlach & Ellis, 2010). Despite their importance, collocations are consistently reported as one of the most problematic aspects of L2 learning, particularly for EFL learners whose exposure to authentic input is limited (Nesselhauf, 2005; Laufer & Waldman, 2011). Research has demonstrated that learners often possess partial or fragmented knowledge of collocations, showing discrepancies between receptive and productive mastery (Henriksen, 1999; Webb, 2008; Sonbul et al., 2022). Traditional instructional approaches typically rely on decontextualized vocabulary lists, uniform exercises, and teacher-fronted explanations, which fail to accommodate individual differences in proficiency, learning pace, and cognitive needs (Boers & Lindstromberg, 2009; Szudarski, 2018). Consequently, collocational knowledge develops slowly and remains resistant to instruction (Boers et al., 2014). Differentiated instruction (DI), grounded in sociocultural and constructivist perspectives (Vygotsky, 1978; Tomlinson, 2014), has been proposed as a pedagogical response to learner diversity. DI emphasizes adapting content, process, and learning outcomes according to learners’ readiness, interests, and profiles (Wang & Walberg, 1985; Waxman et al., 2013). However, implementing DI in EFL classrooms—particularly in large classes—poses practical challenges, including time constraints, monitoring learner progress, and providing individualized feedback. Recent advances in artificial intelligence (AI) offer promising solutions to these challenges. AI-powered tools and adaptive learn
|