Abstract
Short-form video platforms such as TikTok have become central to how learners encounter English, yet their role in vocabulary development remains underexplored in TESOL research. Existing studies often describe TikTok as engaging but rarely identify which design features most effectively support vocabulary growth. This study addresses that gap by investigating ESL learners’ perceptions of TikTok’s multimodal features and by introducing the TikTok-Inspired Multimodal Vocabulary Framework (T-MVF). Ninety-two pre-intermediate learners at a Malaysian university watched curated TikTok videos that represented seven feature categories (subtitles, visuals, narrative, video format and duration, gestures, audio, and interactivity) before completing a 42-item bilingual questionnaire. Descriptive and correlational analyses revealed that subtitles, visuals, and narrative formed the strongest cluster, while gestures, audio, and interactivity were effective only when integrated with this core. Strong correlations confirmed that feature integration, rather than isolated use, underpins perceived vocabulary support. The study contributes empirically by identifying learner-valued feature clusters, theoretically by proposing the T-MVF to explain how design features moderate input and mediate outcomes, and pedagogically by offering evidence-based strategies for integrating short-form video into vocabulary instruction. The findings underscore the global potential of TikTok as a low-cost, widely accessible, and engaging tool for vocabulary development. By clarifying which features matter most, the study advances multimodality research, informs ESL and EFL classroom practice, and aligns with Sustainable Development Goal 4 on Quality Education.