AI-Based Vocabulary Games: Enhancing Memory Recall and Speaking Confidence among University Students

Abstract

Vocabulary knowledge remains one of the most decisive predictors of oral proficiency, yet many university students continue to struggle with retaining new words long enough to use them spontaneously in speech. Artificial intelligence (AI) based vocabulary games, which combine adaptive repetition algorithms with interactive game mechanics, have recently been proposed as a means of addressing this persistent gap. This study investigates how an AI-based vocabulary game influenced memory recall and speaking confidence among eight university students enrolled in a general English course at a private university in Indonesia. Employing a qualitative case study design, data were gathered over a five-week intervention through pre- and post-intervention interviews, reflective journals, and observation of students' in-app performance. Findings reveal that the adaptive, spaced-repetition features of the AI application supported more durable vocabulary retention than students' prior rote-memorization habits, while the game-like scoring and instant feedback mechanisms reduced the psychological pressure typically associated with speaking practice. Participants reported that repeated, low-stakes retrieval of target words within the game gradually carried over into spontaneous classroom speech, resulting in fewer hesitations and greater willingness to volunteer answers orally. The study also identifies remaining challenges, including uneven engagement outside scheduled sessions and the game's limited capacity to model natural conversational register. These results contribute to the growing body of literature on AI-mediated informal digital learning of English (AIIDLE) and offer practical implications for integrating adaptive gamified applications into vocabulary instruction and speaking-skill development in higher education.