Academic writing motivation of Chinese undergraduates in Gen-AI mediated English contexts: an integrated motivational framework

In light of recent calls to investigate Gen-AI’s impact on academic writing motivation, this study introduces an integrated motivational model. It examines the relationships among three key motivational constructs within the model: self-efficacy, future selves, and the perception of the context regarding its effectiveness in fulfilling basic psychological needs. Surveys were administered to 272 undergraduate English majors who attended Gen-AI mediated academic writing courses at the beginning (Time 1) and end of a semester (Time 2), followed by interviews with 9 students. The study yields three major findings. First, Time 1 and Time 2 structural equation modelling results indicate that the integrated model positions self-efficacy as the central motivational driver of academic writing performance, with basic psychological needs and future selves serving as indirect influences. Second, latent profile analysis reveals a motivational evolution among students, characterised by two motivational profiles at Time 1 and three at Time 2. Third, interview data show that Gen-AI usage tends to shape undergraduates’ self-efficacy, ideal L2 self, and basic psychological needs, but not their ought-to L2 self. These findings suggest pedagogically actionable strategies for fostering self-efficacy in Gen-AI mediated writing instruction.