<p>Self-efficacy - the belief in one’s ability to complete a task effectively - is a key determinant of academic success and graduate outcomes in higher education. While active learning enhances engagement and self-efficacy in neurotypical students, little is known about how neurodiverse students experience its social and sensory demands. Grounded in Bandura’s social cognitive theory of self-efficacy, this pilot study explored how active learning impacts the self-efficacy of neurodiverse students in the BSc Medical Biosciences and MSc Genomic Medicine programmes at Imperial College London. An explanatory sequential mixed methods design was employed: a self-efficacy questionnaire provided quantitative breadth, with findings informing subsequent semi-structured interviews that explored the conditions under which self-efficacy was built or undermined. Three themes emerged: connection, control, and balance, corresponding to Bandura’s sources of self-efficacy (vicarious learning and social persuasion, mastery experience, and physiological and emotional states respectively). Participants valued peer and instructor connection and thrived when adequately prepared and in control of their learning. However, self-efficacy was fragile and dynamic, vulnerable to sensory overload and negative feedback. The findings demonstrate that active learning can support neurodiverse students’ self-efficacy, but only when environments are designed to promote psychological safety, predictability, and autonomy. Practical recommendations for neuroinclusive active learning design are offered.</p>