Abstract
Conversational Agents (CAs) is rapidly transforming higher education by providing personalised learning support, adaptive feedback, and interactive educational experiences. However, most research has focused on learning performance rather than students psychological wellbeing. This scoping review examines how Self-Determination Theory (SDT) has been applied in the design of conversational AI to promote digital wellbeing and compassionate educational technology. Guided by the PRISMA Extension for Scoping Reviews (PRISMA-ScR), a systematic search of four major databases identified 28 eligible studies published between 2018 and 2026. Thematic analysis revealed that conversational AI predominantly supports students competence through intelligent tutoring, adaptive scaffolding, and personalised feedback, while explicit support for autonomy and relatedness remains limited. Digital wellbeing is rarely considered an intended design outcome. Based on these findings, the review proposes a preliminary SDT informed conversational AI design framework comprising four interconnected principles which are supporting learner autonomy, strengthening competence, fostering relatedness through compassionate interactions, and promoting responsible AI use to enhance digital wellbeing. Integrating motivational psychology with human-centred AI design, this review provides practical guidance for developing AI systems that support learner agency, psychological wellbeing, and sustainable engagement, contributing to the design of compassionate educational technologies for the future of higher education.
Keywords
Conversational AI, Self Determination Theory, Digital Wellbeing, Compassionate Technology, Educational Technology.
