Empowering Interactions: Integrating Self-Determination Theory into Companion Technology Design

Abstract

Virtual assistants like Alexa, Siri, and AI chatbots are widely used for managing routine tasks but are often seen as tools rather than companions. These systems typically do not support users’ basic psychological needs, which limits engagement. This research,grounded in Self-Determination Theory (SDT), explores how digital companions can be designed to support autonomy, competence, and relatedness. The work investigates: (a)how companion systems can empower users by meeting psychological needs, (b) which features are perceived as companion traits, and (c) what design guidelines can help build systems that support user well-being. Using a mixed-methods approach, we evaluate proactive and reactive companion systems across case studies in training and skill building,information access, and digital well-being. Proactive systems like AR/VR companions and muscle stimulation feedback supported learning and collaboration, while reactive systems like a lockable phone box fostered user control and self-reflection. Autonomy is reinforced through clear opt-out mechanisms, adjustable boundaries, and non-intrusivenudges. Competence is supported through tailored feedback, repeatable task structures,and physical guidance. Relatedness is enhanced through dialogue, social accountability,and shared narratives. The dissertation provides future system designers and researchers with actionable guidelines for the design of companion technologies based on empirical research. Companion technologies should be adaptable, unobtrusive, and responsive to diverse user preferences for meaningful interaction. It contributes to the understanding of how technologies can act as great partners, addressing not only functional needs but also psychological well-being. It also calls for further research into long-term user interactions,cultural differences in technology use, and the integration of AI capabilities.