Abstract: The absence of transparency in robot processes presents a substantial obstacle to facilitating effective human- robot collaboration, particularly in response to the escalating need for autonomous human-robot interaction. This challenge assumes greater significance in non-industrial contexts, as it hinders humans from understanding a robot’s intentions, progress, and decision-making rationale, critical elements for fostering smooth interactions. To address this critical issue, this research introduces two distinct methodologies for integrating transparent behaviours into social robots: a by-design and a by-learning approach. Our preliminary results showed that in- corporating non-verbal cues significantly enhanced the robot’s transparency, leading to a more engaging interaction for human participants.