Social robots hold promise for reducing job interview anxiety, yet designing agents that provide both psychological safety and instructional guidance remains challenging. Through a three-phase iterative design study (N = 8), we empirically mapped this tension. Phase I revealed a "Safety-Guidance Gap": while a Person-Centered Therapy (PCT) robot established safety (d = 3.27), users felt insufficiently coached. Phase II identified a "Scaffolding Paradox": rigid feedback caused cognitive overload, while delayed feedback lacked specificity. In Phase III, we resolved these tensions by developing an Agency-Driven Interaction Layer. Synthesizing our empirical findings, we propose the Adaptive Scaffolding Ecosystem, a conceptual framework that redefines robotic coaching not as a static script, but as a dynamic balance between affective support and instructional challenge, mediated by user agency.
@article{arxiv.2601.10824,
title = {Bridging Psychological Safety and Skill Guidance: An Adaptive Robotic Interview Coach},
author = {Wanqi Zhang and Jiangen He and Marielle Santos},
journal= {arXiv preprint arXiv:2601.10824},
year = {2026}
}