This article outlines a reference architecture for social head gaze generation in social robots. The architecture discussed here is grounded in human communication, based on behavioral robotics theory, and captures the commonalities, essence, and experience of 32 previous social robotics implementations of social head gaze. No such architecture currently exists, but such an architecture is needed to: (1) serve as a template for creating or re-engineering systems, (2) provide analyses and understanding of different systems, and (3) provide a common lexicon and taxonomy that facilitates communication across various communities. A constructed reference architecture and the Software Architecture Analysis Method (SAAM) are used to evaluate, improve, and re-engineer two existing head gaze system architectures (Human–Robot Collaboration architecture and Robot Behavior Toolkit architecture). SAAM shows that no existing architecture incorporated the summation of functionalities found in the 32 studies. SAAM suggests several architectural improvements so that the two existing architectures can better support adaptation to a new environment and extension of capability. The resulting reference architecture guides the implementation of social head gaze in a rescue robot for the purpose of victim management in urban search and rescue (US&R). Using the proposed reference architecture will benefit social robotics because it will simplify the principled implementations of head gaze generation and allow for comparisons between such implementations.