Summary: Much has been studied concerning the use of virtual reality (VR) as a display. However, relatively less has been performed with VR as a sensing tool, or as I lay out in this dissertation, an observatory. Using three studies, I demonstrate that the collection of head and hands pose data in current consumer VR, an instantiation of the paradigm of VR as an observatory, enables collection of body motion data in high temporal and spatial fidelity in large groups which is valuable for quantitative analysis regarding the study of nonverbal communication and poses risks to immersant privacy through biometrics. The three studies are (1) a study of person-to-person interactions through gaze and proxemics, (2) a study of synchrony measurement by exploring content validity, consistency, and predictive validity among 9300 measures of synchrony, (3) a demonstration of the identifiability of this data, and the effects of time upon identifiability.